PDF

Dietary Caffeine and Sugar and their
Relationship with Sleep and Behaviour in

Children and Adults
By
Emily Jane Watson

Bachelor of Psychology (Honours)
Submitted in fulfilment of the requirements for the degree of
Doctor of Philosophy in Psychology
School of Psychology
Division of Education, Arts and Social Sciences
January 2017
I|Page
Table of Contents:
TABLE OF CONTENTS:............................................................................................................................. 2
LIST OF FIGURES..................................................................................................................................... 6
LIST OF TABLES ...................................................................................................................................... 7
TABLE OF ABBREVIATIONS AND SYMBOLS ............................................................................................. 9
CONTEXTUAL STATEMENT ................................................................................................................... 12
OVERVIEW OF AUTHOR’S CONTRIBUTIONS ......................................................................................... 14
WRITTEN PERMISSION OF CO-AUTHORS ............................................................................................. 16
SUMMARY OF PUBLICATIONS SUBMITTED DURING CANDIDATURE .................................................... 17
SUMMARY ........................................................................................................................................... 19
DECLARATION ...................................................................................................................................... 21
ACKNOWLEDGMENTS .......................................................................................................................... 22
CHAPTER 1 INTRODUCTION TO SLEEP .................................................................................................... 1
1.1 CHAPTER OVERVIEW........................................................................................................................................... 2
1.2 SLEEP .............................................................................................................................................................. 2
1.3 SLEEP NEED ...................................................................................................................................................... 4
1.4 WHY MIGHT SLEEP BECOME DISRUPTED? ............................................................................................................... 6
1.5 SUBJECTIVE SLEEP MEASURES ............................................................................................................................... 7
1.6 CONSEQUENCES OF NOT ENOUGH SLEEP, OR POOR SLEEP........................................................................................... 9
1.6.1 Mechanisms by which sleep could affect cognition and behaviour .................................................... 12
1.6.2 A possible mechanistic link between sleep and diet ........................................................................... 13
1.7 SUMMARY ..................................................................................................................................................... 14
CHAPTER 2 CAFFEINE & SLEEP.............................................................................................................. 15

2.1 CHAPTER OVERVIEW......................................................................................................................................... 16
2.2 CAFFEINE ....................................................................................................................................................... 16
2.2.1 How much caffeine do Australians consume? .................................................................................... 16

2.2.2 Caffeine Guidelines ............................................................................................................................. 20
2.3 DIETARY ASSESSMENT METHODS......................................................................................................................... 21
2.3.1 Caffeine food frequency questionnaires ............................................................................................. 25

2.4 EFFECTS OF CAFFEINE ....................................................................................................................................... 28
2.4.1 How caffeine impacts sleep ................................................................................................................ 28

2.4.2 Caffeine and sleep: Adults ................................................................................................................... 29
2.4.3 Caffeine and sleep: Children ................................................................................................................ 39
2.5 CAFFEINE AND BEHAVIOUR ................................................................................................................................ 51
2.6 CAFFEINE, SLEEP AND BEHAVIOUR IN CHILDREN...................................................................................................... 52
CHAPTER 3 SUGAR AND SLEEP IN CHILDREN ........................................................................................ 54

3.2 DIETARY SUGAR .............................................................................................................................................. 55
II | P a g e
3.2.1 Measurement of sugar consumption .................................................................................................. 57
3.3 SUGAR CONSUMPTION GUIDELINES ..................................................................................................................... 58
3.4 HOW MUCH SUGAR ARE CHILDREN CONSUMING? .................................................................................................. 59
3.5 SUGAR AND THE BODY ...................................................................................................................................... 63
3.6 SUGAR AND SLEEP............................................................................................................................................ 63
3.7 SUGAR AND BEHAVIOUR.................................................................................................................................... 71
CHAPTER 4 STUDY AIMS ...................................................................................................................... 73

4.1 STUDY ONE..................................................................................................................................................... 74
4.2 STUDY TWO .................................................................................................................................................... 75
CHAPTER 5 GENERAL METHODOLOGY ................................................................................................. 77

5.1 OVERVIEW ..................................................................................................................................................... 78
5.2 ETHICAL APPROVAL.......................................................................................................................................... 78
5.3 STUDY ONE .................................................................................................................................................... 78
5.3.1 Informed Consent ................................................................................................................................ 78

5.3.2 Participants ......................................................................................................................................... 79
5.3.3 Participant Recruitment ...................................................................................................................... 80
5.3.4 Measures............................................................................................................................................. 80
5.4 STUDY TWO ................................................................................................................................................... 84
5.4.1 Informed Consent ................................................................................................................................ 84

5.4.2 Participant Recruitment ...................................................................................................................... 84
5.4.3 Participants ......................................................................................................................................... 85
5.4.4 Measures............................................................................................................................................. 85
CHAPTER 6 VALIDATION AND REPRODUCIBILITY OF AN AUSTRALIAN CAFFEINE FOOD FREQUENCY
QUESTIONNAIRE .................................................................................................................................. 93
6.2 ABSTRACT: ..................................................................................................................................................... 94
6.3 PILOT STUDY ................................................................................................................................................... 95
6.4 INTRODUCTION ............................................................................................................................................... 96
6.5 MATERIALS AND METHODS ............................................................................................................................... 97
6.5.1 Participants ......................................................................................................................................... 98

6.5.2 Procedure ............................................................................................................................................ 98
6.5.3 Measures ........................................................................................................................................... 99
6.6 RESULTS ...................................................................................................................................................... 103
6.7 DISCUSSION.................................................................................................................................................. 107
6.8 CONCLUSION: ............................................................................................................................................... 112
CHAPTER 7 CAFFEINE CONSUMPTION AND SLEEP QUALITY IN AUSTRALIAN ADULTS ........................ 113
7.1 CHAPTER OVERVIEW....................................................................................................................................... 114
7.2 ABSTRACT: ................................................................................................................................................... 114
7.3 INTRODUCTION ............................................................................................................................................. 115
7.4 MATERIALS AND METHODS ............................................................................................................................. 117

III | P a g e
7.4.1 Participants ....................................................................................................................................... 117
7.4.2 Procedure .......................................................................................................................................... 117
7.4.3 Measures........................................................................................................................................... 117
7.4.4 Statistical Analysis............................................................................................................................. 118
7.5 RESULTS ...................................................................................................................................................... 119
7.6 DISCUSSION.................................................................................................................................................. 125
7.7 CONCLUSIONS............................................................................................................................................... 127
CHAPTER 8 THE RELATIONSHIP BETWEEN CAFFEINE, SLEEP AND BEHAVIOUR IN CHILDREN .............. 129
8.2 ABSTRACT .................................................................................................................................................... 130
8.3 BRIEF SUMMARY: .......................................................................................................................................... 131
8.4 INTRODUCTION ............................................................................................................................................. 132
8.5 METHOD ..................................................................................................................................................... 134
8.5.1 Participants ....................................................................................................................................... 134

8.5.2 Measures........................................................................................................................................... 134
8.5.3 Procedure .......................................................................................................................................... 137
8.5.4 Statistical analyses ............................................................................................................................ 138
8.6 RESULTS ...................................................................................................................................................... 140
8.7 DISCUSSION.................................................................................................................................................. 147
CHAPTER 9 TOTAL DIETARY SUGAR CONSUMPTION DOES NOT INFLUENCE SLEEP OR BEHAVIOUR IN
AUSTRALIAN CHILDREN ..................................................................................................................... 153
9.2 ABSTRACT .................................................................................................................................................... 154
9.3 INTRODUCTION ............................................................................................................................................. 156
9.4 METHOD ..................................................................................................................................................... 158
9.4.1 Participants ....................................................................................................................................... 158

9.4.2 Measures........................................................................................................................................... 159
9.4.3 Procedure: ......................................................................................................................................... 162
9.4.4 Statistical analysis: ............................................................................................................................ 162
9.5 RESULTS ...................................................................................................................................................... 163
9.6 DISCUSSION.................................................................................................................................................. 168
CHAPTER 10 GENERAL DISCUSSION ................................................................................................... 172

10.1 OVERVIEW ................................................................................................................................................. 173
10.2 CAFFEINE FOOD FREQUENCY QUESTIONNAIRE .................................................................................................... 175
10.3 IMPACT OF CAFFEINE .................................................................................................................................... 176
10.4 RELATIONSHIP BETWEEN CAFFEINE AND SUGAR.................................................................................................. 178
10.5 SUGAR INTAKE IN CHILDREN ........................................................................................................................... 179
10.6 IMPACT OF SUGAR ....................................................................................................................................... 179
10.7 FUTURE RESEARCH ....................................................................................................................................... 180
10.8 CONCLUDING REMARKS ................................................................................................................................ 183
IV | P a g e
REFERENCES....................................................................................................................................... 184
APPENDICES....................................................................................................................................... 212
V|Page
List of Figures
Figure 1.1: Two-process sleep model: homeostatic and circadian processes and their interaction...... 3
Figure 1.2: Bi-directional relationship of diet and sleep ....................................................................... 14
Figure 2.1 Proportion of caffeine from soft drinks, energy drinks, flavoured milk, and coffee, adapted
from the Australian Bureau of Statistics (2014b). ................................................................. 18
Figure 2.2: Proportion of caffeine from soft drinks, energy drinks, flavoured milk, and coffee,
adapted from the Australian Bureau of Statistics (2014b) .................................................... 19
Figure 3.1: A pictorial breakdown of foods into the basic sugar components. .................................... 56
Figure 3.2: Proportion of children meeting the sugar recommendations based on the 2003 dietary
guidelines. 60
Figure 3.3: Proportion of total sugar from different food groups, adapted from Australian Bureau of
Statistics (2014b). .................................................................................................................. 61
Figure 3.4: Bar graphs illustrating the overall decline in [A] sugar consumption (g) and [B] sugar as a
percent of energy (%E). ......................................................................................................... 62
Figure 6.1: Bland Altman plot of the difference between caffeine intake measured by the Food Dairy
and C-FFQ1 plotted against the mean caffeine intake for the two methods. .................... 105
Figure 6.2: Scatterplots showing individual responses of caffeine intake from the food diary and C-
FFQ. ...................................................................................................................................... 106
Figure 7.1: Description of participant flow throughout the study ...................................................... 120
Figure 7.2: Percent of caffeine intake from different sources by age groups. ................................... 122
Figure 7.3: Scatterplots showing the relationship between total caffeine consumed and sleep
variables: sleep efficiency, time in bed and sleep onset latency. ....................................... 123
Figure 8.1: Description of recruitment strategies and participant flow from each source ................ 140
Figure 8.2: Direct pathway analysis with Beta (β) values between the controlled variables (age and
socio-economic status) and between total caffeine consumption, morning tiredness and
internalising behaviours. ..................................................................................................... 145
Figure 8.3: Final Model. Indirect pathway analysis with Beta (β) values between the controlled
variables (age and socio-economic status) and between total caffeine consumption, sleep
routine, restless sleep, morning tiredness and internalising behaviours. .......................... 146
Figure 9.1: Description of recruitment strategies and participant flow from each source ................ 164
Figure 10.1: A scatterplot showing the relationship between caffeine consumption (mg) and sugar
consumption (g), r=0.228, p=0.001 ..................................................................................... 178
VI | P a g e
List of Tables
Table 1.1: Summary of recommended sleep duration, and sleep duration that may be appropriate
for each age group, adapted from Hirshkowitz et al. (2015) .................................................. 5
Table 2.1: Differing caffeine amounts for different beverages and foods in Australia. ....................... 17
Table 2.2: Mean daily caffeine intake (mg/day) for adults from the Australian Bureau of Statistics
(2015a)................................................................................................................................... 18
Table 2.3: Mean daily caffeine intake for children from the Australian Bureau of Statistics (2015a)
and the Australian Children’s Survey (2008) ......................................................................... 19
Table 2.4: Summary of the positives and negatives of the different diet measures ............................ 25
Table 2.5: Validity and reproducibility values for available caffeine-specific food frequency
questionnaires ....................................................................................................................... 27
Table 2.6: A summary of the literature regarding cross-sectional studies investigating the association
between sleep and caffeine in adults.................................................................................... 34
between sleep and caffeine in children ................................................................................ 45
Table 3.1: A summary of the literature regarding cross-sectional studies investigating carbohydrate,
sugar-sweetened beverages and sugar intake with sleep in children .................................. 66
between sugar intake and sleep in children.......................................................................... 70
Table 5.1: An outline of the exclusion criteria for study one................................................................ 79
Table 5.2: Different caffeine amounts for each category of the caffeine food frequency
questionnaire......................................................................................................................... 82
Table 5.3: An overview of the questionnaires used in the study, the length of the questionnaires, and
total time needed to complete them. ................................................................................... 86
Table 5.4: An overview of the scoring of puberty status ..................................................................... 91
Table 6.1: Different caffeine amounts for each category of the caffeine food frequency
questionnaire....................................................................................................................... 100
Table 6.2: An overview of the questionnaire, its categories and what it is measuring ...................... 101
Table 6.3: Means (standard error) and medians (interquartile range) of total caffeine consumption
calculated from the food diaries, C-FFQ1, and C-FFQ2. Correlations, mean differences and
statistical significance between the food diaries and C-FFQ1, and between the C-FFQ1 and
C-FFQ2. ................................................................................................................................ 104
Table 6.4: An overview of the quintile analysis showing the proportion of the C-FFQ1 values in the
same quintile, adjacent quintile or incorrectly classified compared to the food diary values
............................................................................................................................................. 107
Table 6.5: An overview of the quintile analysis comparing C-FFQ1 and C-FFQ2 for total caffeine
consumption, caffeine intake from coffee and tea, and caffeine intake from chocolate
consumption. ....................................................................................................................... 107
Table 7.1: Mean (SD), median (IQR) and range of responses of caffeine consumption and sleep
measures for participant ..................................................................................................... 121
Table 7.2: A descriptive table showing the means, standard deviations, medians, and interquartile
ranges for all variables for good and poor sleepers and any significant differences. ......... 124
Table 8.1: Means (standard deviation), medians (interquartile range), and range of values for
caffeine variables, sleep variables and behaviour variables. .............................................. 141
Table 8.2: Spearman correlation values for all relationships between variables ............................... 145
Table 8.3: A Pathway analysis summary, with standardized and unstandardized regression weights
for the model ....................................................................................................................... 146
Table 8.4: Summary of the Model fit indices ...................................................................................... 147
VII | P a g e
Table 9.1: Means (standard deviation), medians (interquartile range), and range of values for sugar
variables, sleep variables and behaviour variables. ............................................................ 165
Table 9.2: Spearman correlation values for all relationships between variables ............................... 166
Table 9.3: Correlations between sleep and behavioural variables and percentage of energy derived
from confectionary or sweet drinks .................................................................................... 168
Table 10.1: An overview of the findings from studies one and two ................................................... 174
VIII | P a g e
Table of Abbreviations and Symbols
↓ decrease or lower
↑ increase or higher
→ no association
% percentage
%E percent of energy
%SUG sugar intake as a percentage of energy intake
χ2 chi square
™ trademark
ACAES-FFQ Australian Child and Adolescent Eating Survey – Food Frequency

Questionnaire
Australian Children’s Survey 2007 Australian National Children’s Nutrition and Physical
Activity Survey
AUSNUT Australian nutrient composition database
BA bedtime anxiety
BMI body mass index
BMI-Z body mass index, z-scores
C Chi square test
CBCL Child Behaviour Checklist
CC caffeine consumption
C-FFQ (1&2) caffeine food frequency questionnaire (visit one and two)
CFI comparative fit index
CHO carbohydrate
CI confidence interval
CSHQ child sleep habits questionnaire
DECD Department for Education and Child Development
df degrees of freedom
Ext external behavioural problems
ESS Epworth sleepiness scale
FDA The Food and Drug Administration

FSANZ Food Standards Australia and New Zealand
FFQ food frequency questionnaire
g grams
IX | P a g e
h hours
I independent samples t-test
Int internal behavioural problems
IQR interquartile range
K Cohen’s K value
kg kilogram
kJ kilojoules
KSS Karolinska sleepiness scale
M mean
M Mann Whitney U-test
mg milligrams
mL millilitres
mins minutes
MT morning tiredness
mths months
n number of participants
N/A not available
NA night arousals
NHMRC National Health and Medical Research Council
NFI normed fit index
NREM non-rapid eye movemen6t
NSF national sleep foundation
NUTTAB nutrient tables for use in Australia
OR odds ratio
PCS puberty category scale
PSG polysomnography
PSPSI paediatric sleep problem survey instrument
PSQI Pittsburgh sleep quality index
REM rapid eye movement
RMSEA root mean square error of approximation
RS restless sleep
X|Page
SD standard deviation
SDB sleep disorder breathing
SDSC sleep disturbance scale for children
SE sleep efficiency
SE standard error
SES socio-economic status
SOL sleep onset latency
SR sleep routine
SSBs sugar-sweetened beverages
SWS slow wave sleep
TCC total caffeine consumption
TIB time in bed
TST total sleep time
UK United Kingdom
USA United States of America
VAS visual analogue scales
WASO wake after sleep onset
WHO World Health Organisation
y years
YAQ Rockett Youth/Adolescent Questionnaire
XI | P a g e
Contextual Statement
This thesis is submitted as a ‘Thesis Containing Published Research’ in accordance with the
University of South Australia’s Academic Regulations for Higher Degrees by Research Clause
16.1.1.b. This presentation method allows students to submit a combination of
conventional chapters and publications. It is required that publications be produced,
published or accepted for publication in peer-reviewed journals or conference proceedings
during the student’s candidature. The requirements of a ‘Thesis Containing Published
Research’ are as follows:
- Each publication must be peer-reviewed and recognised for the University’s

publication collection.
- Each publication must be published or accepted for publication pending revisions for
publication in significant journals, books and/or conference proceedings relevant to
the discipline.
- Conference papers published in conference proceedings and book chapters can only
be included where there is evidence of peer-review of the full work.
- Each co-authored publication included in the thesis must be accompanied by a
statement by the candidate describing the contribution of each author to each
publication. The statements must be signed by each author and incorporated in to
the thesis (see ‘Overview of Author’s Contributions’ and ‘Written Permission of Co-
Authors’).
- The candidate must be the greatest or equal greatest contributor to any co-authored
publications included in the thesis.
- The subject matter of the publications must be closely related and form a cohesive
narrative.
- The publications must be compliant with Copyright law and intellectual property
requirements.
- Where copyright has been assigned to the publisher, permission will need to be
sought to reproduce the work in the print and digital versions of the thesis. It is the
candidate’s responsibility to obtain permission to reproduce the work in the thesis.
Permission statements from publishers must be included as an appendix to the
thesis.
XII | P a g e
- Publications published, accepted, or submitted for publication prior to commencing
the degree, either at University of South Australia or at another university, must not
be included in the thesis.
Explanation of how published works included within the thesis are critically placed within
the area and the field of study, and the context of the thesis:
The three publications within this thesis form Chapters 6, 7 and 8. These publications have
been accepted in peer-reviewed journals and the permissions pertaining to the following
publications are included in Appendix A.
Chapter 6
Watson, E., Kohler, M., Banks, S., & Coates, A.M. Validation and Reproducibility of an
Australian caffeine food frequency questionnaire. International Journal of Food
Sciences and Nutrition. (In Press) doi: 10.1080/09637486.2016.1268102
The above paper was submitted for publication to the journal International Journal of Food
Sciences and Nutrition in October 2016. The submitted manuscript was peer-reviewed by
two independent peers and accepted for publication in November 2016.
Chapter 7
Watson, E. J., Coates, A. M., Kohler, M., & Banks, S. (2016). Caffeine Consumption and Sleep
Quality in Australian Adults. Nutrients, 8(8), 479.
The above paper was submitted for publication to the journal Nutrients in May 2016. The
submitted manuscript was peer-reviewed by three independent peers and accepted for
publication in August 2016.
Chapter 8
Watson, E. J., Banks, S., Coates, A. M., & Kohler, M. The relationship between caffeine, sleep
and behaviour in children. Journal of Clinical Sleep Medicine. (In Press).
The above paper was submitted for publication to the Journal of Clinical Sleep Medicine in
August 2016. The submitted manuscript was peer-reviewed by two independent peers and
accepted for publication in December 2016.
XIII | P a g e
Overview of author’s contributions
The following section describes the author’s contributions to the papers included within this
thesis.
Study One:
Chapter 6: Validation and Reproducibility of an Australian Caffeine Food Frequency

Questionnaire
Chapter 7: Caffeine Consumption and Sleep Quality in Australian Adults
I developed the project, carried out the study, collected data, undertook analysis (including
deciding which statistical tests would be most appropriate, conducting the statistical tests
and interpreting the data), and prepared the first draft of the manuscripts. As the
corresponding author, I submitted the publications to Nutrients and International Journal of
Food Sciences and Nutrition. Reviewers’ comments were received following submissions
and I was responsible for making the requested revisions and resubmitting the article for
publications. My supervisors (Mark Kohler, Alison Coates, and Siobhan Banks) assisted in
developing the project, analysis and revising the manuscript, as well as supervising the
study, deciding which journal is most appropriate, and revisions.
In addition, I helped supervise Psychology Honours students and third year Psychology and
Nutrition students who assisted on the study. The pilot data was collected over August and
September 2013 and the follow up study was collected March through to July 2015.
Study Two:
Chapter 8: The relationship between caffeine, sleep and behaviour in children
I developed the project, carried out the study, collected data, entered the data, undertook
analysis (including deciding which statistical tests would be most appropriate, running the
appropriate tests and interpreting the data), and prepared the first draft of the manuscripts.
As the corresponding author, I submitted the manuscript to Journal of Clinical Sleep
Medicine. Reviewers’ comments were received for chapter eight, following submission, and
I was responsible for making the requested revisions and resubmitting the article for
publication. My supervisors (Mark Kohler, Alison Coates, and Siobhan Banks) assisted in
XIV | P a g e
developing the project, analysis and revising the manuscript, as well as supervising the
study, deciding which journal is most appropriate, and with revisions.
In addition, I helped supervise Psychology Honours students and Psychology and Nutrition
students who worked on the study, ensuring that all data was double entered. The data
was collected over a period of 22 months from October 2013 to August 2015.
XV | P a g e
Written permission of co-authors
The co-authors listed below contributed to the publications included within this thesis. The
signatures provided below confirm each author’s contributions and acknowledge my input
into these two studies described in the overview of the author’s contributions.
Dr Mark Kohler
Dr Siobhan Banks
Associate Professor Alison Coates
XVI | P a g e
Summary of publications submitted during candidature
Peer-reviewed publications within the thesis
Chapter 6
Chapter 7
Watson, E. J., Coates, A. M., Kohler, M., & Banks, S. (2016). Caffeine Consumption and Sleep
Quality in Australian Adults. Nutrients, 8(8), 479.
Chapter 8
Watson, E. J., Banks, S., Coates, A. M., & Kohler, M. The relationship between caffeine, sleep
and behaviour in children. Journal of Clinical Sleep Medicine. (In Press).
First authored conference abstracts/presentations arising from thesis:

Watson, E. J., Banks, S., Coates, A. M., & Kohler, M. (2016). Impact of sugar on sleep and
behaviour in children. Oral presented at Australasian Sleep Association, Adelaide
Watson, E. J., Banks, S., Coates, A. M., & Kohler, M. (2016). Impact of caffeine on sleep and
behaviour in children. Poster presentation at European Sleep Research Society,
Bologna, Italy.
Watson, M., Kohler, M., Banks, S. & Coates, A. (2015) Sugar intake in a cohort of Australian
Children. Journal of Nutrition and Intermediary Metabolism. Poster session
presented at the Nutrition Society of Australia, Wellington New Zealand. DOI:
10.1016/j.jnim.2015.12.324
Watson, M., Banks, S., Kohler, M. & Coates, A. (2015) Relationship between caffeine
consumption and sleep in Australian children. Journal of Nutrition and Intermediary
Metabolism. Oral session presented at the Nutrition Society of Australia, Wellington
New Zealand. DOI: 10.1016/j.jnim.2015.12.190
Watson, E., Coates, A., Banks, S., Priestley, L. & Kohler, M. (2015) Development of a caffeine
intake questionnaire. Poster presentation session presented at the Australasian
Sleep Association, Melbourne, Australia.
Watson, E., Coates, A., Banks, S., & Kohler, M. (2014). How does a diet high in sugar add to
the impact of poor sleep on behaviour in children? Poster session presented at the
congress of the International Paediatric Sleep Association, Porto Alegre, Brazil
XVII | P a g e
Watson, E., Banks, S., Coates, A., & Kohler, M. (2013). Is glucose the mediator between sleep
and cognitive function in prepubescent children? Sleep and Biological Rhythms,
11(Suppl.2)
XVIII | P a g e
Summary
The importance of sleep for general wellbeing and behaviour is well-documented. Recently
the impact of diet on sleep and behaviour has been investigated, particularly for commonly
consumed dietary components thought to interfere with sleep processes including caffeine
and sugar. Caffeine is the most widely consumed stimulant, and despite experimental
studies beginning to characterise the impact of caffeine on sleep, minimal information
regarding real-world caffeine consumption in adults and children exists. This is most likely
due to difficulty in accurately measuring consumption. Additionally, most caffeinated
beverages are high in sugar, and excessive sugar consumption is frequently associated with
poor sleep and behaviour in children. This thesis aimed to investigate the impact of caffeine
on adult sleep, and the impact of caffeine and total dietary sugar on child sleep and
behaviour.
Initially, a novel caffeine food frequency questionnaire (C-FFQ) was developed, and the
association between caffeine consumption and sleep was investigated in 90 participants
(males: 36%; age: 39.6±20.8years). The C-FFQ showed fair validity, and strong
reproducibility, providing an effective measure of total caffeine intake and caffeine from a
variety of sources including: coffee and tea (hot and cold), soft drink, chocolate (food and
drink) and energy drinks. Furthermore, this study showed that while caffeine consumption
in adults had no association with sleep efficiency, and sleep onset latency, greater caffeine
consumption was associated with shorter time in bed, and was higher amongst those who
reported poor overall sleep.
Secondly, the effects of sugar and caffeine consumption on healthy children’s sleep and
behaviour were investigated. A total of 352 children and their parents/guardians (males:
49%; age: 10.7±1.3years) completed a questionnaire battery. Children completed a broad
food frequency questionnaire along with the C-FFQ. Parents completed surveys of their
child’s sleep quality/problems and daytime behaviour, as well as demographic information.
Results showed that greater caffeine consumption is associated with worse sleep and
daytime behaviour. However, the relationship between caffeine and daytime behaviour
was partially mediated by sleep factors. Finally, the study showed that total sugar was not
significantly associated with parent-reported sleep or child behaviour.
XIX | P a g e
The development and application of a novel measure of caffeine consumption has allowed
the accurate assessment of dietary caffeine intake in a group of Australian children, and
compared this intake to sleep and behaviour as well as sugar consumption. Overall, the
results demonstrate the negative impact that even small amounts of caffeine can have on
sleep in adults and children, and also daytime behaviour in children. Perhaps more
informative in the context of sleep is the significant interaction of caffeine, sleep and
behaviour. Given the large percentage of children reported to be consuming caffeine on a
regular basis, these results indicate a need for more salient package labelling and nutritional
guidelines. Also, the need to consider both sleep and caffeine intake when determining
appropriate management of problematic daytime behaviour in children is highlighted.
Finally, the finding that sugar has no significant association with sleep and behaviour in
children informs current misperceptions of the role of sugar intake in related problems.
XX | P a g e
Declaration
This thesis presents work carried out by myself and does not incorporate, without
acknowledgment, any material previously submitted for a degree or diploma in any
university; to the best of my knowledge it does not contain any materials previously
published or written by another person except where due reference is made in the text; and
all substantive contributions by others to the work presented, including jointly authored
publications, are clearly acknowledged.
Emily Watson
XXI | P a g e
Acknowledgments
Firstly, thankyou to all the participants and schools who gave up their time to take part and
assist in these studies, without each of them these studies would not have been possible.
Furthermore, to all the third year and honours students who assisted with data collection
and data entry during the study, thankyou. I would also like to thank the Commonwealth
government for their support as I was a recipient of the Australian Government Research
Training Program Scholarship.
Secondly, a huge thankyou to my principal supervisor Dr Mark (Maddog) Kohler; the time
and guidance you have given me has been invaluable. I cannot thankyou enough for the
opportunities, support, and encouragement you have provided and for being with me each
step of the way. Also Associate Professor Siobhan Banks; thankyou for your guidance and
support over many years, your encouragement and positivity has been much appreciated.
Last, but not least, Associate Professor Alison Coates; your passion and excitement for
research is contagious and your honesty and continuous support have helped get me here
today and words cannot thankyou enough for that. I honestly believe that I hit the jackpot
having you three as my supervisors.
Also, a thankyou to Professor Kurt Lushington, Professor Mary Carskadon, and Associate
Professor Jill Dorrian whose general advice, assistance in statistics, and project funding was
invaluable. Further thankyou to Louise Massie, Karen O’Brien, Pam Hart and Shiller
Peirawan, for all their help, such as but not limited to collecting surveys, answering endless
repetitive questions and continuous general support.
Thank you to everyone in the Centre for Sleep Research: Ciabatta, Gabbi, George, Crystal, Jill
Ryan (honorary), Rachel Faulkner, Chards, Cassie, Girl Alex, Boy Alex, Diana, Gemma, and
Will. I am so fortunate to have worked with such a great group of people and each of you
have added to making this experience enjoyable and memorable.
To all my friends outside of the university world, especially Ellie, Jacinta, Jules, Steph, and
Jane; thankyou for always making me laugh, listening to me, and lending your support
throughout the PhD, you have made this process easier and I am the luckiest person to have
each of you in my life. But mostly, thank you to my family: Mum, Dad, Nanna, Jamesy and
Harry. Thankyou for being so understanding, patient, helpful and accommodating, I cannot
thank you enough for your constant support and unwavering encouragement. Thank you!
XXII | P a g e
Chapter 1
Introduction to Sleep
1|Page
1.1 Chapter overview
This chapter provides a preliminary overview of sleep which is the foundation of the
current thesis; including a short discussion of why we sleep and sleep physiology,
estimated sleep need, common disruptions to sleep in the normal population, as well as
a summary of the major implications of disruption to sleep. This chapter also introduces
two important concepts in relation to sleep - diet and behaviour.
1.2 Sleep
From an evolutionary perspective sleep is a puzzling behaviour. While originally sleep

was perceived to be a simple and passive state, currently sleep is thought of as a
reversible behavioural state in which a person is neurologically and physiologically
active, whilst also being unresponsive to external stimuli and undergoing a decrease in
body movement (Carskadon & Dement, 2005; Gozal & Pope Jr, 2001). Two core
physiological stages of sleep exist in virtually all mammals; non-rapid eye movement
(NREM) sleep and rapid eye movement (REM) sleep. There is still a lack of consensus
regarding the precise function for sleep, however it is widely accepted that sleep is
essential, and benefits the human body in a number of ways.
Non-rapid eye movement sleep is subdivided into three stages, Stages 1-3, with sleep
onset beginning in Stage 1 of NREM (NREM-1) and then progressing through subsequent
stages of NREM. NREM-1 is relatively short and can be described as a transitional stage
between wake and sleep. NREM-2 follows NREM-1, and the body spends majority of its
sleep time in this stage. As slow EEG wave activity becomes prominent the body
transitions into NREM-3 (or slow wave sleep; SWS). In the first cycle of sleep, NREM-3
lasts longer, decreasing across cycles throughout the night (Carskadon & Dement, 2005;
Rechtschaffen & Kales, 1968; Silber et al., 2007). Following the first cycle of NREM sleep
an individual will typically briefly ascend back through to NREM-2, and subsequently the
first REM sleep episode will begin. Rapid eye movement sleep is characterised most
commonly by bursts of rapid eye movements. Rapid eye movement sleep is relatively
short during the first sleep cycle (approximately one to five minutes); however, it
becomes longer across the sleep episode, coinciding with shorter NREM-3. The stages of
sleep continue to rotate in this cyclical manner throughout the sleep episode. Over the
2|Page
course of the lifespan sleep architecture changes for example the percentage of REM
and NREM-3 sleep decreases with age and the percentage of NREM-1 and NREM-2
increases with age (Ohayon et al., 2004).
To explain the regulation of sleep and wakefulness Alexander Borbély described a model
of sleep regulation which involves two independent but interacting physiological
processes; Process C, the circadian process, and Process S, the sleep-dependent
homeostatic process (Borbély, 1982; Borbély, Daan, Wirz‐Justice, & Deboer, 2016; Figure
1.1). The two-process model is the most widely accepted model of sleep regulation, and
helps describe such aspects of sleep as timing, duration and intensity, as well as
fluctuations in a range of other functions including performance, alertness and mood.
AM PM AM PM
Figure 1.1: Two-process sleep model: homeostatic and circadian processes and their interaction.
Notes: interaction shown between the homeostatic process, Process S (red line), and circadian process,
Process C (blue line). During a sleep period (shaded section), homeostatic sleep susceptibility reduces.
The circadian drive for wakefulness peaks during the biological day (blue section) and is at its lowest at
night (Figure is adapted from Borbély and Achermann [1999]).
Process C, the Circadian Process, controls the body’s daily rhythms to a cycle of
approximately 24-hours. These daily rhythms are observed in a range of physiological
and behavioural functions as well as hormonal concentrations, and include sleep
propensity, core body temperature, heart rate, blood pressure, task performance,
melatonin production, cortisol levels, alertness, cognitive performance and mood
(Campbell & Murphy, 2007; Clark, Watson, & Leeka, 1989; Hastings, O’Neill, &
Maywood, 2007; Lack & Lushington, 1996; Wright, Hull, & Czeisler, 2002).
Process S, ‘sleep homeostasis’, refers to the accumulation of sleep need during periods
of wakefulness which then decreases after a sleep episode (Achermann, 2004; Borbély,
1982). It is believed that for infants and young children the build-up and dissipation of
3|Page
sleep pressure is greater than for adults, which may explain the numerous sleep/wake
bouts in infants and napping behaviour of young children (Jenni & Carskadon, 2007).
These two sleep processes interact to promote sleep when the homeostatic Process S
approaches the peak of Process C, and similarly waking is triggered when Process S
reaches the nadir of Process C. Typically, Process S and Process C oppose each other
throughout the wake period (approximately 0700 to 2300), which enables adults to have
fairly consistent vigilance levels throughout the day (Figure 1.1; Czeisler et al., 1999).
1.3 Sleep need
Sleep recommendations play an important role in informing families, public policies,

guidelines, and interventions of healthy sleep behaviours. Sleep recommendations have
previously been determined by expert opinion and consequently sleep
recommendations can differ. For example, among school-aged children
recommendations can differ by as much as 1.75 hours per night (Matricciani, Blunden,
Rigney, Williams, & Olds, 2013). Therefore, there has been a need for a more rigorous
guideline development process, with contributions from experts in the field as well as
reference to comprehensive systematic reviews of empirical investigations, to develop a
consensus (Matricciani et al., 2013).
In 2015 the National Sleep Foundation (NSF) convened a multidisciplinary expert panel
to formulate age-specific sleep duration recommendations (Hirshkowitz et al., 2015).
The NSF’s sleep duration recommendations take into consideration overall health and
well-being as well as cognitive, emotional and physical health (Hirshkowitz et al., 2015).
Similarly, in 2015, AASM and Sleep Research Society put together a panel of 15 experts
to determine amount of sleep needed for adults to promote optimal health (Watson et
al., 2015). Consequently during 2016 the American Academy of Sleep Medicine (AASM)
then used a panel of 13 experts in sleep medicine to determine the ideal amount of
sleep needed for optimal health for children (Paruthi et al., 2016). While sleep need can
differ from person to person, the above-mentioned reports provide broad
recommendations of the optimal sleep duration for each age group throughout the
lifespan (see Table 1.1).
4|Page
Table 1.1: Summary of recommended sleep duration, and sleep duration that may be appropriate for
each age group, adapted from Hirshkowitz et al. (2015)
National Sleep Foundation Recommendations (Hirschkowitz et al., 2015):
Age Recommended Amount May be Notes

Range of sleep per day appropriate
School-aged 6-13y 9-11h 7-8h/12h A post-pubertal adolescent typically

children sleeps less than a younger pre-
pubertal school-aged child.
Teenagers 14-17y 8-10h 7h/11h Early school-start times are a concern
for teenagers especially those who
exhibit evening chronotypes.
Young adults 18-25y 7-9h 6h/10-11h This age range has a large variance
with sleep; with those who enter the
workforce differing from college
students.
Adults 26-64y 7-9h 6h/10h Restricted sleep time particularly
affects 45-54-year-olds, the age range
when time at work usually reaches its
maximum in the life span.
Older adults 65+y 7-8h 5-6h/9h Older adults have more opportunities
to sleep compared with younger adults
and often nap.
American Academy of Sleep Medicine Recommendations (Watson et al., 2015; Paruthi et al., 2016):
School aged 6-12y 9-12h 12-13h Sleeping the recommended hours is

children associated with better health
outcomes. Sleeping fewer than the
number of recommended hours is
Teenagers 13-18y 8-10h Not associated with attention, behaviour,
mentioned and learning problems as well as
increased adverse health outcomes.
Adults 18-60y 7+h 9+ for some. Sleeping fewer than the number of
recommended hours is associated with
greater accidents and errors as well as
increased adverse health outcomes.
Sleeping more than 9 hours may be
appropriate for young adults, adults
recovering from sleep debt or illness.
Abbreviations: y, years; h, hours; +, greater than.
In adults, there is no evidence to suggest a consistent decrease in self-reported sleep

duration from the 1960’s to 2000s (Bin, Marshall, & Glozier, 2012, 2013). However,
there are concerns that children are not getting enough sleep, with sleep
5|Page
recommendations consistently exceeding reported sleep time by an average of 37
minutes (Matricciani et al., 2013). The same article further found that between 1905
and 2008 sleep duration has declined in older children by 0.91 minutes per year and
0.41 minutes per year for primary school children (Matricciani et al., 2013). This has
been further supported in an Australian specific population with a study conducted
during 2007 finding that total sleep time (TST) for school-aged children (10-15 years) has
decreased on average 28-33 minutes since 1974, due largely to later bed times
(Dollman, Ridley, Olds, & Lowe, 2007). This is concerning given the negative outcomes
associated with short sleep. There has been a suggestion that a decline in sleep duration
indicates that children can simply “cope” with less sleep now compared to 100 years ago
(Matricciani et al., 2013). However, more research is still required in this field before
such conclusions can be drawn.
1.4 Why might sleep become disrupted?
While recommendations for sleep duration are a helpful guide in educating on good
sleep practises, sleep quality also needs to be considered. Sleep disturbances reduce
sleep quality and can be a contributing factor to chronic sleep deprivation. The critical
role of sleep and sleep disturbances in daytime functioning is becoming increasingly
apparent. Sleep can be disrupted by sleep disorders, environmental disturbances,
illness, and more recently, evidence has emerged that diet can influence or even disturb
sleep.
At least 10% of the entire population suffers from a sleep disorder that is clinically
significant; insomnia, sleep disordered breathing and restless legs syndrome being the
most common (Partinen & Hublin, 2005). Environmental disturbances which impact
sleep include environmental noise (Griefahn, 2002; Muzet, 2007), technology such as
televisions and game consoles (Calamaro, Yang, Ratcliffe, & Chasens, 2012), and
increased room temperature (Haskell, Palca, Walker, Berger, & Heller, 1981; Muzet,
Libert, & Candas, 1984). Medical illnesses such as endocrine problems (Grunstein, 2005;
Grunstein & Sullivan, 1988) and cancer are also associated with disturbed sleep
(Koopman et al., 2002). Pain also disturbs sleep with studies showing people who
6|Page
experience chronic pain having a lower sleep efficiency (Okura et al., 2008) and changes
in sleep architecture (Lavigne, McMillan, & Zucconi, 2005).
Diet has recently been investigated as a potential means by which sleep can be
disturbed (Peuhkuri, Sihvola, & Korpela, 2012), however there are currently few studies
which have investigated this relationship, especially in children. Overall, clinical
interventions and cross-sectional observations suggest there is a connection between
the intake of macronutrients and sleep (Peuhkuri et al., 2012). Carbohydrates and fats
may modulate sleep quality by influencing the ratio of REM to non-REM sleep and the
amino acid Tryptophan has been shown to promote sleep (Hartmann, 1983; Peuhkuri et
al., 2012; Silber & Schmitt, 2010). Regardless of its cause, chronic sleep deprivation can
contribute to many negative outcomes in day to day functioning, of which are outlined
below in section 1.6.
1.5 Subjective sleep measures
Laboratory and experimental studies commonly measure sleep using such methods as
polysomnography (PSG) and actigraphy to obtain an objective assessment of sleep
activity. Unfortunately, these methods are typically not feasible in large cross-sectional
studies or field research as they are costly, burdensome and relatively labour intensive.
Both subjective and objective approaches to measuring sleep contain inherent strengths
and weaknesses. Objective measures of sleep tend to not capture the subjective sense
of having a poor sleep, while subjective measures fail to capture sleep stage differences
and have problems with the self-report bias (Gregory & Sadeh, 2012). Regardless,
survey study designs remain the only cost effective and logistically viable way to gather
sleep information from large populations in the community.
Adults
A number of questionnaires have been developed to assess adult sleep. In general,
these questionnaires ask participants about their sleep habits over a specified time
frame, commonly 2-4 weeks (Currie, 2006). One sleep variable measured by
questionnaires is daytime sleepiness, that is the subjective inclination to fall asleep. The
Stanford Sleepiness Scale (Hoddes, Zarcone, Smythe, Phillips, & Dement, 1973) and the
7|Page
Epworth Sleepiness Scale (Johns, 1991) are the most commonly used questionnaires to
measure daytime sleepiness. Other common adult questionnaires that examine sleep-
incompatible behaviours and sleep hygiene at bedtime are the Sleep Behaviour Self-
Rating Scale (Kazarian, Howe, & Csapo, 1979) and the Sleep Hygiene Awareness and
Practice Scale (Lacks & Rotert, 1986).
In adults, the Pittsburgh Sleep Quality Index (PSQI) is one of the most widely used sleep
questionnaires to determine sleep quality. The PSQI was founded by Buysse, Reynolds,
Monk, Berman, and Kupfer (1989) and has been used across the lifespan (Buysse et al.,
1989; Buysse, Reynolds, Monk, & Hoch, 1991), validated in populations with
psychological disorders (Backhaus, Junghanns, Broocks, Riemann, & Hohagen, 2002;
Insana, Hall, Buysse, & Germain, 2013), and has been translated into 48 different
languages (Buysse et al., 2008). The PSQI measures ‘sleep quality’ by a global sleep
quality score made up of a series of subscales: subjective sleep quality, sleep latency,
sleep duration, habitual sleep efficiency, sleep disturbance, use of sleeping medication
and daytime dysfunction.
Children
Total sleep time and sleep disturbances are a common way to measure sleep and sleep
quality in children as amount of sleep and sleep disturbances are related to many
different emotional and behavioural difficulties in children (Gregory & Sadeh, 2012).
During the past 20 years, the number of paediatric sleep measures has increased
significantly as previously, adult questionnaires were used and adapted for children
(Lewandowski, Toliver-Sokol, & Palermo, 2011). Questionnaires that have been well-
established and commonly used in children to measure sleep quality or sleep
disturbances include the Children’s Sleep Habits Questionnaire (CSHQ; Owens, Spirito, &
McGuinn, 2000), Paediatric Sleep Questionnaire (Chervin, Hedger, Dillon, & Pituch,
2000), Sleep Disturbance Scale for Children (SDSC; Bruni et al., 1997), and the Sleep
Habits Survey (Wolfson & Carskadon, 1998). Overall, the different measures screen for a
broad range of sleep problems such as sleep habits/hygiene, daytime sleepiness,
parasomnias and night awakenings (Lewandowski et al., 2011).
8|Page
Recently, in Australia the paediatric sleep problem survey instrument (PSPSI) has been
validated in school-aged children to measure, sleep problems and disturbances. The
PSPSI was designed using a combination of the CSHQ and the SDSC (Biggs, Kennedy,
Martin, van den Heuvel, & Lushington, 2012), two questionnaires which have undergone
rigorous validation and reliability analysis (Bruni et al., 1997; Owens et al., 2000). To
determine sleep quality based on sleep disturbance and sleep problems in children the
questionnaire measures six commonly problematic aspects of children’s sleep: sleep
routine (consistency of bedtime and sleep patterns), bedtime anxiety (anxiousness
about going to sleep), morning tiredness (difficulty in getting up and becoming alert in
the morning), night arousals (waking, either spontaneously or due to nightmares, during
the night), sleep disordered breathing (snoring during the night), and restless sleep
(movement when falling asleep or when asleep, and talking during sleep; Biggs et al.,
2012).
1.6 Consequences of not enough sleep, or poor sleep
Disruptions to sleep can occur as a result of a sleeping disorder (for example insomnia,
sleep disordered breathing and restless legs syndrome), environmental disturbance (for
example noise, technology or room temperature), or illness. Generally, it is agreed that
sleep acts to reverse the consequence of being awake, that is, sleep serves a restorative
function (Åkerstedt & Nilsson, 2003; Horne, 1988; Shapiro, Griesel, Bartel, & Jooste;
Siegel, 2005). Therefore, not getting enough sleep, or having a poor night’s sleep due to
any disruption is associated with many negative consequences in both adults and
children.
Physiological
There are non-visible factors that are impacted by sleep deprivation, namely the
physiological impact of reduced sleep. In adults sleep deprivation has been shown to
increase the risk of mortality (Kripke, Garfinkel, Wingard, Klauber, & Marler, 2002),
coronary events (Ayas et al., 2003), obesity (Cappuccio et al., 2008; Gangwisch,
Malaspina, Boden-Albala, & Heymsfield, 2005), cardiovascular events (Kripke et al.,
2002) and sympathetic activation (Knutson & Van Cauter, 2008; Leproult & Van Cauter,
2009). Further effects of sleep deprivation include an elevation in evening cortisol
9|Page
(Knutson & Van Cauter, 2008; Leproult & Van Cauter, 2009), and changes in timing of
the growth hormone (Spiegel et al., 2000). Together these changes may contribute to
alterations in glucose regulation, likely to result in reduced insulin sensitivity in the
morning consequently increasing the risk of developing diabetes (Knutson & Van Cauter,
2008). Sleep deprivation also lowers resistance to illness (Cohen, Doyle, Alper, Janicki-
Deverts, & Turner, 2009) and glucose tolerance (Knutson & Van Cauter, 2008; Leproult &
Van Cauter, 2009).
In children, poor sleep quality is associated with increased hypertension (Guilleminault,

Eldridge, Simmons, & Dement, 1976), headaches (Guilleminault et al., 1976), and obesity
(Eisenmann, Ekkekakis, & Holmes, 2006; Fiese, Everhart, & Wildenger, 2009;
Guilleminault et al., 1976; Gupta, Mueller, Chan, & Meininger, 2002; Kjeldsen et al.,
2014; Liu et al., 2008; Mamun et al., 2007; Noland, Price, Dake, & Telljohann, 2009;
Sekine et al., 2002). Furthermore, it is also associated with reduced immune function
(Ferreira et al., 2000), and physical activity in children (Gupta et al., 2002).
Cognitive
Changes to cognition are important given the impact more broadly on mental actions
and processes during the day. In adults reduced sleep has been linked to reduced
reaction time (Herscovitch & Broughton, 1981; Wyatt, Ritz-De Cecco, Czeisler, & Dijk,
1999), temporal memory (Harrison & Horne, 2000; Morris, Williams, & Lubin, 1960),
declarative memory (Ellenbogen, Payne, & Stickgold, 2006; Marshall & Born, 2007;
Smith, 2001; Walker & Stickgold, 2006), and attention and concentration (Balkin et al.,
2004; Belenky et al., 2003; Lim & Dinges, 2008). Furthermore, reduced sleep has been
associated with increased motor vehicle accidents and decreased driving ability (Philip et
al., 1996; Philip et al., 1999; Stutts, Wilkins, Osberg, & Vaughn, 2003).
In children, cognitive development is important for adequate learning. Children with

disturbed sleep demonstrate poor academic achievement with significantly worse
performance in math, science and spelling and significantly reduced attention capacity,
memory performance and intelligence (Blunden, Lushington, Kennedy, Martin, &
Dawson, 2000; Dewald, Meijer, Oort, Kerkhof, & Bögels, 2010; Gozal & Pope Jr, 2001;
Guilleminault et al., 1976; Meijer, 2008). Sleep disturbances are also related to
decreased learning ability (Gozal, 1998; Gozal & Pope Jr, 2001), inability to recall digits
10 | P a g e
(Sadeh, Gruber, & Raviv, 2002), and achievement motivation (Meijer, Habekothé, & Van
Den Wittenboer, 2000). Finally, children who felt better rested exhibited less boredom
at school (Meijer et al., 2000).
Mood and Mental Health

Understanding how sleep affects mood and mental health in both adults and children is
of upmost importance, with nearly 17.5% of the Australian population currently
diagnosed with a mental health disorder (Australian Bureau of Statistics, 2015b).
Furthermore, nearly all psychiatric and neurological mood disorders express
concomitant abnormalities of sleep (Walker, 2009). In adults reduced sleep worsens
subjective ratings of sleepiness (Belenky et al., 2003; Van Dongen, Maislin, Mullington, &
Dinges, 2003), self-ratings of alertness (Herscovitch & Broughton, 1981), and an increase
in overall mood disturbance (Blagrove & Akehurst, 2001; Brendel et al., 1990; Chuah,
Venkatraman, Dinges, & Chee, 2006; Mastin, Peszka, Poling, Phillips, & Duke, 2005).
Reduced sleep has also been shown to increase tension, anger, fatigue, anxiety,
confusion, depression, and decrease vigour (Angus, Heslegrave, & Myles, 1985; Brendel
et al., 1990; Dinges et al., 1997; Mikulincer, Babkoff, Caspy, & Sing, 1989; Paterson et al.,
2011; Penetar et al., 1993; Rodgers, Paterson, Cunningham, & Noble, 1995).
Furthermore, sleep deprivation has also been shown to reduce the ability to form new
memories (Walker, 2005) and amplify negative emotional consequences and blunting of
positive affective response (Zohar, 2005).
In children, an increase in sleep quality is associated with maintenance of mood

(Peterson & Benca, 2008), an increase in positive self-image (Meijer et al., 2000), and an
increase in ability to express emotions appropriately (Noland et al., 2009; Rosen et al.,
2004). Furthermore, it has also shown to decrease stress (Noland et al., 2009), and
symptoms of depression and anxiety (Alfano, Ginsburg, & Kingery, 2007; Alfano, Zakem,
Costa, Taylor, & Weems, 2009; Ivanenko, Barnes, Crabtree, & Gozal, 2004).
Diet and Hunger

More recently, sleep has been shown to influence diet and hunger. While most of the
research has been conducted in adults there is a small amount of research focussing on
children. In adults, decreased sleep is linked to an increase in appetite and food
consumption (Markwald et al., 2013; Morselli, Leproult, Balbo, & Spiegel, 2010; Spaeth,
11 | P a g e
Dinges, & Goel, 2013). Furthermore, food choices are largely impacted by sleep
deprivation, with healthy sleep improving diet in both children under the age of 18 years
(Al-Disi et al., 2010; Chen, Wang, & Jeng, 2006; Moreira et al., 2010; Weiss et al., 2010;
Westerlund, Ray, & Roos, 2009) and adults (Grandner, Kripke, Naidoo, & Langer, 2010;
Imaki, Hatanaka, Ogawa, Yoshida, & Tanada, 2002; Shi, McEvoy, Luu, & Attia, 2008). In
adults, reduced sleep duration is associated with changes in food choice such as
increased consumption of energy-rich foods, higher proportions of calories from fats or
refined carbohydrates, lower proportions of vegetables and fruits, and more irregular
meal patterns including consumption of more snacks, in particular sweet snacks (Heath
et al., 2012; Hogenkamp et al., 2013; Kim, DeRoo, & Sandler, 2011; Peuhkuri et al.,
2012). Furthermore, the incidence and prevalence of skipping breakfast are typically
higher in adults with low sleep duration than in those with normal sleep duration (Kim et
al., 2011). In summary, research to date suggests that an unhealthy diet is associated
with shorter sleep duration and irregular sleeping patterns.
Behavioural problems
A lack of sleep or poor sleep quality in children also affects their day-to-day behaviour.
Behavioural problems associated with sleep disturbances include aggressiveness (Ali,
Pitson, & Stradling, 1993; Meijer et al., 2000; Rosen et al., 2004; Wiggs & Stores, 1996),
oppositional behaviour (Rosen et al., 2004; Wiggs & Stores, 1996), hyperactivity and
inhibitory control (Ali et al., 1993; Guilleminault et al., 1976; Rosen et al., 2004; Sadeh et
al., 2002; Smedje, Broman, & Hetta, 2001; Wiggs & Stores, 1996), conduct problems
(Smedje et al., 2001), and abnormal behaviour including irritability and temper tantrums
(Ferreira et al., 2000; Wiggs & Stores, 1996). Furthermore, poor sleep has been shown
to increase social and peer problems (Noland et al., 2009; Rosen et al., 2004; Smedje et
al., 2001), an inability to express emotions (Rosen et al., 2004), emotional problems
more generally (Smedje et al., 2001), and also decrease adjustment to school (Bates,
Viken, Alexander, Beyers, & Stockton, 2002).
1.6.1 Mechanisms by which sleep could affect cognition and behaviour
There are a number of underlying mechanisms by which insufficient or disrupted sleep

may adversely impact cognitive functioning and behaviour. One mechanism, referred to
12 | P a g e
as the trace reactivation or replay hypothesis, proposes that sleep, specifically NREM-3,
aids memory consolidation through reactivation of traces of neuronal activity patterns
that encoded information during the preceding waking period (Born & Wilhelm, 2012;
Sejnowski & Destexhe, 2000; Sutherland & McNaughton, 2000; Wilson & McNaughton,
1994). The reactivation is proposed to aid transfer of information: from temporary
hippocampus-dependent storage to long-term hippocampus independent neocortical
storage. Consequently, if someone does not have adequate NREM-3 sleep or are
disturbed during NREM-3 sleep this will interfere with consolidation, enhancement, and
reorganization of explicit memory processes (Sutherland & McNaughton, 2000).
Another theory is the formal hypothesis, which suggests sleep plays a role in maintaining
functional integrity of the frontoparietal brain networks that support sustained
attention or the capacity to remain attentive and respond to stimuli for a prolonged
period of time. When a child has difficulty paying attention this can be interpreted as
poor behaviour and inability to maintain attention is the diagnostic criteria for several
behavioural disorders (American Psychiatric Association, 2013). Support for this idea
comes from studies that have shown sustained attention requires activation of an
extended, mostly frontoparietal, neuronal network that is compromised after sleep
deprivation (Choo, Lee, Venkatraman, Sheu, & Chee, 2005; Drummond et al., 1999;
Weissman, Roberts, Visscher, & Woldorff, 2006). During sleep deprivation, neural
changes are occurring rapidly and frequently in the brain making attention hard to
maintain (Chee et al., 2008; Goel, Rao, Durmer, & Dinges, 2009). Whether these
processes are consistent in child populations, given the immaturity of brain growth
comparatively to adults, is currently unclear (Fair et al., 2008; Lanius, 2011; Uddin,
Supekar, & Menon, 2010).
1.6.2 A possible mechanistic link between sleep and diet

The available literature suggests it is possible that the interaction between diet and
sleep is bi-directional (Figure 1.2). The research regarding the impact of diet on sleep is
still in its infancy and the mechanisms are yet to be completely understood; however,
there are some suggestions that the diet composition as well as appetite stimulant and
suppression hormones have important effects on sleep. Previous studies have found
when given ghrelin, an appetitive hormone, adults show a change in sleep architecture
13 | P a g e
including a decrease in REM sleep and a consequent increase in NREM-3 sleep (Garcia-
Garcia, Juarez-Aguilar, Santiago-Garcia, & Cardinali, 2014; Kluge et al., 2010; Weikel et
al., 2003). While ghrelin may play a role in sleep architecture in adults, the reason why
different macro/micro nutrients may impact sleep is unknown.
Figure 1.2: Bi-directional relationship of diet and sleep
The mechanisms behind the effect of poor sleep on diet have been more thoroughly
researched. One of the possible mechanisms hypothesized for an increase in appetite
and food consumption following sleep restriction is, again, regarding the disruption to
diet and appetite hormones. Studies have demonstrated that during sleep restriction
there is an increase in ghrelin and decrease in leptin, an appetite suppressor hormone
(Schmid, Hallschmid, Jauch-Chara, Born, & Schultes, 2008; Spiegel, Tasali, Penev, &
Cauter, 2004). The premise is that these differences in leptin and ghrelin during sleep
reduction increase appetite (Taheri, Lin, Austin, Young, & Mignot, 2004).
1.7 Summary
Overall, sleep is important for the healthy function of the human body. Whilst adults
appear to be having similar amounts of sleep today compared to previous generations
(Bin et al., 2012; 2013), there is evidence that children are having less (Dollman et al.,
2007; Matricciani et al., 2013). Previous studies provide evidence of the negative
impacts of not having enough sleep on day-to-day life, including diet and behaviour.
However, there is also evidence that diet may disrupt sleep, reducing total sleep. This
suggests that the relationship between sleep and diet is bi-directional. While behaviour
has already been well-established as a negative side effect of not enough sleep in
children, the additional impacts of different dietary components is unknown. The
following two chapters will discuss the literature regarding two specific aspects of diet,
caffeine and sugar, and how they influence sleep and behaviour.
14 | P a g e
Chapter 2
Caffeine & Sleep
15 | P a g e
This chapter provides a summary of caffeine consumption in Australia, an overview of

the dietary guidelines for caffeine consumption, and how caffeine can be measured in
field research. Furthermore, the chapter will discuss the impact of caffeine on different
aspects of human function, in particular sleep in adults and children and behaviour in
children, with a discussion on possible mechanisms for these interactions. Finally, the
literature relating caffeine and sleep in both adults and children and the literature
relating caffeine and behaviour in children is reviewed.
2.2 Caffeine
Caffeine (1,3,7-trimethylxanthine) is a natural stimulant found commonly in foods and

beverages, including coffee, tea, soft drinks, energy drinks and chocolate. Caffeine
metabolism occurs primarily in the liver (Arnaud, 1999; Stavric & Gilbert, 1990) and is
almost entirely metabolised, with only 1-5% of ingested caffeine recovered in urine
(Stavric & Gilbert, 1990). Caffeine is almost entirely absorbed in the stomach and small
intestine and distributed to all bodily tissues, including the brain. It is rapidly absorbed
in about 30-120 minutes with its highest concentration occurring in the blood and brain
within about 30-40 minutes (Smith, 2002). The impact of caffeine usually lasts five to
seven hours but can be influenced by many factors, including gender, age, medications,
pregnancy, and smoking (Kaplan et al., 1997; Rogers, 2007; Smith, 2002). Genetic
variations have also been considered to potentially impact caffeine metabolism (Grosso
& Bracken, 2005). Frequent caffeine users also differ in their response and tolerance to
caffeine (Attwood, Higgs, & Terry, 2007; Evans & Griffiths, 1992; James & Rogers, 2005)
and genetics can also influence the amount of caffeine consumed as well as the
response to caffeine (Yang, Palmer, & de Wit, 2010).
2.2.1 How much caffeine do Australians consume?

As previously stated caffeine is in a wide range of commonly consumed products
including coffee, tea, soft drinks, energy drinks and chocolate. Furthermore, the amount
in each source can change depending on the brand, region, differences in reference
volumes (different sized cups), product sources (plant variety) and methods of
16 | P a g e
preparation (methods of brewing; Barone & Roberts, 1996; Stavric et al., 1988; see
Table 2.1). Recently two surveys have been completed in a large sample of Australians
to identify the amount of caffeine being consumed in adults and children nationally.
Table 2.1: Differing caffeine amounts for different beverages and foods in Australia.
Notes: the information in this table was determined by the information from manufacturers’ websites,
NUTTAB (Food Standards Australia New Zealand, 2015), or personal communication with the
manufacturer.
Caffeine (mg/100ml)
Energy drinks 30- 32
Soft drink 10-32
Coffee
Iced 31-194
Coffee flavoured milk 15
Hot Coffee varieties 31-194
Tea
Iced 7
Hot tea varieties 0-19
Chocolate (mg/100g)
Chocolate Milk 10-37
Milk chocolate 16-20
White chocolate 5
Dark chocolate 30-59
Abbreviations: mg, milligrams; g, grams; mL, millilitres
Adults
The Australian Bureau of Statistics (ABS) collects data on a wide range of economic,
social, population and environmental matters and for the first time in 2011-2012 the
ABS collected data on caffeine consumption. All nutrition data was collected between
May 2011 and June 2012 with a total sample size of 12,153 people. Adults and children
(aged 2 years and older) participated in face-to-face 24-hour recall and a second 24-hour
recall at least eight days after, administered by trained interviewers. The method used
was the automated multiple pass method (Raper, Perloff, Ingwersen, Steinfeldt, &
Anand, 2004; Steinfeldt, Anand, & Murayi, 2013), adapted to reflect Australian food
supply. Children aged two to four years were interviewed by proxy and children aged 5-
11 years were asked to provide recall with assistance from an adult household member.
17 | P a g e
Males and females had similar average intakes of caffeine across all age ranges (Table
2.2), with an average caffeine intake of 159.6mg per day (equivalent to a 500mL can of
Red Bull). Additional information regarding where the caffeine comes from was also
collected. The breakdown of the percentage of caffeine across different beverages is
shown in Figure 2.1. The different sources of caffeine show a decrease in soft drink
consumption and a rise in tea across adult age groups, with a relatively stable coffee
consumption.
Table 2.2: Mean daily caffeine intake (mg/day) for adults from the Australian Bureau of Statistics
(2015a)
Caffeine mg/day
Age groups Males Females All
19-30y 104.1 101.5 102.8
31-50y 187.4 171.1 179.2
51-70y 188.4 177.8 183.0
71+y 147.6 154.2 151.2
19+y 163.8 155.5 159.6
Abbreviations: mg, milligrams; y, years; +, greater than
Figure 2.1: Proportion of caffeine from soft drinks, energy drinks, flavoured milk, and coffee, adapted
from the Australian Bureau of Statistics (2014b).
Notes: caffeine from foods was not included.
18 | P a g e
Children
Data from the ABS showed that, on average, children aged between 2-18 years were
having 20.3mg of caffeine per day (approximately equivalent to a 200mL can of cola;
Table 2.3). The rise in caffeine intake across childhood is potentially explained by
increased freedom and availability in food and beverage choice. On average, males and
females over 18 years consumed more caffeine than younger children. Coffee
consumption seemed to dramatically increase in the 14-18-year age group while after
age 3-years tea consumption remained consistent. Caffeine from soft drink
consumption peaked in the 9-13-year age range (Figure 2.2).
Table 2.3: Mean daily caffeine intake for children from the Australian Bureau of Statistics (2015a) and
the Australian Children’s Survey (2008)
Australian Bureau of Statistics Data Australian Children’s Survey

Caffeine mg/day Caffeine mg/day
Age groups Boys Girls All Age groups Boys Girls All
2-3y 2.0 2.1 2.0 2-3y 3.4 3.3 3.4
4-8y 5.6 6.2 5.9 4-8y 8.5 7.7 8.1
9-13y 19.2 13.8 16.5 9-13y 19.7 18.7 19.2
14-18y 52.1 42.8 47.5 14-16y 46.8 36.4 41.7
2-18y 22.1 18.4 20.3
Abbreviations: mg, milligrams; y, years
Figure 2.2: Proportion of caffeine from soft drinks, energy drinks, flavoured milk, and coffee, adapted
from the Australian Bureau of Statistics (2014b)
Notes: caffeine from foods was not included.
19 | P a g e
The 2007 Australian National Children’s Nutrition and Physical Activity Survey (Australian
Children’s Survey) assessed food and nutrient intake in a sample of children aged 2-16
years randomly selected from across Australia. Data were collected on two occasions
from 4,487 participants or their caregivers between February and August 2007. A
computer assisted 24-hour recall interview was conducted in the child’s home, followed
7-21 days later by a computer assisted telephone 24-hour recall interview. In
collaboration with Food Standards Australia New Zealand (FSANZ), the food and
beverage intake data were translated to daily nutrient intake data using the most recent
Australian nutrient composition database (AUSNUT 2007). The data from the survey
showed similar caffeine intake values across the age groups compared to the ABS data
(Table 2.3) indicating caffeine intake has remained stable since 2007 (Commonwealth
Scientific Industrial Research Organisation [Australia] et al., 2008). Unfortunately, the
Australian children’s survey did not break down caffeine consumption to allow for
identification of which areas of the diet the caffeine was coming from.
2.2.2 Caffeine Guidelines

Currently, in Australia, there are no accepted daily intake guidelines for caffeine
consumption for either adults or children. The effects of caffeine on human health were
explored by Nawrot et al. (2003) in a systematic review. This review concluded that in a
general population of healthy adults, moderate caffeine intake at a dose level of 400mg
per day is not associated with adverse effects such as general toxicity, cardiovascular
effects, effects on bone status and calcium balance, changes in adult behaviour,
increased incidence of cancer, or effects on male fertility. Furthermore, Nawrot et al.
(2003) recommended an upper intake of 2.5mg per kilogram of body weight per day in
children; for example, for an average six-year-old weighing 20kg this is equal to 50mg of
caffeine per day (equivalent to a 250g block of solid milk chocolate). Based on this
recommendation, other countries have determined caffeine guidelines for adults and
children.
Europe, the United Kingdom (UK), United States of America (USA), and Canada have
guidelines for caffeine intake. In Europe and the UK, the caffeine guidelines state that
for healthy adults daily habitual caffeine consumption should be no more than 5.7mg
20 | P a g e
per kilogram of body weight (European Food Safety Authority, 2015); equivalent to
approximately 400mg per day. Europe and the UK also provide guidelines for children’s
habitual caffeine intake, stating that 3mg per kilogram of body weight per day is not
unsafe (European Food Safety Authority, 2015). Health Canada released guidelines for
habitual caffeine intake for adults of no more than 400mg and for children a maximum
daily caffeine intake of 2.5mg per kilogram of body weight (Health Canada, 2012). The
Food and Drug Administration (FDA) (2013) in the USA concluded that for the healthy
adult population, moderate daily caffeine intake at a dose level up to 400mg day is not
associated with adverse effects. There are no guidelines provided for children in the
USA (Food and Drug Administration, 2013). It appears that on average Australians are
consuming within the recommended international guidelines, however the proportion
consuming above these recommended doses is not known.
2.3 Dietary assessment methods
Measuring dietary intake is often tedious and costly, particularly in large populations.
There are two broad categories of measuring dietary intake: assessments that record
food consumed at the time of eating (food diaries), and those that collect data
retrospectively over a time period (24-hour recalls and food frequency methods [FFQ]).
Food diaries usually cover several days of recording while retrospective methods
typically cover recent diet history (24-hour food recalls) or habitual diet (FFQ; van
Staveren et al., 2012). These tools differ in many ways such as the time frame captured
(average intake over a period of time or current intake), administration method
(interview, questionnaire, observation), measurement of foods consumed (weighing or
estimation) and conversion into macro- and micronutrients (nutritional database; van
Staveren, Ocké, & de vries, 2012). The following sections describe the various methods
and discuss their respective strengths and weaknesses.
24-hour recall
The 24-hour recall method is one of the most common methods of recording dietary
intake. The 24-hour food recall consists of an interview where the participant is asked to
remember what they consumed during a preceding period of time (typically 24 hours;
Thompson & Byers, 1994). This method may be conducted via a personal interview
21 | P a g e
(Thomson & Subar, 2001), telephone interview (Casey, Goolsby, Lensing, Perloff, &
Bogle, 1999 Perloff, & Bogle, 1999) or computer assisted interview (Conway, Ingwersen,
Vinyard, & Moshfegh, 2003 & Moshfegh, 2003; Slimani et al., 1999). The interviewer
asks for detailed information about everything the participant had to eat and drink
including details concerning food preparation methods. Food consumption is typically
measured using household measurements but food models or photographs may also be
used (Erdman Jr, MacDonald, & Zeisel, 2012). To obtain accurate and complete dietary
recalls, the interviewer requires training to ask specific probing questions (e.g. ‘What
type of milk did you use to make your tea/coffee?’; Barrett-Connor, 1991; Erdman Jr et
al., 2012). A common method of 24-hour recall is the automated multi pass method
which uses five steps and memory cues with standardised wording to encourage recall
of all possible foods and beverages consumed during the previous 24 hours (Moshfegh
et al., 2008).
The 24-hour recall method is generally used to describe the mean intakes of a group of
individuals (Beaton et al., 1979). The advantage of the 24-hour food recall method is
that information is collected while an interviewer can ask further questions to ensure
accurate detail. Further advantages include ease of administration, minimal burden on
the participant’s time, and does not require literacy (Barrett-Connor, 1991; Erdman Jr et
al., 2012; Thompson & Byers, 1994; Willett, 2012). However, 24-hour recalls also have
limitations including the cost of administration by trained professionals which often
makes this method unsuitable for large scale studies measuring one nutrient. Also,
recalls are often not estimated correctly, as intakes (particularly portion sizes) can be
difficult to remember, the method is not representative of the participant’s overall diet,
and if the record is only taken for one day within-individual variation cannot be
determined (Erdman Jr et al., 2012; Thomson & Subar, 2001). Furthermore, there is a
tendency for participants to show optimism bias whereby they under-report foods and
food portion sizes for foods which are perceived as less healthy (Barrett-Connor, 1991;
Bathalon et al., 2000; Krall, Dwyer, & Coleman, 1988; Kroke et al., 1999; Miles & Scaife,
2003; Thompson & Byers, 1994; Wein, Sabry, & Evers, 1990; Willett, 2012). Under-
reporting appears to be related to factors such as weight, gender, education and social
desirability (Hebert, Clemow, Pbert, Ockene, & Ockene, 1995; Krebs-Smith et al., 2000).
22 | P a g e
Food diary/weighted record
A food diary requires all food and/or beverages to be recorded in detail by the
participant typically for a period of three to seven days (Thompson & Byers, 1994). The
food diary records current intake and thereby estimates absolute rather than relative
intake of energy and other food components (Willett, 2012). Food intake is ideally
recorded in detail by the participant at the time the food is eaten to minimise the
reliance on memory, allowing a high precision of estimating participants’ nutrient intake
(Thompson & Byers, 1994; Willett, 2012). Consequently, it is less likely that a participant
will forget portion sizes and therefore food diaries are often considered as the ‘gold
standard’ (Barrett-Connor, 1991; Thompson & Byers, 1994). However, it is important
that participants are taught methods of keeping complete and accurate food diaries and
provided with forms and instructions to guide recording of the appropriate levels of
detail. Typically, the participant describes information such as the brand name of the
food, preparation method (for example boiling or frying) and recipe details (Thomson &
Subar, 2001). If it is not practical to weigh the food, the participant may describe food
consumed (for example restaurant meals) and the researcher will estimate weights.
Portion sizes are usually described using measurement utensils commonly found in
homes, such as cups and tablespoons (Erdman Jr et al., 2012).
Food diaries have limitations including being inefficient for both researchers and
participants when measuring one nutrient, due to the time and cost involved in detailed
analysis using dietary assessment software (Thomson & Subar, 2001). Also, participants
must be literate, physically able to write, and motivated to keep a detailed diary as
reporting involves a relatively high burden to participants (Barrett-Connor, 1991;
Brunner, Juneja, & Marmot, 2001; Thomson & Subar, 2001; Todd, Hudes, & Calloway,
1983). Furthermore, participants may change their typical eating pattern over the
recording period (Rebro, Patterson, Kristal, & Cheney, 1998) and under-reporting is
common (Black et al., 1993; Taren et al., 1999). In particular, some populations such as
those with a high body mass index (BMI) and women are more likely to under-report
food intake (Lichtman et al., 1992; Johnson, Goran, & Poehlman, 1994).
23 | P a g e
Food Frequency Questionnaires
Since 24-hour-recall and food diaries are generally expensive, unrepresentative of usual
intake, and can be inappropriate for assessment of past diet, investigators have sought
alternative methods for measuring long-term dietary intake. An ideal alternative is the
use of food frequency questionnaires (FFQ) which unlike the 24-hour recall and food
diaries are suited to providing an overview of the person’s average diet (Kristal, Peters,
& Potter, 2005). A FFQ requires respondents to consider a specific period of time
(typically 6 or 12 months) and estimate their habitual consumption, frequency of
consumption, and amount of food from a list of foods (Willett, 1994; Zulkifli & Yu, 1992).
Food frequency questionnaires are typically administered in larger epidemiological
studies and are used to estimate the food intake of individuals or groups.
This method reduces the burden on the participant and the research team compared to
collecting a diet history via interview (Erdman Jr et al., 2012). Questionnaires are usually
self-administered and take around 30–60 minutes to complete, with response rates
typically high. In addition, the cost of using this technique is low (usually data can be
scanned or completed electronically), and interviewer variation is reduced when food
lists are standardised. However, similar to previously described methods, the limitations
of a FFQ include reduced accuracy as it is difficult to remember food intake over longer
periods of time and cannot be used to investigate day-to-day variation. Further
weaknesses of the FFQ are that the detail recorded is food specific and may not capture
all sources of important nutrients if certain foods are not included on the questionnaire,
and quantification is often difficult as FFQs may be inaccurate due to portion sizes
differing across eating occasions (Erdman Jr et al., 2012; Thomson & Subar, 2001).
In summary, all methods of collecting dietary information have strengths and

weaknesses (Table 2.4). Although dietary records are considered the gold standard for
collecting dietary intake data, they are not suitable for larger scale studies. It is
important therefore to consider factors such as cost, participant burden, and specifics of
the group to be studied (for example literacy or a group known to under-report). When
assessing diet in a large population the use of FFQs is an ideal alternative.
24 | P a g e
Table 2.4: Summary of the positives and negatives of the different diet measures
Positives Negatives
24-hour recall  Not a burden on participants  Relies on memory
 Does not require literacy  Costly
 Data entry is timely
 Interviewer requires training
 Not suitable for large studies
 Under-reporting is common
Food diary  Does not require memory  A burden on participants

 Considered ‘Gold Standard’  Requires literacy and motivation from the
participants
 Data entry is timely and costly
 Participants require instruction
 Under-reporting is common
Food  Not a burden on participants  Requires literacy from the participants

Frequency  Not costly  Relies on memory
Questionnaire  Minimal data entry  May not capture all sources of important
nutrients
2.3.1 Caffeine food frequency questionnaires

Measuring one specific nutrient such as caffeine is difficult in large populations and
therefore FFQs are typically used due to cost and ease (Subar et al., 2003; van den
Brandt et al., 1990). Europe, UK and USA have validated caffeine-specific FFQs (Boylan
et al., 2008; Bühler, Lachenmeier, Schlegel, & Winkler, 2013; Ferraroni et al., 2004;
Rudolph, Färbinger, & König, 2012; Schliep et al., 2013). Of the caffeine FFQs that have
reported validity, correlation values were quite varied and ranged from 0.29-0.82
(Boylan et al., 2008; Ferraroni et al., 2004; Rudolph et al., 2012; Schliep et al., 2013). Of
the studies that reported reproducibility, correlation values ranged from 0.61-0.86
(Ferraroni et al., 2004; Schliep et al., 2013). However, most of the questionnaires do not
measure all caffeinated food and drinks, and those that did often did not specify
between sizes and caffeine contents, or reproducibility was not reported (Table 2.5). Of
the four studies that reported a methodology (Boylan et al., 2008; Ferraroni et al., 2004;
Rudolph, Färbinger, & König, 2012; Schliep et al., 2013), three utilised similar
methodologies whereby FFQ forms were completed before the 24-hour recall or food
diary. This means that the FFQ was consistently not measuring the same period of time
as the other dietary measure, and therefore not measuring the same intake. Due to the
differences in day-to-day caffeine intake, not only in preference but also make-up of the
25 | P a g e
beverages, this is likely to impact the validity measure of the FFQ. Currently in Australia
there is no caffeine-specific FFQ, and due to the differences in caffeine amounts in
products and product availability, individual country FFQs are necessary for research
within a particular context.
26 | P a g e
Table 2.5: Validity and reproducibility values for available caffeine-specific food frequency questionnaires
Study (year) Country Critical evaluation Method Validity Reproducibility
A retrospective FFQ was FFQ v FD: k= 0.29-0.69 depending

Pregnant women
Boylan et al. administered at the same time as a on pregnancy level
England Thorough examination of foods, Not reported
(2008) three-day FD. Saliva samples were
drinks and drugs FFQ v saliva: k= 0.47-0.50
also taken during the three-day FD
Doesn’t allow for different sizes or

Bühler et al. Germany different flavours such as cola and Not reported Not reported Not reported
(2013) diet cola
Interview administered FFQ r= 0.56-0.69 (no p value reported) r= 0.61-0.74 (no p value reported)
Ferraroni et Italy Only coffee and tea
al. (2004) followed by a seven-day FD Same quintile: 39.2-91.1% Same quintile: 61.0-91.9%
Thorough examination of foods, A retrospective FFQ was

Rudolph et drinks and drugs administered at the same time as a FFQ v FD: r=0.82, p<0.01
Austria Not reported
al. (2012) three-day FD. Saliva samples were FFQ v saliva: r=0.43, p<0.01
Based on Boylan et al, 2008 also taken during the three-day FD
r=0.68-0.73, p<0.001 r=0.86, p<0.001
Same quartile: 50% Same quartile: 71%

Schliep et al. USA Only drinks – no chocolate 24-hour recall followed by a FFQ Sig. difference between FFQ and No sig. difference between FFQ1
(2013)
24-hour recall, p=0.01 and FFQ2, p=0.92
k=0.51 k=0.76
Abbreviations: sig, significant; FD, food diary; FFQ, food frequency questionnaire; USA, United States of America
27 | P a g e
2.4 Effects of Caffeine
If the consumption of coffee, tea, chocolate, energy drinks and soft drinks is combined,
the general population consumes a considerable amount of caffeine per day (Echeverri
et al, 2010). The effects of caffeine on the body are numerous however, in children,
there is a lack of understanding about the effects of caffeine on the body (Ludden &
Wolfson, 2010). Physiological effects after caffeine consumption include increased
blood pressure (Childs & de Wit, 2006; James, 2004; Robertson et al., 1978) and an
overall increase in heart rate (Robertson et al., 1978). Caffeine has further shown to
have addictive qualities with both adults and adolescents showing caffeine dependent
behaviours (Ludden & Wolfson, 2010; Satel, 2017; Berstein et al 2001). Furthermore, a
small amount of research has suggested that in children and adolescent’s caffeine could
be a gateway drug for future nicotine and alcohol consumption (Collins et al., 1997;
Pallanti et al., 2006). Acute caffeine consumption has also been shown to impact
cognition including improved simple and choice reaction time (Attwood et al., 2007;
Brice & Smith, 2002; Lieberman, Wurtman, Emde, Roberts, & Coviella, 1987; Smith,
2002), and improved attention (Smith, 2002; Regina et al., 1974; Clubley et al., 1979;
Lieberman, 1992). Mood changes after caffeine consumption have also been recorded,
with the ingestion of caffeine shown to increase anxiety (Attwood et al., 2007; Childs &
de Wit, 2006; Lieberman et al., 1987; Smith, 2002), vigour, arousal, and positive mood
(Childs & de Wit, 2006).
Caffeine consistently reduces feelings of fatigue or decreased alertness (Arnaud, 1999;

Attwood et al., 2007; Brice & Smith, 2001; Childs & de Wit, 2006; Smith, 2002; Stavric &
Gilbert, 1990), and these are the reported main reasons for caffeine consumption
amongst adults (Malinauskas, Aeby, Overton, Carpenter-Aeby, & Barber-Heidal, 2007).
Therefore, caffeine’s relationship with sleep has been a topic of discussion and research
in recent decades.
2.4.1 How caffeine impacts sleep

The most accepted biological mechanism explaining the association between caffeine
and sleep involves the neuromodulator adenosine (Basheer, Strecker, Thakkar, &
McCarley, 2004). Adenosine is a by-product of energy metabolism occurring during the
breakdown of adenosine triphosphate (ATP) and cyclic adenosine monophosphate
28 | P a g e
(cAMP), which functions in the regulation of sleep and wakefulness by inhibiting the
release of most brain excitatory neurotransmitters including dopamine (Bjorness &
Greene, 2009; Davis et al., 2003). Consequently, an adenosine agonist decreases
wakefulness and increases sleep, while an adenosine antagonist increases wakefulness
and decreases sleep (Bjorness & Greene, 2009). Caffeine functions as an adenosine
antagonist acting on adenosine receptor subtypes (A1, A2a, and A2b), thereby blocking
and counteracting adenosine activity (Echeverri, Montes, Cabrera, Galán, & Prieto, 2010;
Fredholm, Bättig, Holmén, Nehlig, & Zvartau, 1999). This produces a net increase in
central nervous system activity and leads to the release of dopamine, and increased
wakefulness and alertness (Bjorness & Greene, 2009; Echeverri et al., 2010; Smith,
2002). This mechanism is further supported by evidence of extracellular adenosine
concentration increasing during wake and decreasing during sleep (Basheer et al., 2004).
The duration and depth of sleep after wakefulness also appears to be modulated by
elevated levels of adenosine (Porkka-Heiskanen et al., 1997). However, sensitivity to the
adenosine inhibitory effects of caffeine varies across individuals (Rétey et al., 2006).
2.4.2 Caffeine and sleep: Adults

Research on the relationship between caffeine and sleep in adults has been expansive,
both experimentally and cross-sectionally (Clark & Landolt, 2016). Whilst experimental
research has provided a large amount of insight into the impact caffeine has on the
human body, metabolic response and behavioural reactivity varies between individuals.
The factors determining this individual response include genetic variation, medication,
tolerance, gender, age and pregnancy (Grosso & Bracken, 2005; Kaplan et al., 1997;
Rogers, 2007). Furthermore, people are becoming more aware about the effects
caffeine can have on their sleep. Therefore, while experimental research in caffeine is
important, it is also important to understand how this translates in day-to-day life.
Therefore, this review will focus on cross-sectional field studies to determine if caffeine
has a ‘real-world’ impact on sleep.
Of the previous cross-sectional studies, available in adults, sleep quality and total sleep
time (TST) are the most common measures of ‘good’ sleep. A summary of previous
cross-sectional research investigating caffeine and the association with sleep are shown
in Table 2.6. Overall, previous literature indicates that increased caffeine consumption
29 | P a g e
is associated with reduced sleep quality (Baker, Wolfson, & Lee, 2009; Hollander et al.,
2001; Huntley & Juliano, 2012; Lohsoonthorn et al., 2013; Mniszek, 1988), however
studies have also found no relationship (Brick, Seely, & Palermo, 2010; Doi, Inoue,
Minowa, Uchiyama, & Okawa, 2003; Hicks, Hicks, Reyes, & Cheers, 1983; Lund, Reider,
Whiting, & Prichard, 2010). In all previous studies, sleep quality was measured using the
Pittsburgh Sleep Quality Index (PSQI; Buysse et al., 1989), questions on subjective sleep
quality/satisfaction and presence of sleep problems such as periodic limb movement or
use of the St Mary’s Hospital Sleep Questionnaire (Ellis et al., 1981). Of these studies,
not one utilised a validated caffeine measure.
Previous literature also has shown that increased daily caffeine consumption is
associated with reduced sleep duration (Chinawa, Chukwu, & Obu, 2014; Dorrian et al.,
2011; Hicks et al., 1983; Irish et al., 2014; Kant & Graubard, 2014; Mossavar-Rahmani et
al., 2015; Sanchez-Ortuno et al., 2005), with only two studies finding no association
(Gottlieb et al., 2006; Heaton & Griffin, 2015). In these studies, sleep duration was
measured sleep diaries, actigraphy, and questions regarding bedtime, sleep time, wake
after sleep onset (WASO) and sleep onset latency (SOL). Of these studies, only five have
utilised validated caffeine measures (Dorrian et al., 2011; Heaton & Griffin, 2015; Irish et
al., 2014; Kant & Graubard, 2014; Mossavar-Rahmani et al., 2015).
Of the studies that used validated caffeine measures, three studies used food diaries to
measure caffeine intake (Dorrian et al., 2011; Heaton & Griffin, 2015; Irish et al., 2014).
Two studies used 24-hour dietary recall (Kant & Graubard, 2014; Mossavar-Rahmani et
al., 2015) and one of these studies used the multiple-pass method but did not state what
database they determined energy and macro- and micronutrients from (Kant &
Graubard, 2014), the second study utilised the Nutrition Data System for Research and
determined energy and micronutrients from the National Cancer Institute method
(Mossavar-Rahmani et al., 2015).
Firstly, Dorrian et al. (2011) set out to investigate and describe sleep, stress and
methods of coping in a population of nurses and midwives. The sample consisted of 62
participants (21 midwives, 41 nurses) with an average age of 38.9 ±11.9 years. The
participants were asked to complete a 14-day diary at the same time each evening
(approximately 18:00); the diary required the participants to record information
30 | P a g e
regarding sleep and caffeine use. Sleep information included self-reported sleep
duration (for all sleep periods, including naps), time at which participants fell asleep and
awoke, and whether they had disrupted sleep. In the diary the participants recorded
their caffeine consumption including beverages and grams of chocolate, converted into
grams of caffeine using the Australian food database Food Standards Australia New
Zealand (FSANZ). The study found that sleep duration was a significant predictor of
caffeine intake, such that reduced sleep was associated with increased intake.
Secondly, Irish et al. (2014) conducted a study examining waking health behaviours and
sleep characteristics in women going through menopause. Ages of the 303 participants
ranged from 48-58 years with an average of 52.2 ±2.2 years. For seven consecutive
days, each night before bed the participants were required to fill in a diary that recorded
the number of caffeinated beverages consumed that day. Sleep was measured with the
use of a sleep diary and actigraphy. The study found that increased number of caffeine
beverages across the seven days predicted longer SOL and, similar to Dorrian et al.
(2011), shorter TST.
Thirdly, Heaton and Griffin (2015) set out to describe caffeine use among truck drivers
who habitually consume caffeine. Final analysis included 97 truck drivers (male: 95%),
with a mean age of 44.2 ±12.0 years. For 30 days’ participants noted when they
consumed a caffeinated beverage or food. Food was later removed from analysis and
therefore total caffeine consumption estimates included only beverages and were
stated in ounces of caffeine drink. Daily sleep was measured with the use of a sleep
diary and actigraphy. Interestingly, and in contrast to the above studies, there was no
association between TST and caffeine use, regardless of whether the caffeine was
consumed within five hours of sleep. The differences between the studies could be due
to the measure of caffeine, for example Dorrian et al. (2011) included caffeinated food
in the total caffeine value whereas Heaton and Griffin (2015) only included caffeinated
beverages. Furthermore, the study by Heaton and Griffin (2015) mostly consisted of
males in comparison to Dorrian et al. (2011) and Irish et al. (2014) which included mostly
females.
The benefit of food diaries is that the food intake on a diet record is recorded by the
participant at the time the food is eaten to minimise the reliance on memory allowing a
31 | P a g e
high precision of estimating participants’ nutrient intake (Thompson & Byers, 1994;
Willett, 2012; section 2.3). Both Dorrian et al. (2011) and Irish et al. (2014) asked their
participants to complete their diary once per day which could have reduced the accuracy
of the recording, and it was unclear how often the truck drivers in Heaton and Griffin’s
(2015) study were expected to record in the diary. Furthermore, all three studies used
populations who are well known for their problematic sleep whether due to
physiological changes (women with menopause) or shift work (truck drivers and
nurses/midwives). Consequently, research using validated sleep and caffeine measures
is still required to look at a healthy sample representative of the typical population.
As previously mentioned only two studies have utilised 24-hour recall when measuring
caffeine intake (Kant & Graubard, 2014; Mossavar-Rahmani et al., 2015). Kant and
Graubard (2014) examined the association of sleep duration with eating behaviours in a
sample of 15,199 men and women all over the age of 20 years (mean age not stated).
An interview with participants was conducted and diet was measured using 24-hour
recall via the automated multiple-pass method to collect information on the time an
eating episode began and description, amounts, and source of all foods and beverages
reportedly consumed by the respondent in the preceding 24 hours. Caffeine was
reported in milligrams per day. During the 24-hour recall participants were asked about
the duration of sleep on days that they worked, however weekend sleep duration was
not recorded. Sleep was categorically organised as less than six hours (short), seven to
eight hours (average), and greater than nine hours (long). It was found that caffeine
intake was highest in short-duration sleepers and lowest in long-duration sleepers.
Mossavar-Rahmani et al. (2015) investigated diet and sleep patterns in Hispanics/Latinos

in the USA. There were 11,888 Hispanic/Latino men and women recruited with an
average age of 45 years (range 18-74 years). Sleep duration was determined by asking
participants about bedtimes and wake times on both weekdays and weekends. Sleep
duration was computed as the difference between wake-up time and bedtime and a
weighted average of weekday and weekend durations was computed for habitual sleep
duration. Sleep was again categorically organised into short (three to six hours),
intermediate (six to nine hours) and long duration (9-14 hours). Diet was measured
using two separate 24-hour recalls (second occurring five days to three months later).
32 | P a g e
Caffeine was expressed as mg per 1000 calories. Similar to Kant and Graubard (2014),
the study found that caffeine intake was significantly lower in long sleepers compared to
other sleep duration groups.
Both of these studies gathered large samples of the population and showed very similar
results; caffeine intake is greater in people who have shorter sleep duration. While both
studies utilised validated measures of diet recall, neither study utilised valid measures of
sleep duration. Furthermore, sleep duration was organised categorically based on hours
of sleep into ‘short’, ‘average’ and ‘long’ sleep and therefore variance amongst TST was
lost. As discussed in section 2.3, 24-hour recalls can be costly and time consuming and
are therefore not ideal for such large samples. Food frequency questionnaires are often
used as a quick, low burden option for gathering large amounts of data on one nutrient.
Overall, no study has utilised a validated caffeine-specific FFQ and a validated sleep
measure to determine the relationship between these two factors in day-to-day life of a
healthy sample of adults.
33 | P a g e
Table 2.6: A summary of the literature regarding cross-sectional studies investigating the association between sleep and caffeine in adults
Authors Findings
↓ decreased/lower
(date) Participants Study design Sleep measure Caffeine measure Mean Values
↑ increased/higher
Origin → no association
Baker et al. n= 959 Cross- Questions regarding Question about daily soft drink, Not reported Caffeine Consumption (CC)
(2009) M ±SD: 45 ±11.7y sectional subjective sleep coffee and tea consumption ↓ sleep quality
Range: 18-64y survey quality, sleepiness,
USA Reported as number of
Males: 0% and any diagnosed
caffeinated beverages
sleep disorders
consumed per day
Brick et al. n= 314 Cross- PSQI Caffeine intake was measured Coffee cups per day: 1.67 CC → no relationship to
(2010) M ±SD: 27.8 ±4.0y sectional by amount of coffee, soda, ±1.48 global or subscale measures
Range: 21-43y survey caffeinated tea, and energy of PSQI
USA PSQI global score: 6.37 ±2.57
Males: 42.1% drinks consumed in a typical day
Reported as the number of
equivalent cups of coffee per
day
Chinawa et al. n= 241 Cross- PSQI Not reported Median sleep duration CC ↓ TIB (r=−0.148, p<0.03)
(2014) M ±SD: 28.5 ±5.3y sectional Weekdays: 6h
Range: 20-26y survey Weekends: 7h
Nigeria
Males: 62%
Curless et al n= 181 Cross- PSQI Questions about their daily Median CC: 360mg per day CC → no association with
(1993) Median: 68y sectional consumption in the previous sleep quality
Range: 44-98y survey month of coffee, tea, cocoa,
UK
Males: 57% cola, and chocolate-containing
foods
Reported as mg per day
34 | P a g e
Authors Findings
↓ decreased/lower
Origin
→ no association
Doi et al. n= 1,889 Cross- PSQI – Japanese Not reported Not reported CC → not associated with
(2003) M ±SD: N/A sectional Version periodic limb movement
Range: 20+y survey syndrome
Japan Males: N/A
Dorrian et al. n= 62 Field Sleep diary 14-day food diary Sleep duration: CC ↓ Sleep duration (95% CI:
(2011) M ±SD: 38.9 Workdays -4.94; -0.95, p=0.004)
±11.9y Reported as grams of caffeine Nurses: 7.9 ±2.6h
Australia Range: N/A per day Midwives: 6.5 ±2.1h
Males: 15%
Days off
Nurses: 8.7 ±2.5h
Midwives: 7.1 ±1.9h
Gottlieb et al. n= 5,910 Cross- Sleep Habits Cups of coffee per day. Median TST: 7h CC → No association TST
(2006) M ±SD: 63.1 sectional Questionnaire
±10.7y survey Reported as cups of caffeine per
USA Range: 40-100y day
Males: 48%
Heaton & n= 97 4 week Actigraphy and sleep 30-day food diary. On duty CC → no association with TST
Griffin (2015) M ±SD: 44.2 observational log TST: 399.96mins
±12.0y Reported in ounces of caffeine Caffeine drinks: 16.4 ounces
USA Range: N/A drink
Males: 95% Off duty
TST: 376.99mins
Caffeine drinks: 15.1 ounces
35 | P a g e
Authors Findings
↓ decreased/lower
Origin
→ no association
Hicks et al n= 170 Cross- Questions asking Question asking the frequency Not reported CC ↓ TST
(1983) M ±SD: N/A sectional about TST and sleep of coffee, tea and cola
CC → No relationship with
Range: University survey satisfaction consumption per day 2
USA sleep satisfaction (χ (4)=6.99)
Undergraduates
Males: N/A Reported as number of caffeine
drinks per day
Hollander et n= 436 Longitudinal St. Mary’s Hospital Question asking about typical CC: 0.64 ±0.71 cups per day CC ↓ sleep quality (OR: 1.25;
al. (2001) M ±SD: 41 ±3.5y cohort study Sleep Questionnaire consumption over the past year 95% CI: 1.04, 1.49; p=0.02)
Range: 35-49y assessing sleep
USA Reported as cups of caffeine per
Males: 0% quality
day
Hughes & n= 144 Cross- VAS: Very Poor vs. Questions asking daily CC: 805mg (SD not reported) CC ↓sleep quality
Boland (1992) M: 34y sectional Excellent Sleep consumption of cups of coffee,
Range: N/A survey tea and cans of cola
Canada
Males: 100%
Converted to mg/day
Huntley & n= 1,046 Cross- PSQI Questions asking number of Not stated CC ↓ sleep quality
Juliano (2012) M ±SD: 27.5 ±8.8y sectional servings, typical serving size,
Range: N/A survey typical brand, and number of
USA
Males: 23% days per week they consumed
coffee, tea, caffeinated soft
drinks, energy drinks, chocolate,
foods, and caffeine-containing
medications/dietary
supplements
Reported as mg per day
36 | P a g e
Authors Findings
↓ decreased/lower
Origin
→ no association
Irish et al. n= 303 Longitudinal Actigraphy seven-day food diary CC: 1.49 ±1.31 cups per day CC ↑ SOL (p=0.01)
(2014) M ±SD: 52.2 Reported as the number of SE: 76.49 ±10.64% CC ↓ TST (p=0.002)
USA ±21.5y caffeinated beverages per day SOL: 25.88 ±28.45mins CC → no association with
Range: 48-58y WASO: 57.26 ±27.07mins WASO, SE and fragmentation
Males: 0% TST: 351.76 ±56.81 mins index
Fragmentation index: 32.24
±12.16
Kant & n= 1,519 Cross- Sleep duration 24-h dietary recall CC per day per sleep duration CC ↓ TST (p=0.0001)
Graubard M ±SD: N/A sectional Reported as mg per day category:
(2014) Range = ≥20y survey ≤6h: 200mg
USA Male = 50% 7-8h: 184mg
≥9h: 160mg
Lohsoonthorn n= 2,854 Cross- PSQI Question regarding whether PSQI score: 5.76 ± 2.50 CC ↓ sleep quality (p<0.001)
et al. (2013) M ±SD: 20.3 ±1.3y sectional they consumed more than one
Thailand Range: N/A survey caffeinated drink per week and
Males: 33% if so the specific type of energy
drink
Reported as yes/no
Lund et al. n= 1,125 Cross- PSQI & ESS Question asking about average TST: 7.02h ±1.15h CC → no difference between
(2010) M ±SD: N/A sectional frequency and intake of caffeine PSQI groups (p=0.59) and
USA Range: 17-24y survey drinks per week/weekend day was not a significant
Males: 37% Reported as number of drinks predictor of the PSQI score
per day (p=0.213)
Authors Participants Study design Sleep measure Caffeine measure Mean Values Findings
37 | P a g e
(date) ↓ decreased/lower
Origin ↑ increased/higher
→ no association
Mniszek n= 2,696 Cross- Questions asking the Question asking how many cups Not reported 1-4 cups of coffee within 6h
(1988) M ±SD: N/A sectional participants of tea and coffee consumed of bedtime had shorter SOL,
Range: 20-45y survey regarding SOL, each evening less WASO and less early
UK Males: 56.5% WASO, early waking, compared to those
awakening, Reported as cups of coffee, or who drank more coffee or
satisfaction with cups of tea per day abstained
perceived sleep Tea consumption ↑ SOL and
duration
↑ WASO
Mossavar- n= 11,888 Cross- Sleep Heart Health 24-h dietary recalls TST: 8.0 SE ± 0.02h. CC ↓ sleep duration (p<0.05)
Rahmani et M ±SD: 45 ±14y sectional Study Sleep Reported as mg per 1000kcal
al. (2015) Range: 18-74y survey Questionnaire -
Male: 36.4% Spanish version
USA
Sanchez- n= 1,498 Cross- Questions asking Question asking about daily CC: 225 ±161 mg per day Whole sample:
Ortuno et al. M ±SD: 51 ±3.2y sectional about SOL, nocturnal consumption of caffeinated equivalent to 2.6 ±1.9 cups of CC ↓ TST (r=-0.058, p<0.05)
(2005) Range: 44-58y survey awakenings, TIB and beverages, including coffee, tea coffee 97% of the sample (that is
Males: 64% TST and sodas TST: 413 ±48.5 mins those consuming <8cups of
France
Reported as the number of caffeine per day):
equivalent cups of coffee per CC → TST (r=-0.004, p=0.87)
day CC↓ TIB (r=-0.13, p<0.001)
CC ↑SE (p<0.01)
CC → No relationship with
daytime sleepiness
Abbreviations: ↑, increase or higher; ↓, decrease or lower; →, no association; %, percentage; CC, caffeine consumption; CI, confidence interval; ESS, Epworth sleepiness
scale; h, hours; KSS, Karolinska sleepiness scale; M, mean; mg, milligrams; mins, minutes; mths, months; n, number of participants; N/A, not available; PSQI, Pittsburgh
Sleep Quality Index; SD, standard deviation; SE, sleep efficiency; SE, standard error; SOL, sleep onset latency; TIB, time in bed; TST, total sleep time; UK, United Kingdom;
USA, United States of America; VAS, visual analogue scales; WASO, wake after sleep onset; y, years
38 | P a g e
2.4.3 Caffeine and sleep: Children
In children, there is less research on the impact of caffeine and sleep as compared to
adults. Of the available research, most is cross-sectional by design, and until recently
has focussed on adolescents and older children (see a summary of the studies in Table
2.7). Recent studies have shown that school-aged children are commonly consuming
caffeine (Australian Bureau of Statistics, 2014b; Commonwealth Scientific Industrial
Research Organisation [Australia]) et al., 2008), and it is therefore timely and important
to understand the effects this may have on their day-to-day function due to the
numerous effects sleep has on child health, well-being and performance (see section
1.6). In comparison to adults and adolescents, school-aged children (5-12 years) may
not drink caffeinated beverages to purposefully stay awake, or to self-medicate for
tiredness or loss of sleep, but rather the consumption of caffeine may disturb their sleep
leading children to become sleepy during the day (Calamaro et al., 2012).
Total sleep time

Total sleep time (TST) has been an important sleep measure when investigating the
effects of caffeine. Of the studies that have looked at caffeine and TST in children under
the age of 18 years, the results have mostly shown that increased caffeine consumption
reduces TST (Aepli, Kurth, Tesler, Jenni, & Huber, 2015; Boergers, Gable, & Owens, 2014;
Calamaro et al., 2012; Huang, Wang, & Guilleminault, 2010; Li et al., 2010; Lodato et al.,
2013; Mindell, Meltzer, Carskadon, & Chervin, 2009; Pollak & Bright, 2003; Warzak,
Evans, Floress, Gross, & Stoolman, 2011). Only one study found that caffeine has no
effect on TST in children under 13.3 years of age, but did show decreased TST in children
older than 13.3 years (Drescher, Goodwin, Silva, & Quan, 2011). Another study found
that caffeine consumption had no relationship with TST in females but a negative
relationship in male children such that greater caffeine intake was associated with a
shorter TST (Paciencia, Barros, Araujo, & Ramos, 2013).
Of the studies focussing on school-aged children (Calamaro et al., 2012; Drescher et al.,
2011; Li et al., 2010; Mindell et al., 2009) the majority found that caffeine consumption
was associated with less sleep. Each of the studies used large cohorts to investigate the
relationship; 319 children aged 6-17 years (Drescher et al., 2011), 625 children aged 6-10
39 | P a g e
years (Calamaro et al., 2012), 637 children aged 6-10 years (Mindell et al., 2009) and
20,778 children aged 5-12 years (Li et al., 2010).
In the very large study of Chinese school-aged children by Li et al. (2010) a validated
sleep questionnaire, Child Sleep Habits Questionnaire (Owens et al., 2000), was used
and caffeine was measured with an author-devised question asking about caffeinated
drinks consumed after 18:00 (usually, often, and occasional/no). The study found that
caffeine after 6pm was associated with more than 20% increased odds of short sleep
duration (OR=1.22, CI:1.05–1.41, p<0.001).
Using telephone interviews Calamaro et al. (2012) enquired about children’s normal
sleep schedule during the previous two weeks, and TST was determined as the reported
number of hours the child slept at night and during the day. Questions were asked
about their child’s sleep hygiene, including a question about the usual number of
caffeinated beverages (e.g., Coke, Pepsi, Mountain Dew, and coffee or iced tea) that the
child drank each day. Children who drank caffeinated beverages had on average 15
minutes less sleep per night than children who did not drink caffeinated beverages (b=–
0.27, p=0.002).
Mindell et al. (2009) examined the associations between sleep hygiene and sleep
patterns in children aged 0-10 years. Using telephone interviews, parents were asked
questions regarding SOL, TST at night, night awakening, bedtime routine and late
bedtimes, as well as number of caffeinated beverages consumed per day. School-aged
children who consumed at least one caffeinated beverage per day reported, on average,
sleeping almost 20 minutes less than children without daily caffeine intake
(F[1,630]=17.55, p=0.001).
Drescher et al. (2011) set out to investigate the relationships between important risk
factors for obesity in adolescents including dietary intake and sleep duration. Children
and parents completed a battery of questionnaires including two validated sleep and
diet measures; the Sleep Habits Questionnaire (Wolfson & Carskadon, 1998), and the
Rockett Youth Adolescent Questionnaire (YAQ; Rockett et al., 1997), which provided
amount of caffeine (in mg) consumed. It is important to note the YAQ does not include
energy drinks. This study originally collected data in children aged 6-11 years and then
40 | P a g e
conducted a five year follow up; during analysis, the cohort was grouped together and a
median split of the group was done to compare children under and above the age of
13.3 years. In contrast to the above-mentioned studies, TST and caffeine consumption
were not significantly related in children under the age of 13.3 years (r=0.005; 95% CI:
−0.379, 0.389; p=0.979).
In comparing the above-mentioned studies, it is evident that majority of studies

focussing on school-aged children indicate that increased caffeine consumption is
associated with reduced TST. Interestingly, the only study to utilise validated
questionnaires for both sleep and caffeine (Drescher et al., 2011) showed no
relationship between TST and caffeine consumption. However, measures of caffeine for
this study did not include energy drink intake, and it was unclear if chocolate
consumption was included. Asking parents and children about caffeine consumption
assumes that everyone is aware of what products include caffeine. Another
consideration is that while TST is a reliable measure of sleep quantity in children, there
are other sleep variables that also need to be considered, particularly those pertaining
to sleep quality.
Sleepiness
Sleepiness during the day is characterised by a persistent desire or ability to fall asleep
and general lack of energy. Most studies have found that increased caffeine
consumption is associated with greater daytime sleepiness and increased fatigue or
tiredness (Astill, Verhoeven, Vijzelaar, & Someren, 2013; Giannotti, Cortesi, Sebastiani, &
Ottaviano, 2002; Gibson et al., 2006; Huang et al., 2010; James, Kristjansson, &
Sigfusdottir, 2011; Kristjansson, Sigfusdottir, Allegrante, & James, 2011; Orbeta,
Overpeck, Ramcharran, Kogan, & Ledsky, 2006). Huang et al. (2010) found that coffee
intake in 17-18 year-olds was significantly correlated with school daytime sleepiness but
this same relationship did not occur in younger groups. Furthermore, one Swiss study
showed no relationship between subjective and objective tiredness and caffeine
consumption in children aged 11-17 years (Aepli et al., 2015).
Only one study to date has investigated caffeine consumption and daytime sleepiness in
school-aged children aged 5-12 years (Calhoun et al., 2011). The study set out to
identify associations between demographic, emotional, and medical factors and the
41 | P a g e
quantity and quality of sleep in young children with excessive daytime sleepiness.
During the study children underwent a nine-hour polysomnogram (PSG), and parents
completed the Paediatric Sleep Questionnaire. Caffeine was reported as ‘weekly
caffeine intake’, however it was unclear how caffeine was measured and there were no
reported means or ranges of weekly caffeine values. The final sample of 508 children
consisted of 431 children without excessive daytime sleepiness and 77 children with
excessive daytime sleepiness. Weekly caffeine intake was not reported to be
significantly associated with excessive daytime sleepiness (r=0.05; p=0.58; OR: 1.2; CI
0.66, 2.1).
Overall these associations suggest that while older children may utilise caffeine to
reduce daytime sleepiness, the same association does not appear in younger children.
This further supports the hypothesis that younger children are not self-medicating with
caffeine to reduce tiredness or fatigue during the day.
Sleeping problems and other sleep quality variables

As discussed in section 1.4, sleep quality is also an important consideration in addition to
sleep duration when considering good sleep. Sleep quality is measured by considering
the number of sleep disturbances, as they reduce sleep quality and can be a contributing
factor to chronic sleep deprivation. Overall, caffeine consumption has been shown to
increase sleep disturbances such as nocturnal awakenings and WASO (Pollak & Bright,
2003; Li et al., 2014), generic sleep problems (Kristjansson, Sigfusdottir, Mann, & James,
2014), and difficulty sleeping (Orbeta et al., 2006). However, studies have shown that
caffeine consumption has no impact on sleep difficulty (Quach, Hiscock, & Wake, 2012),
restless legs syndrome (Zhang et al., 2014), and enuresis (Warzak et al., 2011).
Specifically looking at school-aged children, there is not a wide range of studies
regarding caffeine consumption and different sleep quality variables.
Generic sleep problems were studied by Quach et al. (2012) and Kristjansson et al.
(2014), with both studies aiming to specifically determine if caffeine intake was
associated with sleep problems. Quach et al. (2012) collected complete questionnaires
from 1,512 children aged four to eight years (mean: 5.7 ±0.4 years). The study utilised
the validated School-Aged Sleep Habits Questionnaire (Wolfson & Carskadon, 1998) and
asked about different sleep problems such as sleep difficulty more than four nights a
42 | P a g e
week, night time asthma/wheezing, snoring, going to sleep, not happy to sleep alone,
wakes up at night, bedwetting, nightmares/night terrors, tired in the morning, and
‘other’ sleep problems. An additional question was asked regarding caffeine
consumption after school and was coded as either yes/no. The study found that
caffeine use after school was not associated with any of the reported sleep problems.
In contrast, Kristjansson et al. (2014) asked 11,132 Icelandic children aged 10-12 years
two questions regarding intake of caffeine, sugar sweetened beverages, “How many
cans/bottles/or glasses of cola drinks (e.g., Coca Cola, Pepsi Cola)” and “How many
energy drinks (e.g., Red Bull, Magic, Burn, Monster, XL) do you typically consume each
day?” There were six categorical responses ranging from none through to six or more
glasses/cups. Caffeine intake was weighted, with greater values assigned to drinks that
typically contain more caffeine, but it was unclear if the authors expressed caffeine in
number of glasses or milligrams of caffeine. Sleeping problems were measured with a
question asking, “How often, if ever, have you experienced sleeping problems during the
last seven days?” The responses formed two categories: ‘never/almost never/seldom’
and ‘sometimes/often’. The study found a dose–response relationship between cola
and energy drink consumption and the frequency of sleeping problems in children.
Specifically, as cola and energy drink consumption increased the number of sleeping
problems in children increased.
Specific sleep problems, nocturnal awakenings and enuresis were investigated by Li et

al. (2014) and Warzak et al. (2011) respectively. Li et al. (2014) set out to investigate the
factors associated of frequent nocturnal awakening in 20,505 children aged 8.98 ±1.59
years. Using the Children’s Sleep Habits Questionnaire, child sleep behaviours were
measured and a question on whether the child consumed drinks with caffeine after
18:00 (usually, often, and occasional/no) was included. The study found that frequently
having drinks with caffeine after 6pm was significantly associated with an increased
likelihood of nocturnal awakening.
Warzak et al. (2011), on the other hand, asked a sample of 228 parents (children aged 5-
12 years) about their child’s caffeine consumption and if their child wet the bed.
Respondents reported their child’s average daily consumption of various beverages and
snacks, emphasizing those that contained caffeine. Parents indicated whether a child
43 | P a g e
consumed an item and reported how many servings, and of what size, were consumed
per day. Illustrated depictions of serving sizes were provided to help parents report
accurate data. The survey also included items that measured enuresis. Caffeine
consumption was not significantly correlated with the number of nights the child wet
the bed.
Taken together the studies reviewed in this section highlight the inconsistencies in the
literature. This could be because only a few of the studies have used validated
questionnaires for both sleep and caffeine. Majority of studies indicate that for children
who consume caffeine, TST is reduced. The studies also indicate that caffeine
consumption may decrease sleep quality as measured by nocturnal awakening and sleep
problems in children aged under 12 years. However, further research is needed on
other sleep quality variables with the use of validated questionnaires. Furthermore, due
to the disruption that caffeine has on sleep in children, it is also important to consider
the relationship between caffeine and behaviour.
44 | P a g e
Table 2.7: A summary of the literature regarding cross-sectional studies investigating the association between sleep and caffeine in children
Authors (date) Participants Study Sleep measure Caffeine measure Mean Values Findings
Origin design ↓ decreased/lower
→no association
Calamaro et al. n= 625 Cross- Questions regarding Question regarding usual Drinks per day: 0.4 ±0.7 CC ↓TST by 15mins compared
(2012) M ±SD: 8.6 sectional sleep schedule, number of caffeinated to non-CC (b=-0.27, p=0.002)
TST: 9.5 ±1.0h
±1.4y survey bedtime routine and beverages consumed
USA
Range: 6-10y TST each day
Males: 51.1%
Reported as number of
drinks per day
Calhoun et al. n= 508 Cross- PSG and Paediatric Not reported Not reported CC → not associated with
(2011) M ±SD: 102.0 sectional Sleep Questionnaire excessive daytime sleepiness
±0.08mths survey (p=0.58)
School – aged children
USA
Range: 5-12y
Males: 51.8%
Kristjansson et n= 11,132 Cross- Question regarding Question regarding daily Not reported Cola and energy drink
al. (2014) M ±SD: N/A sectional frequency of sleep cola and energy drink consumption ↑ sleeping
Range: 10-12y survey problem symptoms consumption (none, less problems
Iceland
Males: 50.3% than one per day and at
least one per day)
Li et al. (2014) n= 20,505 Cross- Child Sleep Habits Question regarding Not reported CC after 6pm ↑nocturnal
M ±SD: 8.98 sectional Questionnaire consuming drinks with awakenings
China
±1.59y survey caffeine after 6:00pm
Range: 5.00- (usually, often
11.92y occasional/no)
Males: 49.5%
45 | P a g e
→no association
Li et al. (2010) n= 20,778 Cross- Child Sleep Habits Question regarding Males TST: 9.20h ±50mins Caffeine after 6pm is a
M ±SD: 9.00 sectional Questionnaire consuming drinks with Females TST: 9.21h ±49mins predictor of shorter sleep
China
±1.61y survey caffeine after 6:00pm duration (β=0.160, p< 0.001)
Range: 5.08- (usually, often
12.00y occasional/no)
Males: 49.5%
Mindell et al. n= 1,437 (637 Cross- Questions regarding Questions on daily Not reported Daily CC ↓ TST by 20 mins (p=0
(2009) school-aged) sectional SOL, TST, WASO, caffeine consumption, .001)
M ±SD: N/A survey bedtime routine, late including caffeinated
School – aged children
USA
Range: 0-10y bedtimes beverages such as Coke,
Males: 51% Pepsi, Mountain Dew,
coffee, iced tea (none
and at least one per day)
Quach et al. n= 1,512 Cross- Questions regarding Caffeine consumption Not reported CC after school→ not
(2012) M ±SD: 5.7 ± sectional any sleep difficulties after school: yes/no associated with sleep problems
0.4y survey
Australia
Range: 4.1-8.0y
Male: 51.9%
Warzak et al. n= 201 Cross- Sleep history and Average daily 5-7y (n=104): CC –‘vely correlated with TST
(2011) M ±SD: N/A sectional enuresis consumption of various TIB: 9.46h (p=0.02; unclear about
Range: 5-12y survey beverages Caffeine intake: 52mg/day direction of relationship)
USA
Males: 53.7%
Reported in mg/day 8-12y (n=97): CC → number of nights the
TIB: 8.70h child wet the bed
Caffeine intake: 109mg/day
46 | P a g e
→no association
Aepli et al. n= 32 Cross- Seven-day sleep diary Seven-day caffeine food Non-habitual consumer of CC ↑bed time by 57min.
(2015) M ±SD: 14.4 sectional and beverages diary. caffeine: CC↓TIB by 53min
±0.4y survey Caffeine intake: 0.1 ±0.0
Switzerland Reported as mg/kg/day CC →no association with
Range: 11.8- mg/kg/day
subjective and objective
16.8y TIB: 9h 3mins ±16mins
tiredness
Male: 50%
Habitual caffeine consumer:
CC → no association with TIB
Caffeine intake: 2.5 ±0.4
variation between weekend
mg/kg/day
days and weekdays
TIB: 8h 10mins ±13mins
Astill et al. n= 24 Repeated Actigraphy and sleep Question asking number During holidays: CC → no association to any
Adolescents
(2013) M ±SEM: 17.63 measures diaries of caffeinated drinks TST: 7.38h ±12mins actigraphy sleep estimates
±0.10y study consumed previous day
Netherlands During school week: CC → not associated to
Range: N/A
TST: decreases 58 ±12mins subjective sleep quality
Male: 33.3%.
During stress week: CC ↑fatigue (p=0.04)
TST: decreases additional 18
±8mins
Boergers et al. n= 197 Repeated The School Sleep Weekly caffeine intake T1 At T1 & T2, students reporting
(2014) M: 15.6y measures Habits Survey. of all caffeinated items Weekdays TIB: 7.01 ±1.00h <8h of sleep had ↑ weekly CC
Range: grades study (foods and drinks) Weekends TIB: 9.04 ±1.16h than students reporting at least
USA th th Implemented an
9 -12 8h of sleep
experimental school Reported as number of T2
Males: 41%
schedule, measures caffeine items per week Weekdays TIB: 7.01 ±1.00h
two time points (T1 Weekends TIB: 9.04 ±1.16h
& T2)
47 | P a g e
→no association
Calamaro, et al. n= 100 Cross- The Adolescent The Adolescent Sleep, Median caffeine intake: CC → no association with sleep
(2009) Median: 15y sectional Sleep, Caffeine Caffeine Intake, and 141.0mg per day variables
Range: 12-18y survey Intake, and Technology Use
USA
Male: 42% Technology Use questionnaire
questionnaire
Reported in mg/day
Drescher et al. n= 319 Cross- Sleep Habits YAQ Not reported CC ↓ TST (r=-0.270, p=0.006)
(2011) M ±SD: 13.3 sectional Questionnaire and
Reported as mg of
±1.8y Range: survey home PSG
USA caffeine per day
10-17y
Male: 51.7%
Adolescents
Giannotti et al. n= 6,631 Cross- School Sleep Habits Substance use scale Not reported CC or tobacco use ↑ daytime
(2002) M ±SD: 16.7 sectional Survey (soda coffee or tea); sleepiness (β:0.16, p<0.001)
±1.7y survey every day, several times
Italy
Range: 14.1- every day, once or twice
18.6y a day, never
Male: 39.9%
Gibson et al. n= 3,235 Cross- ESS, bed and wake Not reported Not reported CC after 6pm ↑ sleepiness
(2006) M ±SD: 16.1 - sectional times, bedtime (p<0.0001)
Reported as caffeine
16.3 ±1.37 - survey routines, daytime
Canada after 6pm
2.26y sleepiness, and
Range: N/A questions eliciting
Male: 50.8% symptoms suggestive
of sleep disorders
48 | P a g e
→no association
Huang et al. n= 1,906 Cross- Paediatric Sleep Paediatric Sleep Weekday TST: 7.35 In 17-18y olds only coffee intake
(2010) M ±SD: N/A sectional Questionnaire - Questionnaire - Chinese ±1.23h ↑ daytime sleepiness
2
Range: 12-18y survey Chinese version version] – Coffee intake (χ =12.500 p=0.001)
Taiwan Weekend TST: 9.38
Males: 37.7%
Reported coffee intake ±1.62h Coffee intake ↓TST in whole
sample (r=-0.079, p<0.05)
James et al. n= 7,377 Cross- ESS Question regarding daily Not reported CC ↑daytime sleepiness
(2011) M ±SD: sectional caffeine consumption, (β=0.18, p<0.01)
th th
Range: 9 – 10 survey unclear how caffeine was
Iceland
grade reported
Males: 49.2%
Adolescents
Kristjansson n= 7,348 Cross- ESS Question regarding daily Not reported CC ↑ sleepiness for boys
et al. (2011) M ±SD: N/A sectional caffeine consumption, (β=0.27) and for girls (β=0.32)
Range: 14-15y survey unclear how caffeine was
Iceland
Males: 49.2% reported
Lodato et al. n= 1,522 Cross- Questions regarding FFQ Median caffeine CC ↓ TIB (p<0.001)
(2013) M ±SD: N/A sectional bedtimes and wake-up intake: 23.1mg/day
Reported in mg/day
Range: 13y survey times for weekdays
Portugal
Males: 46.5%
Orbeta et al. n= 15,686 Cross- Questions regarding Health Behaviour in School- Not reported Coffee and soda intake ↑
(2006) M ±SD: N/A sectional morning tiredness and aged Children survey; coffee difficulty sleeping by 1.9 times
th th
Range: 6 -10 survey difficulty sleeping and soda only (<1 per day, 1 and ↑morning tiredness by 1.8
USA
grade per day, 1 per week, and <1 times
Males: N/A per week)
49 | P a g e
→no association
Paciencia et n= 1,771 Cross- Sleep and wake times FFQ TST: 9.0 ±0.80h CC → not associated to TST in
al. (2013) M ±SD: N/A sectional female adolescents.
Reported in mg/day
Range: N/A survey
Portugal CC ↓TST in male adolescents
Males: 46.5%
(p=0.009)
Pollak & n= 191 Cross- Bedtime, wake time, Names of caffeine-containing Caffeine intake: CC ↓nocturnal sleep (p=0.001)
Bright (2003) M ±SD: N/A sectional times of naps, and items consumed in previous 52.7mg/day CC ↑WASO (p=0.044)
Adolescents
Range: 12-15y survey WASO day TST: 8.3h CC ↑daytime sleep (p=0.006)
USA
Males: 35%
Reported as mg/day
Zhang et al. n= 2,182 Cohort Questionnaire Question asking about regular Not reported Consumption of tea and coffee
(2014) M ±SD: 13.4 ±2.6y study regarding sleep consumption of tea, & coffee; → not associated with restless
Range: 6-18y environment, sleep never/seldom or legs syndrome
Hong Kong
Males: 48.1% habits and problems, sometimes/often
and excessive daytime
sleepiness
Abbreviations: ↑, increase or higher; ↓, decrease or lower; →, no association; %, percentage; CC, caffeine consumption; CI, confidence interval; ESS, Epworth sleepiness
scale; h, hours; kg, kilogram; M, mean; mg, milligrams; mins, minutes; mths, months; n, number of participants; N/A, not available; PSG, polysomnography, SD, standard
deviation; SOL, sleep onset latency; TIB, time in bed; TST, total sleep time; USA, United States of America; WASO, wake after sleep onset; YAQ, Rockett Youth/Adolescent
Questionnaire; y, years
50 | P a g e
2.5 Caffeine and Behaviour
Finally, a small amount of research has shown that caffeine intake in children impacts
behaviour. Specifically, it has been found that children with increased caffeine consumption
have increased internalising behaviour problems (Azagba, Langille, & Asbridge, 2014;
Fulkerson, Sherwood, Perry, Neumark-Sztainer, & Story, 2004; Guilleminault et al., 1976;
Richards & Smith, 2015). In children, heightened sensation seeking and elevated depressive
symptoms were associated with an increased likelihood of energy drink consumption in
8,210 Canadian children in grades 7-12 (Azagba et al., 2014). Fulkerson et al. (2004) found,
using a sample of 4,734 adolescents in the USA, that caffeine consumption was positively
associated with depressive symptoms among both males and females. Another study found
that coffee intake was associated with a lower chance of having depressive feelings (Flores,
Andrade, & Lima, 2000). However, this study only investigated caffeine from coffee intake
and located in Brazil where children start drinking coffee socially as early as three years of
age (Flores et al., 2000). Additionally, increased somatic complaints have been associated
with caffeine use, whereby females who reported high caffeine consumption were 1.4 times
more likely to complain of headaches, stomach-aches, and backaches compared with those
who consumed lower levels of caffeine (Ghandour, Overpeck, Huang, Kogan, & Scheidt,
2004).
Furthermore, externalising behaviour problems have also been linked to caffeine intake.
Increased aggression has been shown to be related to higher caffeine consumption
(Guilleminault et al., 1976; Kim, 2013), and increased caffeine use was strongly related to
increased violent behaviours and conduct disorder in one study (Gupta et al., 2002). This
was further supported by a population-wide cross-sectional study conducted with Icelandic
tenth graders, aged 15-16 years, utilising self-report caffeine and behaviour questionnaires
(James, Kristjansson, & Sigfusdottir, 2015; Kristjansson, Sigfusdottir, Frost, & James, 2013).
It was found that higher caffeine consumption was associated with increased anger, violent
behaviour and conduct disorders in both boys and girls (James et al., 2015; Kristjansson et
al., 2013).
51 | P a g e
2.6 Caffeine, sleep and behaviour in children
In children, only one study has looked at the interaction of caffeine, sleep and behaviour.
Kristjansson et al. (2011) utilised population-wide cross-sectional data on a range of
demographic and health-related variables in children aged 14-15 years in all primary schools
in Iceland. The consumption of caffeine was measured as the number of glasses or cups of
coffee, tea, cola drinks, or energy drinks that contained caffeine. The responses were
marked on a seven point Likert scale from ‘never’ to ‘six glasses/cups or more’, however it
was unclear how caffeine consumption was calculated for analyses. Sleepiness was
measured on the Epworth Sleepiness Scale, which was modified for use with children and
adolescents to assess daytime sleepiness (Chan et al., 2009; Melendres, Lutz, Rubin, &
Marcus, 2004). Anger, as the behaviour outcome, was measured on a five-item scale
headed with the sentence, ‘‘During the past week how often did the following statements
apply to you.” Responses were on a 4-point Likert scale ranging from ‘never’ to ‘often.’ A
structural equation model approach was used to identify direct and indirect effects between
the variables. Of the sample, 76.3% consumed caffeine on a typical day and 83.4% of the
sample indicated they had a high level of sleepiness. Caffeine was strongly related to
sleepiness for boys (β=0.27) and girls (β=0.32, critical ratio = 4.12, p<0.01) and furthermore
caffeine was moderately related to anger for both genders (β=0.16 for both). Sleepiness
was moderately related to anger for boys (β=0.19) and girls (β=0.22). The standardized total
effect between caffeine use and anger was moderate, and the indirect effects low. Overall,
the results suggested that about 43% of the total relationship between caffeine and anger
was due to mediation by sleepiness and licit substance use. The same ratio for girls was
48%. Consequently, almost half of the total relationship between caffeine and anger was
mediated by licit substance use and sleepiness.
While this unique finding indicates that caffeine should be considered when investigating
behavioural difficulties in children under 18 years of age, there are some limitations with
the interpretations that can be derived specific to caffeine. Firstly, the unique contribution
of sleepiness was not reported, as sleepiness and licit substance use were reported as a
combination. Due to the well reported relationship between sleep and behaviour (Beebe,
2011; Blunden & Beebe, 2006) these factors should be considered separately, as other licit
52 | P a g e
drugs such as alcohol have been known to change sleep architecture and reduce sleep
duration in adults (Roehrs & Roth, 2001), while smoking has been associated with risk
behaviours and depression/distress in adolescents (Tyas & Pederson, 1998). Also, no study
has investigated this relationship across a broader range of sleep factors, and similarly,
while this study looked at anger in children, it would be beneficial to determine if other
behavioural problems are associated with caffeine and sleep. Finally, the study did not
gather specific caffeine information and required the participants to already know what
beverages contain caffeine and which do not. The use of a validated assessment of caffeine
consumption will therefore be a valuable extension to this work in future research.
53 | P a g e
Chapter 3
Sugar and Sleep in Children
54 | P a g e
In the 1970s, dietary sugar and its implications for health became a controversial topic due
to a growing perception that high sugar diets were associated with hyperactivity, alterations
in brain chemistry, behavioural problems, thiamine deficiency, carbohydrate sensitivity,
obesity, diabetes and heart disease (Liebman & Moyer, 1980). By the 1980s the blame for
many diet-related health problems started to shift away from fat intake and towards sugar
consumption. Media has had a large influence on the negativity surrounding sugar and its
consumption. In 1979, the media in the United States of America (USA) reported a reduced
murder sentence for a man based on the defence that he consumed too much sugar and
was not in control of his actions, and the term ‘Twinkie Defence’ was coined (Pogash, 2003).
Newspaper articles in Australia have compared sugar consumption to drug use (Lambert,
2016), insinuated that sugar kills (Wood, 2016), and regularly suggested how to cut down or
‘quit’ sugar consumption (Burrell, 2016; Gameau, 2016). Furthermore, ‘Sugar Addiction’ has
been the topic of many best-selling books (Appleton, 1985; Appleton & Jacobs, 2009; Rufus,
2005). Sugar has subsequently been considered a major contributor to a myriad of health
problems including obesity (Malik, Pan, Willett, & Hu, 2013), Type-II diabetes and metabolic
disease (Malik et al., 2010), dental caries (Watanabe et al., 2014), and in children, sugar
intake has been assumed to play a role in behavioural disorders (Prinz, Roberts, & Hantman,
1980).
This chapter provides a summary of sugar consumption in Australian children, and

guidelines for sugar consumption. Furthermore, the chapter will discuss the impact of sugar
on sleep and behaviour in children, with a discussion on possible mechanisms for these
interactions. Finally, the empirical literature relating sugar and sleep in children is reviewed,
as is the literature relating sugar and behaviour.
3.2 Dietary Sugar
Macronutrients, namely carbohydrate, fat and protein, are the main components of all food
intake and are needed for growth and energy. Additionally, vitamins, minerals and water
are essential nutrients that provide further benefits for the body (Whitney, Rolfes, Crowe,
Cameron-Smith, & Walsh, 2011). When the body breaks down one gram of carbohydrate or
protein, it is equivalent to 16.7 kilojoules (kJ) of energy, and one gram of fat equates to
55 | P a g e
37.7kJ. Sugar is a type of carbohydrate and the term ‘sugar’ is traditionally used to describe
both monosaccharides and disaccharides that are found naturally in foods or added during
processing. There are three different types of monosaccharides: glucose, fructose and
galactose (Figure 3.1). They all are important in nutrition and all have the same chemical
formula but different chemical structures, which accounts for differing sweetness. Glucose
is more commonly known as blood sugar and serves as an essential energy source for all the
body’s activities. Fructose occurs naturally in fruits and honey, and is commonly added to
processed foods such as soft drinks and ready to eat cereals and desserts, particularly those
imported into Australia. Galactose occurs naturally but tends to partner with other
monosaccharides to form a disaccharide, and is rarely found singularly in foods (Whitney et
al., 2011).
Disaccharides are made up of two different monosaccharides, one of which is always

glucose. There are three different disaccharides - maltose, sucrose, and lactose. Maltose is
made up of two glucose units, and is found only in a few foods such as barley. Sucrose is
made up of fructose and glucose, and is commonly used in the form of table sugar. Finally,
lactose is made up of galactose and glucose, and is the principal carbohydrate of milk.
Sucrose
Disaccharides Lactose
Maltose
Simple
Sugars
Carbohydrate Galactose
Complex
Food
Starches and Fibre

Fat Monosaccharides Fructose
Protein Glucose
Figure 3.1: A pictorial breakdown of foods into the basic sugar components.
The primary role of all carbohydrates in human nutrition is to supply the body’s cells with
glucose for energy. During digestion, the body breaks sugars down into the basic
monosaccharides for the body to absorb and use in the form of glucose. The initial
56 | P a g e
digestion begins in the mouth by the salivary enzyme amylase, and some of the glucose is
absorbed through the mouth lining. For the most part, glucose is absorbed through the
small intestine, where final splitting occurs by the major carbohydrate-digesting enzyme,
pancreatic amylase. Typically, within one to four hours after a meal all the sugars have been
digested, and after absorption monosaccharides are transported in the blood to the liver
and other tissues. As the blood from the intestines circulates through the liver, fructose and
galactose are converted to glucose. Glucose can be stored by the liver and muscles as
glycogen, or converted into fat if intake exceeds need.
Glucose is the metabolic fuel for most cells in the body and is the principal fuel for the cells
in the central nervous system (CNS). Of all the organs and tissues in the body, the CNS is
most dependent upon the continuous supply of glucose from the blood. The CNS modulates
hepatic glucose production via parasympathetic and sympathetic fibres to the liver
(DeFronzo & Ferrannini, 2001; Shimazu, 1987). Without ingested glucose, the body is forced
to break down glycogen stores and protein tissues to make glucose via gluconeogenesis,
alter energy metabolism to make ketone bodies from fats, or produce glucose from liver and
muscle via glycogenolysis (Tirone & Brunicardi, 2001). Glucose is indispensable for
maintaining the functional integrity of nervous tissue. Blood glucose regulation depends
primarily on two pancreatic hormones: insulin to allow the uptake of glucose from the blood
into the cells when levels are high, and glucagon to release glucose from glycogen stores
into the blood when circulatory glucose levels are low. Due to these hormones, glucose
homeostasis is regulated and levels remain relatively stable throughout the day despite
ingestion of dietary carbohydrates (Herman & Kahn, 2006; Tirone & Brunicardi, 2001).
3.2.1 Measurement of sugar consumption
The different ways to measure nutrients and dietary intake in field research were reviewed
in section 2.3Error! Reference source not found.. From these measures, total sugar intake
can be derived, as well as total energy intake, and from these two values sugar as a
percentage of energy can be calculated. Total sugars in grams includes those sugars
naturally present in foods such as in fruit and milk, as well as the sugars added to processed
foods and beverages. Total energy intake is the sum of all daily kilojoules consumed from
food and drink. Energy comes from macronutrients, such as fat, carbohydrate including
57 | P a g e
total sugars, dietary fibre, and protein. In adults, ethanol intake (from alcoholic beverages)
also needs to be considered as it also contributes to energy.
3.3 Sugar consumption guidelines
The National Health and Medical Research Council (NHMRC) Australian Dietary Guidelines
are derived using scientific-based evidence to provide information about healthy food
choices. The guidelines promote health and well-being by minimising the risk of diet-related
diseases within the Australian population (National Health and Medical Research Council,
2013). These guidelines are for healthy individuals and were developed from a
comprehensive review of the literature. The 2003 guidelines were the first revision of the
original dietary guidelines for children and adolescents in Australia published in 1995. The
2003 NHMRC dietary guidelines suggest that care should be taken to “Consume only
moderate amounts of sugars and foods containing added sugars” (Binns, Davidson, &
Forbes, 2003, p.156). The guidelines go on to say that consumption of up to 20% of energy
as total sugars is compatible with a healthy diet for Australians. The subsequent 2013
NHMRC dietary guidelines further recommend that intake of foods and drinks containing
added sugars (e.g. confectionary, sugar-sweetened soft drinks and cordials, fruit drinks,
vitamin waters, and energy and sports drinks) should be limited (National Health and
Medical Research Council, 2013).
Soon after the latest NHMRC guidelines were released, the World Health Organisation
(WHO) guidelines were developed for both adults and children. These guidelines include
sugar-specific recommendations, contain suggestions for clinical practice, and public health
policy, and help the public to make informed decisions (World Health Organisation, 2015).
The WHO guidelines make ‘strong recommendations’ to reduce intake of free sugars
throughout the life course, and, in both adults and children, reduce the intake of free sugars
to less than 10% of total energy intake. The guidelines also provide ‘conditional
recommendations’ for a further reduction of the intake of free sugars to below 5% of total
energy intake.
Of the three recent guidelines, only those published in 2003 (Binns et al., 2003) consider
total sugar intake. The 2013 NHMRC guidelines and the 2015 WHO guidelines only consider
added sugars. Unfortunately, the current Australian dietary database that is commonly
58 | P a g e
used for diet analysis (Food Standards Australia New Zealand, 2015) does not differentiate
between added and natural sugar and can only report total sugars. Therefore, all dietary
measures (food frequency questionnaire, 24-hour recall, food diaries) report a total sugar
value in grams. As a consequence of these limitations, this thesis will only report on total
sugars and refers throughout to the 2003 NHMRC guidelines (Binns et al., 2003).
3.4 How much sugar are children consuming?
1985 National Dietary Survey of Schoolchildren

The National Dietary Survey of Schoolchildren was the first national survey to collect dietary
intake information on school-aged children and was conducted between May and October
1985 (Cook, Rutishauser, & Seelig, 2001). The survey recruited 5,224 children aged 10-15
years. The survey was conducted across all Australian states and territories on all weekdays
(Monday to Friday) from a sample of schools where participants completed a 24-hour
dietary record. The survey found that boys were consuming, on average, 142g of sugar per
day and girls 115g; this equates to 24.5% and 25.3% of total energy intake, respectively.
1995 data from National Nutrition Survey

The National Nutrition Survey (NNS), conducted from February 1995 to March 1996, was
conducted across all Australian states and territories on all days of the week (Cook et al.,
2001). The survey involved an in-home 24-hour recall for those aged two years and over
and proxy interviews were conducted for children aged two to four years. Children aged 5-
11 years were asked to provide their own food intake data with the assistance of an adult
household member. The survey recruited a total of 13,858 participants.
The 1995 NNS found an increase in child sugar consumption compared to the 1985 survey.
Boys and girls were consuming, on average, 174g and 137g of sugar per day, respectively;
equating to 26.2% and 27.0% of total energy intake. The 1995 NNS stated that in terms of
sources of dietary sugar, the main reasons for the increase were significant increases in the
intake of cereal-based foods, confectionery and health bars, and sugar products and dishes
(Cook et al., 2001).
2007 data from Australian National Children’s Nutrition and Physical Activity Survey
Between February and August 2007, the Australian National Children’s Nutrition and
Physical Activity Survey (Australian Children’s Survey) data was collected as outlined in
59 | P a g e
section 2.2.1. Overall, the survey found that boys and girls aged four to eight years were
consuming, on average, 116.7g and 105.9g of sugar per day, respectively, and those aged 9-
13 years were consuming 145.9g and 124.4g of sugar, respectively. On average, children
aged four to eight years and 9-13 years had respective total sugar intakes of 24.3% (24.3%
for boys and 24.3% for girls) and 23.7% (23.6% for boys and for 23.9% girls) of total energy
intake. This proportion of dietary intake represented a decrease from the 1995 survey. The
Australian Children’s Survey found that in children nine years and over, consumption in
excess of guidelines was found for a variety of macronutrients/food groups including
saturated fat, vegetables, fruit, dairy and sugar. Only a minority of children met the
guidelines for having moderate sugar intake (<20% of total energy [Binns et al., 2003];
Figure 3.2).
Figure 3.2: Proportion of children meeting the sugar recommendations based on the 2003 dietary guidelines.
Notes: the data was from the Australian National Children’s Nutrition and Physical Activity Survey
(Commonwealth Scientific and Industrial Research Organisation [Australia] et al., 2008) and the dietary
guidelines were based on the 2003 NHMRC dietary guidelines (Binns et al., 2003).
2011-2012 data from Australian Bureau of Statistics

Australian Bureau of Statistics (ABS) survey data was collected as outlined in section 2.2.1.
The results showed that children under the age of 18 years consumed 112.4g of sugar per
day, with boys consistently consuming more sugar than girls. Overall, in children aged less
60 | P a g e
than 18 years, carbohydrate contributed 50% of total energy intake, with total sugars
contributing 23% of energy.
Figure 3.3: Proportion of total sugar from different food groups, adapted from Australian Bureau of Statistics (2014b).
Similar to the caffeine data reported by the ABS, sugar consumption was further broken
down into food groups. Overall, the food groups contributing the greatest amounts of
sugars for children under 18 (percent of total sugars) were: fruit products and dishes
(15.4%), soft drinks and flavoured mineral waters (9.6%), dairy milk (10.1%), fruit and
vegetable juices, and drinks (9.5%), sugar, honey and syrups (3.4%), and cakes, muffins,
scones and cake-type desserts (6.1%; Australian Bureau of Statistics, 2014c). The proportion
of total sugars from food groups changed across age groups (Figure 3.3), with fruit and
vegetable juices, and drinks appearing to remain consistent, fruit products and dishes
decreasing with age, and soft drinks and flavoured mineral waters increasing with age.
Another important consideration in children is the confectionery food group and the group
consisting of chocolate, fruit, nut and seed bars and muesli or cereal-style bars, which were
consumed by almost half of all children (49% of children aged four to eight years and 48% of
children aged 9-13-years). Other confectionery (mainly consisting of lollies) was consumed
by 11% of the population, but children aged less than 14 years were the greatest consumers
(17-18%; Australian Bureau of Statistics, 2014b).
61 | P a g e
Since the peak of sugar consumption reported in 1995, the more recent surveys indicate
sugar intake has since been decreasing (Figure 3.4). However, the 2011 ABS data still
reported that children were consuming more than the recommended amount of sugar, on
average.
Figure 3.4: Bar graphs illustrating the overall decline in [A] sugar consumption (g) and [B] sugar as a percent
of energy (%E).
Notes: Data from 1985 Survey and 1995 NNS (Cook et al., 2001), 2007 Australian Children’s Survey
(Commonwealth Scientific and Industrial Research Organisation [Australia] et al., 2008) and 2011-2012
Australian Bureau of Statistics (Australian Bureau of Statistics, 2014b).
62 | P a g e
3.5 Sugar and the body
When considering the negative effect sugar has on the body, there are two major health
concerns: obesity and dental caries. In children, sugar consumption has been documented
to impact dental caries as early as 1883, and the impact has been confirmed in a number of
studies since then (Burt & Pai, 2001; Newbrun, 1982; Selwitz, Ismail, & Pitts, 2007). For
dental caries to occur, the rate of demineralisation on the tooth must exceed the
remineralisation, an effect enhanced by frequent exposure to sugars, retention of sugary
foods in the mouth, and the number of eating occasions (Walsh, 2000). Similarly, sugar
consumption has been consistently linked to obesity in children (Chen, Beydoun, & Wang,
2008; Malik et al., 2013; Van Cauter & Knutson, 2008). Obesity is predominantly caused by
an imbalance between energy intake and expenditure, and commonly people consuming
high sugar foods do not adequately compensate for the added energy intake (Vartanian,
Schwartz, & Brownell, 2007). In particular, soft drink consumption has been associated with
increased energy intake, increased bodyweight, and lower intakes of milk, calcium, and
other key nutrients (Vartanian et al., 2007).
In adults, there is evidence to suggest that sugar intake is linked to Type-II diabetes and
metabolic disease (Malik et al., 2010), as well as colorectal cancer (Bostick et al., 1994).
However, Type-II diabetes and metabolic disease could also be explained by the increased
obesity among high sugar consumers (Levinson, 1995; Weiss et al., 2004; Wolf & Colditz,
1998). Similarly, colorectal cancer has been related to overall poor diet (Donaldson, 2004;
Ryan-Harshman & Aldoori, 2007), which can include high consumption of sugar (Vartanian
et al., 2007), thus making it difficult to specifically attribute the association to sugar.
Commonly, high sugar intake is blamed for poor sleep and behaviour in children.
3.6 Sugar and sleep
Determining the impact of diet on sleep in children is a relatively new area of research, and
there is a paucity of literature regarding this interaction. Experimental studies have
focussed on manipulating sleep duration and considering whether sugar intake changes as a
consequence (Beebe et al., 2013; Simon, Field, Miller, DiFrancesco, & Beebe, 2015). There
are currently no published studies experimentally manipulating sugar intake and measuring
sleep duration. Despite this, there are cross-sectional studies that have set out to
63 | P a g e
determine the associations between sleep variables and habitual sugar intakes. This has
been achieved either by measuring sugar intake via carbohydrate intake, consumption of
sugar-sweetened beverages/foods, or total or added sugar intake (see Table 3.1 for a
summary). There is a lack of studies looking at only sugar and this might be due to the cost
and time required to measure a single nutrient.
Carbohydrates include simple sugars and complex starches and fibres. Overall, cross-
sectional studies have found mixed results regarding the relationship between carbohydrate
intake and sleep duration in children less than 18 years. Increased carbohydrate
consumption was found to be associated with less total sleep time (TST) in adolescents (Al-
Disi et al., 2010) and school-aged children (Firouzi, Poh, Ismail, & Sadeghilar, 2014). No
association has also been shown (Awad, Drescher, Malhotra, & Quan, 2013; He et al., 2015)
and one study showed that greater carbohydrate consumption was associated with greater
TST in adolescents (Weiss et al., 2010). While Firouzi et al. (2014) and Al-Disi et al. (2010)
utilised a validated sleep questionnaire, others used a more objective measure of sleep
(Awad et al., 2013; He et al., 2015; Weiss et al., 2010). Using actigraphy, He et al. (2015)
showed that while carbohydrate intake was not related to TST, habitual sleep variability was
related to increased carbohydrate intake in school children aged 6-12 years. Weiss et al.
(2010) found an association between short sleep duration and decrease in carbohydrate
intake in teenagers aged 16-19 years. Furthermore, another study used home-based
polysomnography monitoring and a dietary questionnaire and found that carbohydrate
intake did not influence TST or sleep architecture in children aged 10-17 years (Awad et al.,
2013).
The most common way to measure sugar intake in children has been to measure the
consumption of high sugar beverages and high sugar foods. Overall, increased soft drink
consumption is associated with reduced sleep duration in adolescents (Al-Haifi et al., 2016;
Park, Sherry, Foti, & Blanck, 2012) and children (Franckle et al., 2015; Hjorth et al., 2014;
Kjeldsen et al., 2014; Moreira et al., 2010), although for some studies this was only found in
boys (Al-Haifi et al., 2016). In contrast, one study found that adolescents with longer sleep
duration were consuming higher frequencies of sugar-sweetened beverages (Al-Hazzaa,
2013; Garaulet et al., 2011). However, these studies did not use validated measures of
sleep and/or diet. Finally, naps throughout the day in children aged 10-19 years occurred
64 | P a g e
more often in boys who had increased sugary beverage consumption, whereas in girls,
increased high sugar beverage consumption is associated with fewer daytime naps (Collison
et al., 2010). Overall, the research focussing on carbohydrates and foods and beverages
containing sugar suggests a negative interaction with sleep duration. However, there are
many other dietary factors that could influence sleep duration, such as caffeine, and the
sleep changes cannot be specifically attributed to sugar intake.
65 | P a g e
Table 3.1: A summary of the literature regarding cross-sectional studies investigating carbohydrate, sugar-sweetened beverages and sugar intake with sleep in children
Authors (date) Participants Study design Sleep measure Sugar measure Mean Values Findings
Origin ↓ decreased/lower
→ no association
Al-Disi et al. n= 126 Cross- Question asking 24-hour food recall and FFQ. Not reported CHO intake ↓ TST (p=0.04)
(2010) M ±SD: N/A sectional about TST Reported as %E CHO intake ↑ interrupted sleep
Saudi Arabia Range: 14-18y study (p=0.04)
Males: 0%
Awad et al. n= 319 Field At home PSG Youth/adolescent Not reported CHO intake → no association with
(2013) M ±SD: N/A and a sleep questionnaire TST (p=0.458)
USA Range: 10-17y habits Reported as g/day
Males: N/A questionnaire.
Carbohydrate Studies
Firouzi et al. n= 164 Case control Children sleep FFQ Not reported CHO intake ↓ TST (r=-0.228, p<0.05)
(2014) M ±SD: N/A study habits Reported as g/day
Malaysia Range: 6-12y questionnaire
Males: N/A
He et al. (2015) n= 305 Field Actigraphy Youth/adolescent CHO intake: 232.42 CHO intake → no association with
USA M ±SD: 16.7 ±2.3y questionnaire ±880.80g TST
Range: 6-12y Reported as g/day Weekday TST: 6.91 ±0.90h CHO intake ↑habitual sleep
(baseline) variability
Males: 53% Weekend TST: 7.22 ±1.37h
Weiss et al. n= 240 Field Actigraphy 24-hour food recall, Weekday TST: 7.55 ± 1.14h CHO intake ↑ TST
(2010) M ±SD: 17.7 ±0.4y multipass approach %E from CHO: 50.8 ±8.4%
USA Range: 16-19y Reported as %E
Males: 48%
66 | P a g e
→ no association
Al-Haifi et al. n= 906 Cross- Question Questions related to the TST: SSBs ↓ TST in boys (p=0.001)
(2016) M ±SD: N/A sectional regarding TST number of SSBs consumed Boys: 5.2 ± 1.6h SSBs → no association TST in girls
th th
Kuwait Range: 10 -12 study per week Girls: 5.1 ± 1.7h
grade
Males: 51%
Sugar-sweetened beverages (SSBs) and foods studies
Al-Hazzaa et al. n= 2,868 Cross- Question Questions related to the TST: 7.17 ±1.6h SSBs ↑ TST (p=0.041)
(2014) M ±SD: 16.6 ±1.1y sectional regarding TST number of SSBs consumed
Saudi Arabia Range: 15-19y study per week
Males: 48%
Collison et al. n= 9,433 Cross- Question FFQ Not reported SSBs ↑ napping in boys
(2010) M ±SD: N/A sectional regarding TST Reported as servings of SSBs SSBs ↓ napping in girls
Saudi Arabia Range: 10-19y study per week
Males: N/A SSBs → not associated with TST
Franckle et al. n= 1,870 Cross- Self-reported Questions asking about the SSBs: 1.45 ±1.37per day Regular soda/soft drink ↓ TST
(2015) M ±SD: 10.6 ±1.5y sectional bedtime and frequency of consumption Soda/soft drink: 0.59 ±0.84 (β=0.16, 95% CI: 0.08, 0.24)
th th
USA Range: 4 -7 study wake time of SSBs the day before; per day SSBs ↓ TST (β=0.22, 95% CI: 0.09,
grade none, once, twice, three or 0.35)
Males: 49% more times
Adjusted for covariates SSBs → no
longer associated with TST
Garaulet et al. n= 3,069 Cross- Question asking FFQ Not reported SSBs and sugary foods → not
(2011) M ±SD: N/A sectional about TST Reported as frequency of associated with TST
Europe Range: 12-18y study SSBs and foods ranging from
Males: N/A zero to one per day
67 | P a g e
→ no association
Hjorth et al. n= 441 Cluster- Actigraphy Web-based daily dietary TST: 556 ±26mins SSBs ↓ TST (p=0.002)
(2014) M ±SD: 9.9 ±0.6y randomised assessment – seven days
Median SSBs %E (IQR): 2.7
Sugar-sweetened beverages (SSBs) and foods studies
Range: 8-11y crossover

Denmark SSBs intake reported as %E (1.3-4.3)
Males: 53% design
Kjeldsen et al. n= 676 Cross- Actigraphy and Web-based daily dietary Median SSBs %E (IQR): 2.7 SSBs ↓ TST (p≤0.03)
(2014) M ±SD: N/A sectional Child sleep assessment – seven days (1.3–4.5) SSBs ↑ TST variability (p≤0.003)
Range: 8-11y study habits
Denmark SSBs intake reported as %E
Males: 52% questionnaire
Moreira et al. n= 1,976 Cross- Question asking FFQ validated for SSBs: 106 ±188g A dietary pattern that included fast
(2010) M ±SD: 7.5 ±1.2y sectional about TST Portuguese adults food, SSBs and pastries ↓ sleep
Range: 5-10y study duration
Portugal Fast food, SSBs and pastries
Males: 50%
reported in g/day
Park et al. n= 16,188 Cross- Question about Question asking how much Not reported Higher odds of daily consumption of
(2012) M ±SD: N/A sectional average TST on soda was consumed in the soda among students who slept <8h
th th
Range: 9 -12 study school night past week, 0-4 times per day a night compared with those who
USA
grade slept ≥8h a night (OR, 1.18; 95 % CI,
Males: 49% 1.08, 1.29)
Abbreviations: ↑, increase or higher; ↓, decrease or lower; →, no association; %, percentage; %E, percentage of total energy intake; CHO, carbohydrate; CI, confidence
interval; FFQ, food frequency questionnaire; g, grams; h, hours; IQR, interquartile range; M, mean; mins, minutes; n, number of participants; N/A, not available; OR, odds
ratio; PSG, polysomnography; SSB, sugar-sweetened beverages; SD, standard deviation; TST, total sleep time; USA, United States of America; y, years
68 | P a g e
In 2014, cross-sectional studies started specifically investigating sugar and its
relationship to sleep (see Table 3.2 for a summary). Firstly, Hjorth et al. (2014) aimed to
investigate associations between sleep duration and dietary risk factors for obesity over
a period of 200 days. In 441 children aged 8-11 years (mean: 9.9 ±0.6 years) daily intake
of food and beverages was recorded for seven consecutive days, using a validated
Dietary Assessment Software for Children along with actigraphy measurements for sleep
duration. The study showed reduced sleep duration was associated with higher intakes
of added sugar. Similarly, Kjeldsen et al. (2014) aimed to investigate if sleep duration
was associated with proposed dietary risk factors for 676 overweight and obese 8-11-
year-old children, a similar sample to Hjorth et al. (2014). For seven consecutive days,
children wore actigraphy to measure sleep, while parents and children completed a
sleep diary. For the same time period, daily intake of food and beverages was recorded
using a validated Dietary Assessment Software for Children. Similar to Hjorth et al.
(2014), the study found that short sleep duration was associated with higher intakes of
added sugar as a percentage of total energy.
Finally, Hunsberger et al. (2015) investigated whether the association between being
overweight and short sleep duration could be influenced by a high carbohydrate diet.
From across Europe (Belgium, Cyprus, Estonia, Germany, Hungary, Italy, Spain, and
Sweden) 5,944 children aged 6.1 ±1.8 years took part in the study. Data on diet and
sleep was collected every day of the week (including weekends). Parents completed a
24-hour recall for their child and were asked questions regarding the hours of nocturnal
sleep their child was having, with short sleep defined as less than 10 hours per night.
The study found that short sleep was associated with less sugar intake at midday but
more in the evening. Currently no study has investigated the association of sugar intake
with other sleep variables, such as sleep quality.
69 | P a g e
Table 3.2: A summary of the literature regarding cross-sectional studies investigating the association between sugar intake and sleep in children
Results
Name (date) Sleep ↓ decreased/lower
Sample Design Sugar measure Mean results
Origin measure ↑ increased/higher
→ no association
Hjorth et al. n= 441 Cluster- Actigraphy Web-based daily dietary TST (mins): 556 ± 26 Added sugar ↓ TST by 1h
(2014) M ±SD: 9.9 ±0.6y randomised assessment – seven days (p=0.001)
Added sugar (%E): 11.2 ±
Range: 8-11y crossover
Denmark Reported as %E 4.1
Males: 53% design
Kjeldsen et al. n= 676 Cross- Actigraphy Web-based daily dietary Sugar %E: 11.2 ±4.2 Added sugar ↓ subjective
(2014) M ±SD: N/A sectional and child assessment – seven days and objective TST (p≤0.03;
Range: 8-11y study sleep habits p=0.02)
Denmark Reported as %E
Males: 52% questionnaire
Hunsberger et n= 5,944 Prospective 24-hour recall 24-hour recall Not reported Short sleep was associated
al. (2015) M ±SD: 6.1 ±1.8y cohort study with less sugar intake at
Reported as g/day
Range: 2-9y midday but more sugar
Europe
Males: 51% intake in the evening
Abbreviations: ↑, increase or higher; ↓, decrease or lower; →, no association; %, percentage; %E, percentage of total energy intake; g, grams; h, hours; M, mean; mins,
minutes; n, number of participants; N/A, not available; SD, standard deviation; TST, total sleep time; y, years
70 | P a g e
3.7 Sugar and behaviour
The first major study to assess the relationship between diet and behaviour was
undertaken by Prinz et al. (1980). Using seven-day food diaries on 28 hyperactive
children and 26 controls they found a significant positive correlation between the
amount of sugar consumed and levels of “destructive-aggressive” and “restless”
behaviour in the hyperactive group. Among control children only “movement in the
room” correlated significantly with the dietary measures, with increased activity being
associated with increased sugar intake. However, it has been recognised that Prinz et al.
(1980) did not calculate the nutrient levels correctly. Experimental work since has
largely debunked the notion that sugar consumption impacts behaviour (Bellisle, 2004;
Milich, Wolraich, & Lindgren, 1986). Since Prinz et al. (1980), experimental work has
consistently shown that a large dose of sugar intake in children is not associated with
changed behaviours such as activity, impulsivity, and aggressive, or disruptive behaviour
in healthy children and children diagnosed with Attention Deficit Hyperactivity Disorder
(Bellisle, 2004; Milich, et al., 1986). As previously stated, experimental work allows
deductions of causation, and greater understanding of the mechanisms involved, but
often does not give an accurate depiction of an overall diet, and does not assist in
explaining ongoing behavioural issues.
Only one study has examined the relationship between high habitual sugar intake and
behaviour in school-aged children. Wolraich et al., (1986) compared diets of 32
hyperactive boys and 26 controls aged 7-12 years. A three-day food diary was kept by
the parents to calculate the sugar intake of the children. There were no significant
differences between diets; but, there was a significant difference between the
proportion of parents of hyperactive children and the proportion of parents of children
in the control group who stated they were restricting or had tried to restrict sugar.
However, in 1986, data on sugar content of food was not readily available, therefore
making it difficult to calculate a reliable total sugar value.
Overall, there is a paucity of information regarding habitual sugar intake and behaviour
in school-aged children. While the current information provided suggests no
relationship exists, the literature available is outdated. However, the relationship
71 | P a g e
between sugar intake and sleep suggests a potential pathway for behaviour to be
disrupted in high sugar diets and this possibility is explored in this thesis.
72 | P a g e
Chapter 4
Study Aims
73 | P a g e
4.1 Study one
Validation of an Australian caffeine food frequency questionnaire

Food frequency questionnaires (FFQs) are often used as a quick, low burden option to
gather large amounts of data on one nutrient. Researchers in Europe, United Kingdom,
and United States of America have validated caffeine-specific FFQs (Boylan et al., 2008;
Bühler et al., 2013; Ferraroni et al., 2004; Rudolph et al., 2012; Schliep et al., 2013), but
most of these questionnaires do not measure all caffeinated food and drinks, and those
that are more comprehensive do not specify between item sizes and caffeine content.
Due to the differences in day-to-day caffeine, such item specificity is likely to impact the
validity of FFQ measurement. Currently in Australia there is no caffeine-specific FFQ.
Due to the differences in caffeine amounts in products, and product availability,
individual country FFQs are necessary for valid research outcomes.
While experimental research in caffeine is important (to understand direct relationships

and mechanisms), it is also important to understand how identified effects translate in
day-to-day life. Research on the relationship between caffeine and sleep in adults has
been expansive, both experimentally and in the field (Clark & Landolt, 2016). However,
no study has used a validated caffeine-specific FFQ and a validated sleep measure to
determine the relationship of caffeine and sleep in the day-to-day life of a healthy
sample of adults.
Aim: To develop a caffeine-specific food frequency questionnaire to capture all food and
beverages containing caffeine in the Australian context, then use this measure to
determine the association between caffeine and sleep in adults.
Research questions:
Chapters 6 & 7
1. What is the validity and reproducibility of an Australian caffeine-specific food

frequency questionnaire?
2. How much caffeine is consumed day-to-day in a sample of Australian adults?
3. Is there a relationship between caffeine and sleep quality in adults?
74 | P a g e
4.2 Study two
Association between caffeine and sugar on sleep and behaviour in children

In Australia, there are no accepted daily intake guidelines for caffeine consumption for
either adults or children, but the effects of caffeine on the body are numerous and
include physiological, cognitive, and mood changes. Caffeine consumption has also
been recorded to reduce feelings of fatigue. Compared to adults, there is less research
concerning the relationship between caffeine and sleep in children. This is in spite of
recent studies showing that school-aged children are commonly consuming caffeine
(Australian Bureau of Statistics, 2014b; Commonwealth Scientific and Industrial
Research Organisation [Australia] et al., 2008). It is therefore timely and important to
understand the effects caffeine consumption may have on children’s day-to-day sleep
and subsequent behaviour. In the literature, majority of studies indicate that children
aged under 12 years who consume caffeine have decreased total sleep time, and
decreased sleep quality, however further research is needed on a broader range of sleep
quality variables with the use of validated questionnaires. Due to the disruption that
caffeine has on sleep in children, it is also important to consider the relationship
between caffeine and behaviour.
Since the peak of sugar consumption reported in 1995, the more recent surveys indicate
sugar intake in Australia has since been decreasing. However, children are still, on
average, consuming more than the recommended amount of sugar. There are two
major health concerns when considering the negative effect sugar has on the body:
obesity and dental caries. However, society frequently associates increased sugar intake
with poor sleep and behaviour. Few studies have examined the effect of sugar
consumed in habitual high doses and the relationship with behaviour in school-aged
children. While the current information provided suggests no relationship exists, the
literature available is outdated. However, the association between higher intakes of
added sugar and shorter sleep duration suggests a potential pathway for behaviour to
be disrupted in high sugar diets. Currently no study has investigated other sleep
variables, such as sleep quality and the association with sugar intake.
75 | P a g e
Aim: To gather data on habitual caffeine and sugar intake, sleep quality and daytime
behaviour in healthy, school-aged children.
Research questions:
Chapter 8
1. What is the amount of caffeine consumed by 8-12-year-old Australian children

and what are the sources?
2. What is the relationship between caffeine consumption and sleep variables in
children aged 8-12 years?
3. What is the relationship between caffeine consumption and behavioural
variables in school-aged children aged 8-12 years?
4. Does sleep act as a mediator between caffeine consumption and behaviour?
Chapter 9
5. How much sugar do school-aged children aged 8-12 years consume per day
compared to previous years?
6. Is sugar intake related to sleep problems in children aged 8-12 years?
7. Is sugar intake related to behavioural problems in children aged 8-12 years?
8. Does increased sugar intake exacerbate the problematic behaviour associated
with poor sleep?
76 | P a g e
Chapter 5
General Methodology
77 | P a g e
5.1 Overview
Data reported in this thesis were collected from two studies. Study one involved
developing and validating a caffeine questionnaire and investigating the relationship
between caffeine and sleep in adult populations. Study two involved assessment of diet,
sleep, and behaviour in children aged 8-12 years. This chapter provides an overview of
the methodologies used for each study. Both subjective and objective approaches to
measuring sleep contain inherent strengths and weaknesses. Objective measures of
sleep such as actigraphy and polysomnography tend to capture immediate sleep stage
differences and sleep efficiency during a specific night and have minimal self-report bias
(Gregory & Sadeh, 2012; Baker, Maloney & Driver, 1999). While subjective measures of
sleep measure the subjective sense of having a poor sleep and provide a more general
overview or ‘normal’ measure of sleep quality (Gregory & Sadeh, 2012).
5.2 Ethical Approval
This research (both study one and two) was approved by the University of South
Australia’s Human Research Ethics Committee (HREC number: 30885). This allowed for
recruitment within the community through fliers, event booths and social media. For
child recruitment, further ethical approval was sought from the Department for
Education and Child Development (DECD) and the Catholic Education Office before
recruiting children through schools in South Australia.
5.3 Study One
5.3.1 Informed Consent
Before taking part in the study, all participants were given an information sheet which
outlined the study, stated that the study was voluntary, and explained that participants
could withdraw at any time (Appendix B). After carefully reading the information sheet
each participant provided informed written consent (Appendix C). Furthermore, each
participant could request a personal summary of their results, and enter the draw to win
one of 15 coffee vouchers worth $20.
78 | P a g e
5.3.2 Participants
A pilot study was conducted from July to August 2013 to determine face validity of a
caffeine food frequency questionnaire (C-FFQ). Following this, a validation study of the
C-FFQ was conducted from March to August 2015. Both studies were cross-sectional in
design.
5.3.2.1 Pilot study

Twenty participants were recruited and were screened for eligibility (Table 5.1;
Appendix D).
5.3.2.2 Validation study

Ninety participants were initially recruited and screened to determine if they were
eligible (Table 5.1; Appendix D). One participant was removed as an outlier as their
caffeine intake was greater than three standard deviations above the mean. To test the
validity of the C-FFQ, 12 further participants were removed due to discrepancies with
their food diary (such as incorrect completion, or not completing for seven full days),
and for not completing the C-FFQ1 the immediate day after completing the food diary
(meaning the C-FFQ was no longer measuring the same period as the food diary). To
test reproducibility, 14 participants were removed as they failed to complete the C-FFQ1
and C-FFQ2 on two consecutive days; leaving a final sample of 75 participants.
Table 5.1: An outline of the exclusion criteria for study one
Exclusion Criteria
Did not consume a caffeinated beverage or food at least once a day
Under the age of 18 years
Had paid employment after 19:00

Pilot Study
Clinically diagnosed with any sleep disorders
Taking any prescribed medications or supplements to aid sleep or alertness
Suffering from any conditions that limits caffeine intake e.g. pregnancy
Did not consume a caffeinated beverage or food at least once a day
Under the age of 18 years

Validation Study
Could not read/write in English
Not able to attend the University of South Australia on two consecutive days
79 | P a g e
5.3.3 Participant Recruitment
Using fliers (Appendix E) participants were recruited using community notice boards
throughout the Adelaide city area, the University of South Australia web site and social
media. Further participants were recruited using word of mouth.
5.3.4 Measures
Caffeine food frequency questionnaire (C-FFQ)
The aim of Study One was to test the validity and reproducibility of the C-FFQ (Appendix
F). The C-FFQ was specifically designed to capture caffeine intake from beverages and
foods available in Australia; medications, supplements, and other sources of caffeine
that are not from a normal diet were not included. The C-FFQ is a self-report
questionnaire and has four food and beverage categories: energy drinks, soft drinks
(soda), coffee and tea (both hot and cold), and chocolate (food and beverage).
The C-FFQ requires participants to recall beverages and chocolate foods that they have
consumed during the previous week (seven days). The questionnaire requests the
participant to select the products they consumed, size most commonly consumed, and
the number of times over the last week it was consumed. Pictures of products with
caffeine were shown in the C-FFQ to stimulate participants’ memory, and to increase the
reliability. Total caffeine intake is then reported as an average weekly value in
milligrams, which is then divided by days to report average daily total caffeine intake in
milligrams. Average daily caffeine consumption is also calculated for each food and
beverage category.
The C-FFQ was designed by compiling a list of caffeinated beverages available to

purchase through Australian supermarkets. During piloting, any caffeinated products
missed were added to the questionnaire as well as a section for people to write any
additional drinks they consumed and believed to contain caffeine. If caffeine values
were not readily available on the product or website, then nutrient tables for use in
Australia (NUTTAB 2010 - Australian Food Composition Tables: Food Standards Australia
New Zealand, Canberra) were used and finally if the caffeine amounts were still not
readily available, the manufacturers were contacted directly (Table 5.2).
80 | P a g e
Based on the results from the pilot study (see section 0) a validation study of the
questionnaire was completed to measure the reproducibility and validity of the
questionnaire in a larger sample. The validation study used the same questionnaire
format except for the following changes based on the pilot results (Appendix G):
1. Removal of products that were no longer sold in Australia when the study was
conducted such as caffeinated Vitamin Waters and Powerade containing
caffeine.
2. After the energy drink and soft drink section an extra question was included
labelled ‘other’. This was for participants to include beverage and foods not
commonly available but to ensure all sources were captured.
3. Further investigation of caffeine content in beverage varieties allowed for
flavours of some drinks to be merged. For example, all Red Bull™ flavours have
the same amount of caffeine but all Coca Cola™ flavours do not.
4. Further instructions were included/amended:
a. ‘Place a cross through the pictures of drinks not consumed’ was added to
ensure that all pages were looked at.
b. Wording in the chocolate section was not clear and was restructured for
clarity so more detail could be provided.
81 | P a g e
Table 5.2: Different caffeine amounts for each category of the caffeine food frequency questionnaire.
Caffeine Information source

Energy drinks (mg/100ml)
All Energy drinks 30- 32 Product/ Company websites
Soft drink
Coca-Cola™ Varieties 10-32 Company websites
Pepsi™ Varieties 10-24 Product
Mountain Dew™ 15 Product
Sunkist™ 0 Manufacturer*
Coffee
Iced 31-155 NUTTAB - 2010
Flavoured milk 15 NUTTAB - 2010/Company websites
Hot Coffee varieties 31-155 NUTTAB - 2010
Tea
Iced 7.2 NUTTAB - 2010/Company websites
Hot tea varieties 0-19 NUTTAB - 2010/Company websites
Chocolate (mg/100g)
Flavoured Milk 1-37 NUTTAB - 2010
Milk 16-20 NUTTAB - 2010
White 5 NUTTAB - 2010
Dark 30-59 NUTTAB - 2010
Abbreviations: mg, milligrams; g, grams; mL, millilitres; NUTTAB, Nutrient
Tables for use in Australia; TM, trademark.
*Direct contact with manufacturer
Seven-day Food diary

Participants were required to keep a food diary for seven consecutive days (Appendix H).
A food diary requires all foods and/or beverages to be recorded in detail by the
participant for a period time (Thompson & Byers, 1994). All participants were instructed
to record their consumption of foods containing chocolate, all types of beverages
including water and alcohol, the time of consumption, type of food or beverage and
brand names, and the volume or quantity. If measuring the food was not possible,
participants were asked to record a description of what was consumed in detail so that a
best estimation of the quantity could be made when entering the data. Participants
were asked to record their food or beverage consumption at the time to minimise error.
Participants were provided with diaries and instructions to assist in recording.
Food diaries have a high precision in estimating participants’ nutrient intake and are an
accurate way to measure and represent food intake (Thompson & Byers, 1994; Willett,
2012). However, they are not error free, due to the burden on participants and need for
high motivation. Food diaries have been shown on average to underestimate overall
82 | P a g e
intake of nutrients due to various reasons including dieting and dietary restraint, gender,
socioeconomic status, motivation, social expectations and the nature of the testing
environment (Hill & Davies, 2001). Nevertheless, food diaries are often regarded as the
gold standard among different dietary assessment tools (Barrett-Connor, 1991;
Thompson & Byers, 1994).
Pittsburgh Sleep Quality Index

The Pittsburgh Sleep Quality Index (PSQI) is a subjective measure of sleep which uses
self-report to measure sleep quality and sleep disturbance over a one-month period
(Appendix H). The PSQI contains 19 self-report questions and five questions rated by the
bed partner or roommate. This study utilised only the self-rated questions as they
combined to form the seven component scores and the latter five questions are used for
clinical information only (Buysse et al., 1989). The seven component scores include:
subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep
disturbances, use of sleep medication, and daytime dysfunction. Each item on the PSQI
has a range of 0-3 points. In all cases a score of 0 indicates no difficulty, while a score of
3 indicates severe difficulty. A global PSQI score is calculated from adding together all
seven component scores to give an overall indication of sleep quality. The PSQI global
score ranges from 0-21 points, with 0 indicating no difficulty and 21 indicating severe
difficulties. The PSQI has been used in populations as young as seven years (Yuksel et
al., 2007) and as old as 95 years (Cole et al., 2006).
The PSQI has been validated in many different populations and has been translated into
many different languages due to its ability to identify different groups of patients
(Buysse et al., 1989; Buyesse et al., 2008). Originally, Buysse et al. (1989) reported on
the validity and reliability of the PSQI in good sleepers, poor sleepers and sleep disorder
patients. The validity of the PSQI is difficult to measure due to differences with more
objective measures such as polysomnography. However, the PSQI did show a sensitivity
value of 89.6% and specificity of 86.5%. This means that while the distribution of global
PSQI scores differed between groups, a post hoc cut-off score of 5 correctly identified
poor sleepers in 88.5% of all patients and controls, 84.4% of people with disorders of
initiating and maintaining sleep, 88% of people with disorders of excessive somnolence
and 97% of people with depression (Buysse et al., 1989). Buysse et al. (1989) also
83 | P a g e
showed that the PSQI has strong reliability with the seven component scores showing an
overall reliability coefficient (Cronbach’s alpha) of 0.83. Furthermore, to test the PSQI’s
test-retest reliability 91 participants completed the PSQI on two separate occasions.
There were no significant differences between time one and time two and Pearson
correlations demonstrated good stability. For all sleepers, global PSQI scores had a
correlation coefficient of 0.85, while component scores ranged from 0.65-0.84. Similar
results have been replicated in further psychometric assessment studies (Backhaus et
al., 2002; Carpenter & Andrykowski, 1998).
5.4 Study Two
5.4.1 Informed Consent

Before taking part in the study, all participants were given an information sheet which
outlined the study, stated that the study was voluntary, and explained that participants
could withdraw at any time (Appendix J). After carefully reading the information sheet
each parent provided informed written consent (Appendix K) for their child to take part
in the study. Furthermore, each participant could request a personal summary of their
results and enter the draw to win one of five iPod shuffles. If recruitment was done
within a school, the classes were offered a talk about the benefits of healthy eating and
good sleep by the candidate.
5.4.2 Participant Recruitment

5.4.2.1 Community approach
Community recruitment was completed by placing fliers throughout the Adelaide
metropolitan area (Appendix L). Places included universities, hospitals, community
noticeboards, medical centres, entertainment facilities and primary schools. Interested
participants either emailed or called for further information. From there participants
were sent an information sheet, consent form and all questionnaires to complete.
Participants returned all information via a supplied reply-paid envelope. Only forms
accompanied with completed consent forms were retained for analysis.
5.4.2.2 School approach

South Australian Catholic, public and independent schools were approached between
October 2013 and July 2015 from across Adelaide and a variety of socio-economic areas.
84 | P a g e
Schools were contacted and if the school was interested they distributed information
sheets, consent forms and parent questionnaires to all children aged 8-12 years. Of the
schools contacted in South Australia one independent school took part, 14 public
schools took part and two catholic schools.
5.4.3 Participants
The study was cross-sectional with each participant completing the questionnaires on
one occasion. To collect the most accurate data pertaining to a typical healthy
population, strict inclusion and exclusion criteria were enforced. Each child participant
was between 8-12 years. Children were not included if they had already been diagnosed
with, or currently experiencing diabetes, ADHD or another behavioural disorder, and/or
a sleep disorder. All participants (child and parents) needed to have sufficient English to
be able to understand the questions and answer appropriately, and are able to read and
write. The approached schools in South Australia could not be a special needs school;
that is a school which specifically caters for disabilities. A power analysis was conducted
using G*Power (v 3.1.9.2, Universität Kiel, Germany) to determine the number of
participants needed. Using previous research done by Mindell et al. (2009) investigating
caffeine consumption and sleep in children it was shown that a total sample of 420
children will be needed to detect an effect size f =0.17 with 80% power and alpha at
0.05. Furthermore, previous research done by Kjeldsen et al (2014) on added sugar
consumption and sleep in children showed a sample of 30 children is needed for an
effect size f=0.74 with 80% power and alpha at 0.05.
5.4.4 Measures
Each parent and child answered a questionnaire battery. Parents answered a
demographic questionnaire, a behaviour questionnaire, and a sleep questionnaire. Each
child answered a food frequency questionnaire (FFQ) and a C-FFQ. Either the parent or
child completed the Puberty score questionnaire with the decision left to each individual
family. Each questionnaire will be described in detail in section 5.4.4.1 for parents and
5.4.4.2 for children.
Before being administered to participants each questionnaire was piloted in a small

sample of convenience (aged 8-12 years; no diabetes or diagnosed behavioural or sleep
disorders). Piloting the questionnaire was completed to check length of the
85 | P a g e
questionnaire, ease of administration, and appropriateness of questions. After piloting
the following changes were made to the parent questionnaire: primary school was
added to highest level of education; tanner stage questionnaire was allowed to be
completed by either the adult or child; household income was removed; the first 10
questions of the CBCL were removed as they are more clinically relevant and a further 4
questions removed as they were deemed unnecessary and confronting for parents; and
two additional sleep questions were included (‘How many hours of sleep does your child
get on most nights?’ and ‘How long after going to bed does your child usually fall
asleep?’). A final overview of the questionnaires used, and the length of the
questionnaires, is shown in Table 5.3.
Table 5.3: An overview of the questionnaires used in the study, the length of the questionnaires, and
total time needed to complete them.
Parent: Child:
Task Duration Task Duration
Demographic Information 10mins Caffeine food frequency questionnaire 10mins
Child Behaviour Checklist 20mins Australian Child and Adolescent 20-60mins

Eating Survey
Paediatric sleep problem survey 15mins
instrument
OPTIONAL: Puberty Score 5mins OPTIONAL: Puberty Score 5mins

questionnaire questionnaire
Total time: 45-50mins Total time: 40-85mins
Abbreviations: mins, minutes
5.4.4.1 Parent Questionnaires

Demographic Information
A demographic questionnaire (Appendix M) gathered important information that was
not collected as part of the other questionnaires. Within the demographic
questionnaire there were general questions regarding the child in question; these
included age, gender, date of birth, birth order, how noisy their bedroom was, whether
the child shared a room, ethnicity, weight and height. Furthermore, there were
questions about the parents themselves which included their birthdate, highest level of
education, occupation and if either parent was participating in shift work. Some general
questions were asked about the home including postcode and whether English was the
86 | P a g e
primary language spoken. For further screening of the exclusion criteria it was asked if
the child had been diagnosed with any clinical behavioural disorders, sleep disorders or
diabetes.
Child Behaviour Checklist

To measure each child’s behaviour, the Child Behaviour Checklist (CBCL) for ages 6-18
years was used. The CBCL is a well validated measure of behavioural, emotional, and
social problems and has been thoroughly used in Australia (Achenbach & Rescorla,
2001). The CBCL consists of 113 questions and asks parents or guardians to consider
their child’s behaviour over the past six months and rates each item as 0 =not true, 1
=somewhat or sometimes true, or 2 =very often true or often true. During the piloting
of the questionnaire they were four items deemed inappropriate and not suitable for
this age group; they were items 59 (plays with own sex parts in public), 60 (plays with
own sex parts too much), 73 (sexual problems), and 96 (thinks about sex too much).
These items were automatically assigned a score of 0. Therefore, this study used 109
questions.
The CBCL has 8 syndrome scales: ‘Anxious/Depressed’ (13 items),

‘Withdrawn/Depressed’ (8 items), ‘Somatic Problems’ (11 items), ‘Social Problems’ (11
items), ‘Thought Problems’ (16 items), ‘Attention Problems’ (10 items), ‘Rule Breaking
Behaviour’ (17 items), ‘Aggressive Behaviour’ (18 items) and ‘Other’ (17 items).
Removed items 59 and 60 were associated with thought problems and items 73 and 96
were associated with rule breaking behaviour. Furthermore, some syndrome scales can
be broken into internalising problems and externalising problems. Internalising
problems consists of the scales anxious/depressed, withdrawn/depressed and somatic
problems; problems that are mainly within the self. Externalising consists of rule
breaking behaviour and aggressive behaviour; problems that mainly inflict conflicts with
other people and with their expectations for their child. Additionally, a total problem
score can be calculated by summing all 109 items answered. A total problems score can
range from 0 (no problems at all that is the parents scored every problem as 0) to 240
(many problems that is the parent scored every problem as 2). For all scales and
domains of the CBCL a high score indicates a clinical problem (Achenbach & Rescorla,
2001).
87 | P a g e
Normalised T-scores were assigned to raw scores on each scale. A T-score of 50 was
assigned to all raw scores that were at or below approximately the 50 th percentile of the
normative sample. T-scores from 71-100 indicated the raw scores were above the 98th
percentile in the normative sample. A T-score below 64 is considered normal, 65-69 in the
borderline clinical range and ≥70 in the clinical range (Achenbach & Rescorla, 2001).
The CBCL has been tested for validity and reliability. The content validity of the CBCL
items has been supported by four decades of research, consultation, feedback,
refinement and revision (Achenbach & Rescorla, 2001). Furthermore, all items
discriminate significantly (p<0.01) between demographically matched children
presenting with behavioural problems and ones without behavioural problems. The
criterion related validity of the CBCL scales was supported by multiple regressions, odds
ratios, and discriminant analyses, all of which showed significant discrimination between
children presenting with behavioural problems and ones without behavioural problems.
Test-retest item reliabilities were computed for CBCL scale scores obtained by 73
parents with an eight-day interval. Test-retest reliability was very high for all scales
ranging from r=0.80-0.94. Internal consistency for the CBCL scale scores was supported
by alpha coefficients of 0.72-0.97.
Paediatric Sleep Problem Survey Instrument

The Paediatric Sleep Problem Survey Instrument (PSPSI; Biggs et al., 2012) was used as a
measure of each child’s sleep quality. It is a validated tool for assessing sleep routine,
bedtime anxiety, morning tiredness, night arousals, sleep disordered breathing and
restless sleep in children. The questionnaire has been validated for use in Australian
school-aged children (Biggs et al., 2012). The questionnaire is based on a combination of
children’s sleep domains from two frequently used and validated questionnaires
(Children’s Sleep Habits Questionnaire [Owens et al., 2000] and Sleep Disorders Scale for
Children [Bruni et al., 1996]) and author devised questions. The PSPSI is a retrospective
parental report questionnaire. It requires parents to recall their child’s sleep behaviours
over the last typical school week. Items were rated on a four point Likert scale of
‘never’, ‘rarely’ (once per week), ‘sometimes’ (2-4 times per week), and ‘usually’ (5-7
times per week). The PSPSI measured six factors: sleep routine (6 items); bedtime
88 | P a g e
anxiety (4 items); morning tiredness (4 items); night arousals (4 items); sleep disordered
breathing (4 items); restless sleep (4 items). Within each scale a higher score indicates a
larger problem with a T-score of equal to or greater than 70 representing above the 95th
percentile.
The PSPSI has been measured for reliability using test-retest reliability and internal
consistency. A test-retest analysis was completed on 109 participants over a three time
periods (0-5.9 months, 6-11.9 months, 12-18.1 months). A spearman’s Rho correlational
analysis was conducted on each test-retest time interval for each group. No significant
difference was found between groups for sex, age or socio economic status. Scores
mostly ranged 0.41–0.90 (moderately strong) except for bedtime anxiety which when
retested at the 12-18.1 months’ time period was found to have a correlation of 0.36.
Internal consistency was moderate to strong with ranges of the Cronbach’s alpha from
0.6-0.8.
5.4.4.2 Child Questionnaires

Caffeine food frequency questionnaire (C -FFQ)
The C-FFQ is the same questionnaire as outlined above in section 5.3.4 (Appendix G).
Australian Child and Adolescent Eating Survey Food Frequency Questionnaire

(ACAES-FFQ)
The ACAES-FFQ is the first food frequency questionnaire (FFQ) developed specifically for
Australian school children which ranks the dietary intakes of Australian children and
adolescents for a range of nutrients. The ACAES-FFQ is a 120-item FFQ with 15
supplementary questions regarding age, use of vitamin supplements, food behaviours
and sedentary behaviours. It is designed as a self-administered tool to collect
information about the dietary intake of school children over the previous six months,
however can be completed by a parent or guardian. An individual response for each
food, or food type, is required, with frequency options ranging from, ‘never’ to ‘four or
more times per day,’ but varied depending on the food. The ACAES-FFQ is semi
quantitative with a standard portion size provided for each food item and determined
using ‘natural’ serving size i.e. one slice of bread. In the absence of a natural serving
size, portion sizes were derived from the 1995 National Nutrition Survey. Due to
seasonal availability of some fruits, a separate section was included for seasonal fruit. It
89 | P a g e
was initially designed for children aged 9-12 years however has been found to be
accurate in children as young as eight years old (Burrows et al., 2013).
The validity and reproducibility of the ACAES-FFQ has previously been reported (Watson,
Collins, Sibbritt, Dibley, & Garg, 2009). Validity was tested by comparing the ACAES-FFQ
to a food record. When the unadjusted data for the ACAES-FFQ and the average of the
food records was compared, Spearman's correlation coefficients ranged from 0.09-0.35.
After transformation, the Pearson correlation ranged from 0.13-0.37. The weighted
kappa values for the nutrient intake data from the ACAES-FFQ and the average of the
food records ranged from 0.09-0.36. Further testing of validity of the ACAES-FFQ was
done with quintiles. The proportion of individuals correctly classified into the same
quintile ranged from 19%-31%. The percentage classified within one quintile ranged
from 51%-65% and the percentage grossly misclassified was small 1-7%. Validity was
further analysed with the Bland Altman method which showed the mean difference
between the food records and the ACAES-FFQ was positive for most nutrients, indicating
that when compared to food records, the ACAES-FFQ provided higher estimates for the
intake of all nutrients, except polyunsaturated fat and thiamine.
Test-retest reproducibility was evaluated by calculating correlation coefficients.

Spearman correlations for the unadjusted data ranged from 0.34-0.53. After
transformation, the range of Pearson correlation coefficients was lower ranging from
0.34-0.51. Furthermore, test-retest results of the nutrient intake data from time one
and time two showed weighted kappa values ranging from 0.36-0.54, however most
nutrients showed 'moderate' agreement with values between 0.41-0.60. Furthermore,
when testing the reproducibility, the proportion of individuals correctly classified into
the same quintile ranged from 39%-23%. The percentage classified within one quintile
was between 79%-63% and a very small percentage of individuals were grossly
misclassified, ranging from 0%-5%. A Bland Altman analysis of the reproducibility data
showed a small, positive mean difference for all nutrients, indicating that the testing
during time one tended to give slightly higher estimates of nutrient intake than time
two.
90 | P a g e
Puberty Category Scores Questionnaire
Tanner staging (sexual maturation rating) is based upon a scale of secondary sexual
characteristics. It allows a way to identify the degree of pubertal maturation in an
adolescent by the appearance of pubic hair, the development of breasts, the occurrence
of menarche and the degree of testicular and penile development. Carskadon and
Acebo (1993) developed a puberty stage questionnaire to determine the level of puberty
the child has reached without a physical assessment making it less invasive for the child
(Appendix N). The puberty stage questionnaire is a five-item questionnaire using a four
point Likert scale for questions 1-5 for boys and 1-4 for girls (1- Not started, 2- barely
started, 3- definitely started, 4- seems complete). For girls, the fifth question is ‘Have
you begun to menstruate?’ with a yes, no, and I don’t know answer scale and then a
further question asking how old they were when they started to menstruate. Yes on the
menstruation item =4 points; no = 1 point.
The appropriate puberty category is determined for boys by summing the responses to
body hair growth, voice change, and facial hair growth to determine a puberty score.
For girls, puberty category scores used body hair growth, breast development, and
menarche (Table 5.4).
Table 5.4: An overview of the scoring of puberty status
Pubertal Category Boys Girls
Pre-pubertal 3 3
Early pubertal 4 & 5 (no 3 point responses) 3 & no menarche
Mid pubertal 6-8 (no 4 point responses) 4+ & no menarche
Late pubertal 9-11 ≤7 & menarche
Post pubertal 12 8+ & menarche
Carskadon and Acebo (1993) found that children as young as nine years old indicated a
high correspondence between self-report and Tanner staging determined by clinical
assessment. While the initial questionnaire was designed, and tested using children
aged 10 years and older it has been successfully used in children as young as seven years
91 | P a g e
(Luking, Luby, & Barch, 2014). The questionnaire is designed so that either the child’s
parent/legal guardian or the child themselves can complete the puberty score
questionnaire.
The puberty stage questionnaire was tested for reliability and validity. Reliability was
tested using internal consistency in a sample of 698 children of both genders aged 10-12
years. Using Cronbach’s coefficient results from the survey demonstrated reasonable
reliability for both the child (α= 0.67-0.70) and parent (α= 0.68-0.78) versions of the
puberty scales. To determine validity some children were pre-tested and had their
Tanner stage determined by a paediatrician. Correlations between the child Tanner
ratings and paediatrician ratings were strong with only seven of the 38 adolescents
rating themselves one stage or more different from the paediatrician. The validity of the
Tanner stage questionnaire is further supported by a number of consistent patterns
found within the results for example fifth grade children rated themselves and were
rated by their parents as less mature than sixth graders, and sixth grade girls were
consistently rated more mature than boys of the same age, all of which is consistent
with normal pubertal development (Carskadon & Acebo, 1993).
92 | P a g e
Chapter 6
Validation and Reproducibility of an Australian
Caffeine Food Frequency Questionnaire
This is an Accepted Manuscript of an article published by Taylor and Francis in the

International Journal of Food Sciences and Nutrition
Reference:
93 | P a g e
This chapter discusses the validation process of the caffeine food frequency
questionnaire (C-FFQ) used in the proceeding chapters of the thesis. The C-FFQ was
developed to measure caffeine intake, as no other questionnaire was validated in an
Australian sample. This chapter has been accepted for publication in the International
Journal of Food Sciences and Nutrition. Section 6.3 was accepted as supplementary
material and so has been included before the introduction, and provides an overview of
the piloting of the C-FFQ.
6.2 Abstract:
The aim of this study was to measure validity and reproducibility of a caffeine food
frequency questionnaire (C-FFQ) developed for the Australian population. The C-FFQ
was designed to assess average daily caffeine consumption using four categories of food
and beverages including; energy drinks; soft drinks/soda; coffee and tea; and chocolate
(food and drink). Participants completed a seven-day food diary immediately followed
by the C-FFQ on two consecutive days. The questionnaire was first piloted in 20 adults,
then a validity/reproducibility study was conducted (n=90 adults). The C-FFQ showed
moderate correlations (r=0.60), fair agreement (mean difference: 63mg), and reasonable
quintile rankings indicating fair to moderate agreement with the seven-day food diary.
To test reproducibility, the C-FFQ was compared to itself and showed strong correlations
(r=0.90), good quintile rankings, and strong kappa values (κ=0.65), indicating strong
reproducibility. The C-FFQ shows adequate validity and reproducibility and will aid
researchers in Australia to quantify caffeine consumption.
94 | P a g e
6.3 Pilot study
Before starting the validation/reproducibility process, a pilot study was conducted on

the questionnaire to determine the face validity of the questionnaire. Capturing total
dietary caffeine intake is difficult and has not been done before in Australia using a food
frequency questionnaire (FFQ), therefore, the pilot study ensured the questionnaire
measured caffeine intake of an individual and helped to inform development and
accuracy of the study. The pilot study of the questionnaire was completed from July to
August 2013. The protocol was the same as the protocol discussed in the main study.
Twenty participants were recruited via word of mouth and utilised the same measures
as the validation/reproducibility study.
The final sample of the pilot study consisted of 19 participants; 15 females, 4 males with
an average age of 35.0 ±15.5 years (one participant’s data was removed as an outlier).
Caffeine consumption was not significantly different between the food diary and C-FFQ1
(Food diary: M=230.2mg, SE=43.1; C-FFQ1: M=197.9mg, SE=32.2; t[18]= 1.356, p=0.19)
or between C-FFQ1 and C-FFQ2 (C-FFQ2: M= 212.6mg, SE=35.8; t[18]=1.438, p=0.17).
There was a strong correlation between the caffeine reported on the food diary and C-
FFQ1, r=0.84, p<0.01; and C-FFQ1 and C-FFQ2, r=0.96, p<0.01). The Bland Altman test
indicated moderately good agreement, however there was proportional bias (p=0.038),
with larger caffeine intakes showing greater variance. The mean difference between the
two methods of collecting caffeine intake was 32.2mg and ±2SDs was equal to -170.9
and 235.3mg.
95 | P a g e
6.4 Introduction
Caffeine is the most widely consumed stimulant and is commonly consumed in coffee,
tea or chocolate with some drinks and powders also containing caffeine as a flavour
component (Barone & Roberts, 1996). The Australian Bureau of Statistics (ABS) reported
that during 2011-2012 daily caffeine intakes of Australian adults ranged between 95-
193mg per day (Australian Bureau of Statistics, 2015a). The most common method of
caffeine consumption is coffee with 16.3 million coffees consumed per day in Australia
(2014b). The amount of caffeine in coffee and tea is variable (Stavric et al., 1988) and
depends on the differences in reference volumes (i.e. different sized cups), product
sources (i.e. plant variety) and methods of preparation (i.e. methods of brewing; Barone
& Roberts, 1996). There is currently no quick measure which calculates total caffeine
from all food and beverages in Australia. While other countries such as Germany, Italy,
Austria, England and the United States of America (USA) have previously validated
caffeine food frequency questionnaires, there is a variance in caffeine consumption and
amount of caffeine added to products between countries. Therefore, this study aimed
to validate a caffeine food frequency questionnaire for use in Australia.
Europe, United Kingdom (UK), USA, and Canada have recommended guidelines for
caffeine intake. The caffeine guidelines for all these countries are similar and state that
for healthy adults, daily habitual caffeine consumption should be, on average, no more
than 400mg per day (European Food Safety Authority, 2015; Food and Drug
Administration, 2013; Health Canada, 2012). Canada, Europe and the UK have provided
guidelines for children’s habitual caffeine intake, stating that up to 2.5-3mg per kilogram
of body weight per day is not unsafe (European Food Safety Authority, 2015; Health
Canada, 2012). There are no guidelines provided for children in the USA (Food and Drug
Administration, 2013). In Australia, there are currently no acceptable daily intake
guidelines for caffeine consumption for either adults or children.
In adults, it has been shown that consuming large amounts of caffeine (greater than
400mg in a day) negatively impacts sleep, mood, performance, memory, and health (de
Mejia & Ramirez-Mares, 2014; Smith, 2002). Adverse effects of caffeine on human
health were explored by Nawrot et al. (2003) in a systematic review published in 2003
which concluded that in the general population of healthy adults, moderate caffeine
96 | P a g e
intake (up to 400mg per day) is not associated with adverse effects, such as general
toxicity, cardiovascular effects, effects on bone status and calcium balance, changes in
adult behaviour, increased incidence of cancer or effects on male fertility.
Measuring dietary intake is often difficult and costly, particularly when measuring one
specific nutrient in large populations. The most commonly used measures of nutrient
consumption include food diaries, 24-hour recall, and food frequency questionnaires
(FFQ; Barrett-Connor, 1991; Thompson & Byers, 1994; Willett, 2012). When assessing
diets in large groups the FFQ methods can be a good alternative as they lower
participant burden, are less time consuming, less intrusive, and are relatively
inexpensive as this analysis can be automated and does not require trained interviewers
(Gibson, Ferrucci, Tangrea, & Schatzkin, 2010; Riboli et al., 2002).
Currently in Europe, UK, and USA there are validated caffeine-specific FFQs (Boylan et
al., 2008; Bühler, Lachenmeier, Schlegel, & Winkler, 2013; Ferraroni et al., 2004;
Rudolph et al., 2012; Schliep et al., 2013). Each of these questionnaires provides
important information on caffeine intake, but each contain important limitations. The
key limitations of current caffeine-specific FFQs include lack of reproducibility (Boylan et
al., 2008; Rudolph et al., 2012), considering only caffeinated beverages (Schliep et al.,
2013), or only coffee and tea (Ferraroni et al., 2004), for specific populations only
(Boylan et al., 2008), or not allowing for reporting of different volumes consumed
(Bühler et al., 2013). These should be addressed in further FFQ development, and no
caffeine food frequency questionnaires have been previously validated in an Australian
context.
Therefore, this study tested validity and reproducibility of a caffeine food frequency
questionnaire (C-FFQ) which included a broad range of caffeinated food and beverages
to capture caffeine intake in Australian adults over one week. The primary objectives of
this study were to assess 1) the validity of the C-FFQ compared to a seven-day food diary
and 2) the reproducibility of the C-FFQ.
6.5 Materials and Methods
The study was approved by the University of South Australia’s Human Research Ethics
Committee. Data was collected for this study from July 2013 to August 2015. The study
97 | P a g e
was conducted in two phases; an initial piloting phase to determine face validity of the
C-FFQ (see section 6.3) followed by a larger study to determine validity and
reproducibility of the revised version. Based on pilot results the following alterations
were made to the C-FFQ: foods and drinks sold in Australia were updated, instructions
clarified, and additional categories included capturing other items containing caffeine.
Overall, the results showed strong correlations, reasonable agreement, and good face
validity.
6.5.1 Participants
For the larger study, ninety participants were recruited via University of South Australia
web pages, social media, flyers, and word of mouth. Participants were not eligible for
the study if they were not daily caffeine consumers, under the age of 18 years, not
proficient in reading and writing in English, and not able to attend the University of
South Australia on two consecutive days. No educational background was collected but
it was assumed that all participants could follow the instructions provided and describe
the foods in detail.
6.5.2 Procedure
Participants completed a seven-day food diary where they recorded all drinks consumed
and all foods eaten containing chocolate. Since the C-FFQ is not a measure of habitual
caffeine intake but a measure of caffeine intake over the previous one week, after
completing the seven-day diary, each participant was required to return to the
University of South Australia to complete the C-FFQ on day eight (C-FFQ1) and nine (C-
FFQ2). This approach has been used previously when designing food frequency
questionnaires looking at a specific time period (Dunn et al., 2011). To minimize bias
participants were told they needed to return to the University campus to complete a
second questionnaire; however, they were not told it was a repeat of the same
questionnaire. Average daily caffeine consumption was calculated for the food diary
and both caffeine food frequency questionnaires (C-FFQ1 & C-FFQ2). Due to the
procedure needing to measure the same seven days, participant data was excluded if
they did not follow procedure.
98 | P a g e
6.5.3 Measures
Caffeine food frequency questionnaire
The C-FFQ was specifically designed to capture caffeine intake from beverages and foods
available in South Australia at the time of data collection; medications, supplements,
and other sources of caffeine that are not from a normal diet were not included. The C-
FFQ is a self-report questionnaire designed to assess the average daily caffeine
consumption and the range of caffeinated products consumed. There were four
categories of food and beverages in the C-FFQ including: energy drinks, soft drinks
(soda), coffee and tea (both hot and cold), and chocolate (foods and drinks).
Participants were required to recall drinks and chocolate foods they had consumed
during the previous week and select the products they consumed, size most commonly
consumed, and number of times it was consumed.
The C-FFQ was designed by compiling a list of caffeinated drinks available to purchase
through Australian supermarkets. If caffeine values were not available on the product,
websites of the product or Nutrient Tables for use in Australia (NUTTAB 2010 -
Australian Food Composition Tables: Food Standards Australia New Zealand, Canberra)
were utilized. Finally, if the caffeine amounts were still not readily available the
manufacturers were contacted. Pictures of products with caffeine (including specific
brands) were shown in the C-FFQ along with size options and brewing methods where
appropriate to stimulate participants’ memory, to increase the reproducibility. The
results collected from the C-FFQ were entered into an Excel file to convert into an
average daily caffeine amount in milligrams and the C-FFQ also provided the amount of
caffeine coming from each category (energy drinks, soft drink, coffee/tea and chocolate;
Table 6.1). Caffeine intake was expressed as average milligrams per week and divided
by number of days to obtain average daily consumption. An example of the information
collected in the questionnaire is shown in Table 6.2).
99 | P a g e
Table 6.1: Different caffeine amounts for each category of the caffeine food frequency questionnaire.
Caffeine Information source

Energy drinks (mg/100ml)
All Energy drinks 30- 32 Product/ Company websites
Soft drink
Coca-Cola™ Varieties 10-32 Company website
Original coke and flavored 10
varieties
Diet coke 13
Vitamin Water (energy) 32
Pepsi™ Varieties 10-24 Product
Original 11
Max 19
Diet 10
Kick 24
Mountain Dew™ 15 Product
Sunkist™ 0 Manufacturer*
Coffee
Iced 31-194 NUTTAB - 2010
Instant 31
Café/Espresso 194
Coffee flavoured milk 15 NUTTAB - 2010/Company websites
Hot Coffee varieties 31-194 NUTTAB - 2010
Instant 31
Espresso 194
Long Black 101
Tea
Iced 7 NUTTAB - 2010/Company websites
Hot tea varieties 0-19 NUTTAB - 2010/Company websites
Black 19
Green 16
Herbal 0
Chocolate (mg/100g)
Chocolate flavoured Milk 10-37 NUTTAB - 2010
Chocolate milk 10
Drinking chocolate/café 37
Milk chocolate 16-20 NUTTAB - 2010
White chocolate 5 NUTTAB - 2010
Dark chocolate 30-59 NUTTAB - 2010
Fondant filled 30
All other 59
Abbreviations: mg, milligrams; g, grams; mL, millilitres; NUTTAB, Nutrient Tables for use in Australia; TM, trademark.
*Direct contact with manufacturer
100 | P a g e
Table 6.2: An overview of the questionnaire, its categories and what it is measuring
Energy Drink Soft Drink Coffee Tea Chocolate
TM TM TM TM
Beverage/food types Red Bull , V , Monster Coca Cola , Pepsi , Iced, Espresso, Long Black Iced, Green and Black. Iced, Hot and Chocolate
TM TM TM TM
asked about , Rockstar , Mother , Mountain Dew , foods.
TM TM TM
Wicked , Pink , Red Sunkist , Vitamin
TM TM
Eye Water
Volumes ranging in this 250mL–550mL 200mL–600mL 250mL-600mL 250mL-500mL 250mL-500mL

category
Frequency of 1-4 times per week and 1-4 times per week and 1-4 times per week and 1-4 times per week and 1-4 times per week and
consumption options an other category an other category an other category an other category an other category
101 | P a g e
Food diary
A seven-day food diary was used to assess all participants’ caffeine consumption over a
week. Participants were instructed to record as accurately as possible all foods
containing chocolate, beverages of all kinds including water and alcohol, the time of
consumption, type of food or beverage with brand names (where possible), and the
volume or quantity. If measuring the food was not possible, participants were asked to
provide a description of what was consumed in as much detail as possible and to make
the best estimation of the quantity. Food diaries were analysed using the NUTTAB 2010
database in FoodWorks 7 (Xyris Software, Queensland, Australia). Weekly caffeine
intake was calculated and was converted into an average daily caffeine amount in
milligrams.
Statistical Analyses:
Validity and reproducibility data was analysed using SPSS V21 for Windows (Chicago, IL,
USA). The raw data for the food diaries and C-FFQs were not normally distributed but
positively skewed, as commonly found with dietary data. Statistical and graphical
assessment was used to determine the most appropriate transformation to achieve
normal distribution with logarithmic transformation producing normalized distributions.
Means, standard deviations, medians and interquartile ranges were determined for all
measures of caffeine. Associations between two measures (food diary vs C-FFQ1, and C-
FFQ1 vs C-FFQ2) were measured using Pearson correlations (using logarithmic adjusted
data) and Spearman correlations (for unadjusted data). Difference between means was
tested using the paired samples t-test on logarithmic adjusted data. In all analyses,
results were considered statistically significant if p<0.05 (two tailed).
Agreement and variability between the C-FFQ and the seven-day food diary was
determined by a Bland Altman plot. A Bland Altman plot graphically shows the
individual difference between the two measures against the mean of the measurements
(Bland & Altman, 2010). These plots indicate the direction of bias and whether it is
constant across levels of intake. The ‘limits of agreement’ (the mean difference
±1.96SDs of the difference between the two measurements) determine whether the
agreement between the two methods is acceptable (Bland & Altman, 2010).
102 | P a g e
To evaluate C-FFQ1’s ability to assign participants to the same categories of intake as the
food diary and against the repeated C-FFQ (C-FFQ2), participants were classified into five
categories (quintiles). Quintile analysis allowed the two measures to be ranked and
compared to determine if the participants are ranked similarly for both measures.
Finally, to determine the overall agreement between the C-FFQ1 and C-FFQ2 the kappa
statistic was used (Viera & Garrett, 2005).
6.6 Results
Validity:
Data from one participant was removed as an outlier as their caffeine intake was greater
than three standard deviations above the mean. Twelve further participants were
removed due to discrepancies with their food diary (such as incorrect completion, or not
completing for seven full days), and for not completing the C-FFQ1 the immediate day
after completing the food diary (meaning the C-FFQ was no longer measuring the same
period as the food diary). To test the validation of the C-FFQ, 77 participants were used
(25% male) with an age range of 19-94 years and an average age of 39.5 ±19.8 years (M
±SD). The mean difference (±SE) between the food diary and the C-FFQ1 was 63.4
(±12.6)mg per day. A Pearson correlation was completed on the adjusted values of the
food diary showing a significant and moderately strong correlation, r=0.60 p<0.001
(Table 6.3). Scatterplots in Figure 6.2 further indicate moderate agreement. A paired
samples t-test showed a significant difference between the food diary and the C-FFQ1
(Food diary: M=246.9mg, SE=14.7; C-FFQ1: M=183.6mg, SE=11.1; t[76]=-5.019, p<0.001).
The Bland Altman plot (Figure 6.1) indicated fair agreement, however there is
proportional bias (p=0.004). The Bland Altman plot shows that the differences between
the measurements increase as the average amount of caffeine consumed increases.
That is, as caffeine intake increased the variability between the food diary and the C-
FFQ1 was greater. The Bland Altman plot shows that on average the questionnaire
underestimates caffeine consumption by 63.4mg per day.
103 | P a g e
Table 6.3: Means (standard error) and medians (interquartile range) of total caffeine consumption calculated from the food diaries, C-FFQ1, and C-FFQ2. Correlations,
mean differences and statistical significance between the food diaries and C-FFQ1, and between the C-FFQ1 and C-FFQ2.
Food Diary C-FFQ 1 Correlations Mean

t-test
Mean (SE) Median (IQR) Mean (SE) Median (IQR) r
a
r
b Difference
Total Caffeine (mg) 246.9 (14.7) 232.7 (130.1) 183.6 (11.1) 155.9 (104.6) 0.60* 0.58* 63.4 p<0.001
C-FFQ 1 C-FFQ 2 Correlations Mean t-test
a b
Difference
Mean (SE) Median (IQR) Mean (SE) Median (IQR) r r
Total Caffeine (mg) 188.0 (11.3) 168.7 (131.7) 188.7 (10.8) 168.7 (116.5) 0.90* 0.89* 0.7 p=0.622
Coffee/Tea (mg) 171.4(11.4) 157.4 (116.6) 171.3 (11.0) 157.4 (118.1) 0.98** 0.91** 0.1 p=0.967
Soft drinks (mg) 9.3 (2.8) 0.0 (9.1) 9.8 (2.9) 0.0 (10.4) 0.99** 0.98** 0.5 p=0.335
Chocolate (mg) 3.4 (0.5) 1.8 (4.2) 3.7 (0.6) 1.9 (4.6) 0.94** 0.95** 0.3 p=0.535
Energy drinks (mg) 3.9 (2.2) 0.0 (0) 4.0 (2.3) 0.0 (0) 1.00** 1.00** 0.1 p=0.321
Abbreviations: C-FFQ1, caffeine food frequency questionnaire visit one; C-FFQ2, caffeine food frequency questionnaire visit two; IQR, interquartile range; SE, standard
error; mg, milligrams.
n=77 participants for validity, 75 participants for reproducibility.
a
r – Pearson correlation (log adjusted data)
b
r - Spearman correlations (un adjusted data)
t-test – paired samples t-test of adjusted data
*p<0.01, **p<0.001
104 | P a g e
Figure 6.1: Bland Altman plot of the difference between caffeine intake measured by the Food Dairy and
C-FFQ1 plotted against the mean caffeine intake for the two methods.
Abbreviations: C-FFQ1, caffeine food frequency questionnaire visit one; SD, standard deviation; mg,
milligrams
Notes: The Bland Altman plot indicates moderate agreement with a proportional bias. Circles indicate the
individual data points, the solid horizontal line in the centre indicates the mean difference between the
two methods, (63.4mg) and the dotted lines above and below indicate ±1.96SDs (-153.7mg and 280.4mg).
Reproducibility:
One participant’s data was removed as an outlier as their caffeine intake was greater
than three standard deviations above the mean and 14 participants were removed as
they failed to complete the C-FFQ1 and C-FFQ2 on two consecutive days. To determine
the stability of the C-FFQ, test re-test reliability score was calculated on 75 participants
(31% male) with an age range of 19-94 years with an average age of 39.9 ±20.4 years (M
±SD). Correlations between total caffeine intake for C-FFQ1 and C-FFQ2 showed strong
correlations (r=0.90, p<0.001) and on the four categories (coffee/tea, soft drink,
chocolate and energy drinks) again with strong correlations (r=0.94-1.00, p<0.001). A
paired samples t-test showed no difference between the C-FFQ1 and the C-FFQ2 (C-
FFQ1: M=183.6mg, SE=11.3; C-FFQ2: M=188.70mg, SE=10.8; t(74)=0.495, p=0.62).
Further parametric and non-parametric tests showed there were minimal differences
between the C-FFQ1 and C-FFQ2 (Table 6.3). Scatterplots further indicated strong
correlations between the two measures (Figure 6.2). Cohen’s κ showed substantial
agreement between the C-FFQ1 and C-FFQ2, κ= 0.65 (95%; CI: 0.52, 0.78), p<0.001
105 | P a g e
A
Figure 6.2: Scatterplots showing individual responses of caffeine intake from the food diary and C-FFQ.
Abbreviations: C-FFQ1 and C-FFQ2, caffeine food frequency questionnaire visit one and visit two
Notes: Panel A (n=77) shows the agreement of caffeine intake between the food diary and the C-FFQ1 for
each individual. A moderate agreement (r=0.60) exists for caffeine intake between a food diary and the C-
FFQ1. Panel B (n=75) shows the agreement of caffeine intake between C-FFQ1 and C-FFQ2 for each
individual. A strong agreement (r=0.90) exists for caffeine intake between C-FFQ1 and C-FFQ2. Circles
indicate individual data points, and the line represents the linear relationship between variables.
Quintile Analysis:
When comparing C-FFQ1 and C-FFQ2, further quintile analysis was used to explore
coffee and tea consumption and chocolate consumption as the most consumed caffeine
products. Quintile analysis of the food diary compared to the C-FFQ1 showed that 34%
of subjects were correctly assigned to the same quintiles, and 70% of subjects were
assigned to either the same or adjacent quintile (Table 6.4).
106 | P a g e
Table 6.4: An overview of the quintile analysis showing the proportion of the C-FFQ1 values in the same
quintile, adjacent quintile or incorrectly classified compared to the food diary values
Quintile Same quintile Adjacent quintile Incorrectly classified

1 (n=15) 8 5 2
2 (n=16) 3 5 8
3 (n=15) 5 6 4
4 (n=16) 3 9 4
5 (n=15) 7 3 5
Total no. (%) 26 (34) 28 (36) 23 (30)
Abbreviations: C-FFQ1, caffeine food frequency questionnaire visit one; %,
percentage.
Notes: The table shows the proportion of the C-FFQ1 values in the same quintile,
adjacent quintile or incorrectly classified compared to the food diary values, n=77.
Quintile analysis of the C-FFQ1 compared to the C-FFQ2 showed that for total caffeine
consumption 72% of the subjects were assigned to the same quintile and 95% of
subjects were assigned to either the same or adjacent quintile (Table 6.5). The quintile
analysis for coffee and tea consumption showed that 95% of subjects were assigned to
either the same or adjacent quintile and for chocolate consumption the analysis showed
that 96% of subjects were assigned to either the same or adjacent quintile (Table 6.5).
Table 6.5: An overview of the quintile analysis comparing C-FFQ1 and C-FFQ2 for total caffeine
consumption, caffeine intake from coffee and tea, and caffeine intake from chocolate consumption.
Adjacent Incorrectly
Same quintile Weighte
quintile classified 95% CI
dκ
n % n % n %
Total caffeine 54 (72) 17 (23) 4 (5) 0.65 0.52, 0.78
Coffee and tea 54 (72) 17 (23) 4 (5) 0.65 0.52, 0.78
Chocolate 58 (77) 14 (19) 3 (4) 0.72 0.60, 0.84
Abbreviations: C-FFQ1 and C-FFQ2, caffeine food frequency questionnaire visit one and visit two; CI,
confidence intervals; κ, Cohen’s κ value; %, percentage
Notes: The table shows the proportion of the C-FFQ1 values in the same quintile, adjacent quintile
or incorrectly classified compared to the C-FFQ2 values, n=75. The table also shows the κ
Coefficient and 95% confidence intervals.
6.7 Discussion
This study describes the validation of a FFQ designed to assess caffeine consumption in
Australia. The study had two aims, first to determine the validity of the C-FFQ against a
food diary, and second to determine test re-test reliability of the C-FFQ. The results
107 | P a g e
indicated that the C-FFQ is a reliable tool to measure caffeine intake, it is reproducible,
has moderate correlations with caffeine intake determined from a food diary analysis,
and the results from the quintile assignments show that it is a reasonable tool for
ranking individuals from lowest to highest intakes. The C-FFQ does not replace the use
of a food diary/24-hour recall, however, the C-FFQ is the first of its kind in Australia, and
could be reasonably used as a tool to measure caffeine intake where a food diary/24-
hour recall is not appropriate, such as research where timing and cost is a factor and in
large sample sizes.
The first aim, to test the validity of the C-FFQ, determined fair to moderate validity
which was equal or greater than a number of more general FFQs. The Victorian Cancer
Council FFQ which is widely used in Australia reported correlation coefficients ranging
from 0.14-0.60 (Hodge, Patterson, Brown, Ireland, & Giles, 2000). Collins et al. (2014)
designed another widely used general FFQ for Australian adults and validated it against a
food diary with reported average nutrient correlations of r=0.10-0.78. Other caffeine-
specific FFQs have shown validity coefficients ranging between r=0.29-0.82 (Boylan et
al., 2008; Ferraroni et al., 2004; Rudolph et al., 2012; Schliep et al., 2013). It has been
stated that correlations between FFQ and recall-derived mean nutrient intake is rarely
greater than 0.6 (Cade, Burley, Warm, Thompson, & Margetts, 2004; Kristal et al., 2005).
As such, the C-FFQ showed moderate correlational agreement (r=0.60) with food diaries
and appears to perform in line with most widely used FFQs; however, its limitations
need to be considered.
Assessment of agreement between the FFQ and food diary showed that the two
measures had fair moderate agreement but with proportional bias, i.e. when measuring
small caffeine intakes, the variance is much smaller than when measuring larger caffeine
intakes. The total amount of caffeine calculated by the FFQ was lower than from the
food diary by 63mg, which is equivalent to one instant coffee per day (NUTTAB 2010
database). The agreement appeared to decline more markedly as caffeine consumption
increased above 500mg per day, which was confirmed with the scatterplots. The
underestimation and bias is believed to be due to the inability for the questionnaire to
provide extra information about coffee and tea intake. The variability of caffeine
amounts in coffee and tea is well documented due to differences in preparation (Barone
108 | P a g e
& Roberts, 1996; Stavric et al., 1988) and therefore, in the future the questionnaire
needs to consider ways to ensure stability across all caffeine intakes such as
incorporating more detail about coffee and tea consumption.
Quintile assignment indicated that the FFQ was able to adequately identify individuals
according to their caffeine levels with 34% falling in the same quintile and 70% in the
same or adjacent quintile. Overall the quintile assignment in this study was better than,
or consistent with, other studies (Erkkola et al., 2001; Sullivan, Brown, Williams, &
Meyer, 2008). Erkkola et al. (2001) found when validating a general FFQ, the FFQ could
classify 69% of participants into same or adjacent quintiles according to their nutrient
intakes. Sullivan et al. (2008) found when validating a FFQ measuring long chain n-3
polyunsaturated fatty acids, the FFQ could place 49% of participants in the same quintile
and 87% in either same or adjacent quintile compared to a food diary. A caffeine FFQ
focusing specifically on coffee and tea designed in Italy conducted a quintile and tertile
analysis for coffee and tea respectively (Ferraroni et al., 2004); the quintile analysis for
coffee placed 39.2% and 91.1% of participants in the same quintile for caffeinated coffee
and decaffeinated coffee respectively when compared to a food diary. Furthermore, tea
placed 60.8% of participants in the same tertile when compared to a food diary. A
caffeine food frequency questionnaire designed in the USA (Schliep et al., 2013)
conducted a quartile analysis on total caffeine which placed 50% of the sample into the
same category as the FFQ and 90% into the same or adjacent quartiles. Considering
other studies using this approach the C-FFQ showed similar results to other more
general FFQs and superior classification of consumers compared to caffeine-specific
FFQs.
The second aim of the study was to test the reproducibility of the C-FFQ against itself
one day apart. The Pearson correlation value between C-FFQ1 and C-FFQ2 was r=0.90
indicating that the C-FFQ has strong test-retest reproducibility. Correlation coefficients
between two administrations of an FFQ of 0.5-0.7 are common, however FFQ
administrations one day apart produce higher correlation coefficients (r=0.6-0.7; Cade et
al., 2004). Other caffeine food frequency studies have shown high reproducibility,
ranging from r=0.61-0.86 (Ferraroni et al., 2004; Schliep et al., 2013), although these fall
short of that reported for the C-FFQ.
109 | P a g e
Further quintile analyses were conducted to determine if the ranking of the participants
was the same across both questionnaires. One caffeine FFQ conducted a quintile
analysis showing that 61.0-91.9% of participants were ranked in the same quintile when
testing reproducibility (Ferraroni et al., 2004), while a second stated that 100% of the
sample were placed in the same or adjacent quartiles (Schliep et al., 2013).
Reproducibility quintile analysis in this study showed that reproducing similar ranks of
caffeine consumption was high. Overall, in the current C-FFQ the test re-test reliability
was very good as confirmed by strong associations between retest scores, no difference
between the average of repeated scores, and a high percentage of cases being ranked in
same or adjacent quintiles.
The strengths of the C-FFQ include that it can differentiate between caffeine sources, it
is brand specific, and captures both beverage and food sources of caffeine. The study
has shown that the C-FFQ is a tool that can measure caffeine quickly and easily, yet
effectively. However, there are some limitations to this questionnaire. Firstly, the study
was undertaken on a healthy population and it is unknown how people from clinical or
specific populations (e.g. children or elderly) would perform. The questionnaire is self-
report, potentially leading to optimistic bias (Miles & Scaife, 2003), and social desirability
(Hebert et al., 1995) which might affect the amount of food or drink reported. However,
this is also a concern for participants completing 24-hour recall and food dairy
completions. Further problems that have been identified in similar types of studies is
the ability for participants to record accurate portion size (Livingstone et al., 1990).
However, it has previously been shown that while there were higher association among
studies who allowed the participant to describe their portion size, these results were not
statistically significant and therefore the time taken to allow the participant to estimate
the portion size of their cup/mug does not provide significant benefit (Cade et al., 2004).
The process of keeping a food diary might have improved the accuracy in completing the
C-FFQ but to allow the food dairy and the C-FFQ to measure the same seven-days, the C-
FFQ1 and C-FFQ2 had to be completed one day a part, immediately following the
completion of a seven-day food dairy. It is possible this could have resulted in higher
validity and reproducibility scores as the participant’s memory may have improved by
completing the diary first (Cade et al., 2004), however, previous studies have shown that
110 | P a g e
people tend to forget about dietary consumption within 24 hours (Smith, Jobe, &
Mingay, 1991). This design was necessary to ensure dietary assessment over similar
periods.
Previously other caffeine food frequency questionnaires have compared caffeine

questionnaire data to caffeine levels in saliva (Boylan et al., 2008; Rudolph et al., 2012).
Boylan et al. (2008) collected saliva to validate a three-day FFQ and the participants had
to collect saliva samples every 90 minutes for nine hours across the day by keeping the
Salivette® in their mouth of 10 minutes each time. Furthermore, the women were asked
to avoid caffeinated foods and drinks for one hour before collecting the first sample and
15 minutes before each sample thereafter. While on the second day fewer samples
were collected the same rules applied, these restrictions may have altered their habitual
caffeine consumption. Rudolph, Faerbinger, and Koenig (2014) followed a similar
method when validating a three-day FFQ but only collected data on one day, which does
not allow for variability in caffeine consumption which exists daily. Using saliva as a
caffeine comparison would not have been appropriate in this current study as the time
restraints on caffeine consumption in this ‘free living’ study and the number of saliva
samples needed to measure caffeine concentrations across a seven-day FFQ would not
be feasible. It also isn’t feasible to use other biomarkers such as blood and urine as
caffeine is no longer present after one to five days (Crews, Olivier IV, & Wilson, 2001;
Zylber‐Katz, Granit, & Levy, 1984).
Future studies might consider applying the questionnaire to different populations where
caffeine consumption is a concern and not highly reported, such as high caffeine
consumers including shift workers, or low caffeine consumers such as children.
Furthermore, the questionnaire could be administered in populations where coffee
consumption has a large cultural importance or in lower socio-economic status
populations where there is a larger intake of high sugar drinks such as soft drink and
energy drinks (Lien, Jacobs, & Klepp, 2002; Scully, Dixon, White, & Beckmann, 2007).
Also, recent research has started to focus on the potentially beneficial effects of coffee
or tea intake on several outcomes, including Type-II diabetes (Ding, Bhupathiraju, Chen,
van Dam, & Hu, 2014), metabolic syndrome (Marventano et al., 2016), several cancers
(Wang et al., 2016) and mortality (Grosso et al., 2016). Thus, measuring caffeine to test
111 | P a g e
whether the beneficial effects of coffee are mediated by caffeine or other compounds
would be of interest for future research. Furthermore, due to the variability in caffeine
amounts within coffee and tea varieties, the questionnaire could also benefit from more
clarification in questions regarding types and sizes of coffee and tea.
6.8 Conclusion:
In conclusion, due to the layout and questionnaire set up, the C-FFQ is unique, and in
Australia there are no other caffeine-specific FFQs. The C-FFQ quantifies caffeine intake
in a one-time measure questionnaire capturing not only caffeine intake but also where
caffeine is coming from. The present study has shown that the new C-FFQ was able to
estimate caffeine intakes. The variability of caffeine intakes observed in this study was
not a major surprise, considering the range in caffeine in similar products. The C-FFQ
was able to effectively rank the majority of individuals according to total caffeine
intakes. The C-FFQ is therefore an adequate dietary assessment tool for the estimation
of total caffeine intakes in healthy Australian adults. It is an appealing alternative to
other dietary assessment methods due to its reproducibility, low subject burden and low
cost.
112 | P a g e
Chapter 7
Caffeine Consumption and Sleep Quality in
Australian Adults
Published in Nutrients
Reference:
Watson, E. J., Coates, A. M., Kohler, M., & Banks, S. (2016). Caffeine Consumption and
Sleep Quality in Australian Adults. Nutrients, 8(8), 479. doi:10.3390/nu8080479
113 | P a g e
The current chapter identifies common sources of caffeine in adults and the association
between caffeine consumption and sleep quality using the caffeine food frequency
questionnaire and the Pittsburgh Sleep Quality Index. This is the first study to use
validated measures to capture both sleep quality and caffeine intake in a cross-sectional
study and was published in Nutrients during August 2016.
7.2 Abstract:
Caffeine is commonly consumed to help offset fatigue, however, it can have several
negative effects on sleep quality and quantity. The aim of this study was to determine
the relationship between caffeine consumption and sleep quality in adults using a newly
validated caffeine food frequency questionnaire (C-FFQ). In this cross-sectional study,
80 adults (M ±SD: 38.9 ±19.3 years) attended the University of South Australia to
complete a C-FFQ and the Pittsburgh Sleep Quality Index (PSQI). Caffeine consumption
remained stable across age groups while the source of caffeine varied. Higher total
caffeine consumption was associated with decreased time in bed, as an estimate of
sleep time (r=−0.229, p=0.041), but other PSQI variables were not. Participants who
reported poor sleep (PSQI global score ≥5) consumed 192.1 ±122.5mg (M ±SD) of
caffeine which was significantly more than those who reported good sleep quality (PSQI
global score <5; 125.2 ±62.6mg; p=0.008). The C-FFQ was found to be a quick but
detailed way to collect population based caffeine consumption data. The data suggests
that shorter sleep is associated with greater caffeine consumption, and that
consumption is greater in adults with reduced sleep quality.
114 | P a g e
7.3 Introduction
Caffeine is a widely-consumed stimulant that is found in a variety of commonly

consumed foods and beverages such as chocolate, soft drink (soda), tea, and coffee.
Caffeine is commonly used as a fatigue countermeasure (Dorrian et al., 2011). Due to its
action on adenosine receptors (Porkka-Heiskanen et al., 1997; Strecker et al., 2000)
caffeine improves alertness (Brice & Smith, 2001). The Australian Bureau of Statistics
(ABS) showed that on average in 2011/2012 Australian adults aged 19–70 years had
daily caffeine intakes ranging between 103–183mg per day (Australian Bureau of
Statistics, 2014b), with coffee being the most common source of caffeine (Australian
Bureau of Statistics, 2014b). Currently there are no consumption guidelines for caffeine
in Australia. Therefore, it is important to examine the impact caffeine has on our sleep
to make more informed recommendations on consumption and to better understand
the impact in roles where caffeine consumption is higher.
While sleep need is individual and can differ from person to person, it is recommended
that adults obtain seven to nine hours per night (Hirshkowitz et al., 2015). Sleep has
been shown to be important for many different cognitive and health reasons
(Cappuccio, D'Elia, Strazzullo, & Miller, 2010; Cappuccio et al., 2008; Jennings, Muldoon,
& Hall, 2007; Xie et al., 2013). These benefits of sleep are not only dependent on total
sleep time (TST) but also sleep quality, measured by variables such as sleep efficiency,
and sleep onset latency (SOL). In adults, there are a number of experimental studies
investigating caffeine intake and its influence on sleep, with previous experimental
studies finding that caffeine consumption can impact sleep quality (Drapeau et al., 2006;
Shilo et al., 2002). Experimental laboratory studies have shown that when caffeine is
ingested one to three hours before bedtime it decreases sleep efficiency (Carrier et al.,
2007; Drapeau et al., 2006a; Shilo et al., 2002), decreases TST (Drapeau et al., 2006;
Hindmarch et al., 2000; Shilo et al., 2002), and increases SOL (Carrier et al., 2007;
Drapeau et al., 2006; Shilo et al., 2002). It can also impact sleep architecture by reducing
the amount of slow wave sleep (Carrier et al., 2007; Drapeau et al., 2006). However, the
experimental nature of these studies does not consider the impact of an individual’s
habitual caffeine intake or sleep patterns (Grant, Tang, & Kalow, 1983).
115 | P a g e
The effect of caffeine on sleep in university and general populations has been examined
by several cross-sectional surveys and/or field studies (Brick et al., 2010; Kant &
Graubard, 2014; Lund et al., 2010; Regestein et al., 2010; Reid & Baker, 2008; Sanchez-
Ortuno et al., 2005; Whittier et al., 2014; Zunhammer, Eichhammer, & Busch, 2014). In
these studies, increased caffeine consumption has been associated with decreased TST
(Kant & Graubard, 2014; Regestein et al., 2010), increased naps (Regestein et al., 2010),
decreased time in bed (TIB; Sanchez-Ortuno et al., 2005), increased sleep efficiency due
to decreased TIB (Sanchez-Ortuno et al., 2005), daytime sleepiness (Reid & Baker, 2008;
Whittier et al., 2014), and poor subjective sleep quality (Brick et al., 2010; Reid & Baker,
2008). However, in other studies increased caffeine consumption does not impact TST
(Sanchez-Ortuno et al., 2005), daytime sleepiness (Sanchez-Ortuno et al., 2005) or
Pittsburgh Sleep Quality Index (PSQI) global score (Brick et al., 2010; Lund et al., 2010;
Zunhammer et al., 2014). Of these cross-sectional studies, only one looked at total
caffeine consumption, i.e., all caffeinated beverages and caffeinated food consumed
(Lund et al., 2010), while the others examined only caffeinated beverages (Brick et al.,
2010; Regestein et al., 2010) or a selection of caffeinated beverages (Kant & Graubard,
2014; Reid & Baker, 2008; Sanchez-Ortuno et al., 2005; Whittier et al., 2014;
Zunhammer et al., 2014). Additionally, it was not always clear what question was asked
to calculate caffeine consumption, or how the participants recorded their intake. Also,
many studies did not consider caffeine from chocolate or examine the separate sources
of caffeine individually. This is important because different age groups may prefer
particular sources of caffeine. Furthermore, all the previous questionnaire studies rely
on an individual knowing what drinks/foods contain caffeine.
The mixed sleep patterns in caffeine users could be due to reporting inaccuracies. Few
studies gather detailed information about caffeine consumption or sources of caffeine.
To address this we have developed a caffeine food frequency questionnaire (C-FFQ;
Watson, Kohler, Banks, & Coates, 2016) that is short, does not rely on prior caffeine
content knowledge, and gathers information about a wide variety of caffeine sources
(coffee, tea, soft drink [soda] and chocolate beverages and foods). The overall aim of
this study was to determine the relationship between caffeine consumption and sleep
using this newly validated Caffeine Food Frequency Questionnaire (C-FFQ; Watson et al.,
116 | P a g e
2016) and self-reported sleep quality in adults. Specifically, this study aimed to: (1)
identify what types of foods/beverages contribute to caffeine intake; (2) determine the
impact of caffeine on different sleep quality variables (TIB, SOL and sleep efficiency); and
(3) determine the difference in caffeine intake between self-reported good and poor
sleepers.
7.4 Materials and Methods
This cross-sectional study was designed to investigate the relationship between habitual
caffeine consumption and sleep. It was conducted between March and August 2015 at
the University of South Australia, Adelaide. The project was approved by the University
of South Australia’s Human Research Ethics Committee (HREC number: 30885).
7.4.1 Participants
Adults were recruited via University of South Australia web pages, social media, flyers,
and word of mouth. Participants were ineligible for the study if they did not consume
caffeine daily, were under the age of 18 years, not proficient in reading and writing in
English, experiencing any sleep related conditions, have any conditions that affect
caffeine intake, and not able to attend the University of South Australia. Participants
were also excluded if they were taking any medications (prescription or over the
counter, e.g., sleeping tablet, herbal supplement) to assist sleeping or alertness.
7.4.2 Procedure
All participants gave written informed consent before attending the University of South
Australia. While at the University of South Australia participants were asked to
complete questionnaires recalling caffeine consumption over the past week and sleep
patterns over the past month.
7.4.3 Measures
Caffeine food frequency questionnaire: The C-FFQ is a self-report, validated, and
reliable questionnaire designed to assess the average daily caffeine consumption and
the range of caffeinated products consumed (Watson et al., 2016). The C-FFQ asks
about beverages (energy drinks, soft drinks [soda], both hot and cold coffee and tea, and
chocolate flavoured milk) and foods (chocolate) that were consumed in the previous
week. The questionnaire requires participants to select the beverage or foods they
117 | P a g e
consumed based on images of currently available products in Australia, indicating the
specific brand, size, and the number of times over the last week each was consumed.
Average daily total caffeine consumption and average daily caffeine amounts from
beverage and food sources were expressed as milligrams per day.
Pittsburgh Sleep Quality Index: The Pittsburgh Sleep Quality Index (PSQI) is a frequently
used sleep quality questionnaire that has been shown to be reliable and valid in many
populations (Backhaus et al., 2002; Buysse et al., 1989; Cole et al., 2006). The PSQI is a
subjective measure of sleep which uses self-report to measure sleep quality and sleep
disturbance over a one-month period. The PSQI contains 19 self-report questions and
five questions rated by the bed partner or roommate. This study utilised only the self-
rated questions which combine to form seven component scores (Buysse et al., 1989).
The seven component scores include subjective sleep quality, sleep latency, sleep
duration, habitual sleep efficiency, sleep disturbances, use of sleep medication, and
daytime dysfunction. Each component has a range of 0–3 points. In all cases a score of
0 indicates the best outcome, while a score of 3 indicates the worst outcome. A global
PSQI score is calculated from adding together all seven component scores to give an
overall indication of sleep quality. The PSQI global score ranges from 0–21 points, with 0
indicating no difficulty and 21 severe difficulties. A global score of greater than or equal
to five indicates poor sleep quality (Buysse et al., 1989). This cut-off has been shown to
have high specificity and sensitivity for distinguishing insomnia patients and controls
(Backhaus et al., 2002; Buysse et al., 1989).
7.4.4 Statistical Analysis

The C-FFQ calculates a total caffeine value and caffeine values for beverage and food
subcategories as milligrams per day. All data were checked for normality and caffeine
variables were found to be positively skewed. Log transformation of total caffeine
intake allowed correction to a normal distribution; however, due to the skew of caffeine
from individual beverage and food categories they could not be transformed. Data
extracted from the PSQI included subscales of sleep information and from these a global
score was calculated. Good sleepers were determined by a global score of less than five
and poor sleepers were determined by a global score of greater or equal to five as
118 | P a g e
described by Buysse et al. (1989). Sleep variables were non-normal and could not be
transformed to normal.
Mean, standard deviation, median, interquartile range, and range was calculated for
caffeine values to describe the population. Participants were broken into age groups
similar to the Australian Bureau of Statistics categorization (Australian Bureau of
Statistics, 2014b) to allow for comparison. All outliers for the sleep and caffeine
variables (greater than three SD above the mean) were clarified with the participant
completing the questionnaire, and all variables were subsequently considered to be
accurate and viable. To determine differences between groups both parametric and
non-parametric tests were used. For parametric data, independent samples t-tests were
used, for the data which could not be transformed and were non-normal Mann Whitney
U-tests were used, and for categorical comparisons chi square tests were used.
For further analysis, non-parametric (Spearman) correlations were utilised to assess

relationships with caffeine and sleep variables. Finally, to determine if there were
differences between PSQI categorical sleep variables and total caffeine consumption a
one-way ANOVA were undertaken. The ANOVA dependent variables included subjective
sleep quality, sleep disturbances, use of sleep medication, and daytime dysfunction. In
all tests, significance was determined if p<0.05.
7.5 Results
Study Participants
Of the 104 people who were provided with study information, 90 participants consented
to the study and 84 participants returned the questionnaires. The final data set for
analyses included 80 participants with four questionnaires having missing data (Figure
7.1). The final sample consisted of 54 females and 26 males, aged 19–94 years (mean
±SD 38.9 ±19.3 years), with a mean caffeine intake of 165.1mg/day. Most of the
participants (85%) reported their sleep quality to be either very or fairly good and 86%
of the participants had not taken medications to help them sleep over the past month
(prescribed or over the counter). Of the sample, 80% stated that they had disturbed
sleep less than once a week. Finally, 83% of the sample had minimal problems during
the day due to their sleepiness (further measures of caffeine and sleep are reported in
119 | P a g e
Table 7.1). There was no difference in age between the good sleepers (PSQI global score
<5) and poor sleepers (PSQI global score ≥5) and age had no relationship with sleep
efficiency (p=0.0574), SOL (p=0.756), however as age increased so did TIB (p=0.030).
Figure 7.1: Description of participant flow throughout the study
120 | P a g e
Table 7.1: Mean (SD), median (IQR) and range of responses of caffeine consumption and sleep measures
for participant
Mean (SD) Median (IQR) Range of responses
Total caffeine (mg) 165.1 (105.3) 133.4 (120.1) 41.6–726.6
Coffee (mg) 109.9 (99.4) 87.2 (82.3) 0.0–646.6
Tea (mg) 38.0 (56.2) 13.6 (50.9) 0.0–271.4
Chocolate (mg) 3.5 (4.6) 1.7 (4.3) 0.0–20.6
Soft Drink (mg) 9.0 (24.0) 0.0 (8.6) 0.0–182.4
Energy drink (mg) 4.9 (20.4) 0.0 (0.0) 0.0–135.0
PSQI global score 5.3 (2.5) 5.0 (4.0) 0.0–21.0
SOL (mins) 18.8 (11.8) 20.0 (20.0) 1.0–60.0
SE (%) 88.3 (9.3) 89.1 (13.7) 55.6–100.0
TIB (h) 8.0 (1.0) 8.0 (1.0) 6.0–13.3
Abbreviations: SD, standard deviation; PSQI, Pittsburgh Sleep Quality Index; IQR, interquartile range; mg,
milligrams; mins, minutes; SOL, sleep onset latency; SE, sleep efficiency; TIB, time in bed; global score, PSQI
total score; %, percentage.
Notes: Chocolate includes both food and drink.
Types of Foods/Beverages that Contribute to Caffeine Intake by Gender

There were no differences in total caffeine consumption between men and women
t(78)=0.60, p=0.548. Further, Mann Whitney U-tests were undertaken to determine any
differences in caffeine consumption from different caffeine sources between genders
and showed no differences in caffeine consumption from coffee, tea, soft drink and
chocolate between genders. However, males consumed more caffeine from energy
drinks compared with females (U=589.50, z=−2.36, p=0.018; Male median [IQR]: 0.0
[0.0]mg, range: 0.0–135.0mg; Female median [IQR]: 0.0 [0.0]mg, range: 0.0–22.4mg).
Due to the small intake of energy drinks consumed overall in this sample the differences
between gender in overall caffeine consumption was not considered to impact the
results and males and females were analysed together.
Types of Foods/Beverages that Contribute to Caffeine Intake by Age

There was no significant correlation between total caffeine consumption and age
(r=0.167, p=0.145), with similar mean intakes across age groups and a wide variation
within age groups (18–30 years: 174.6 [±139.4]mg; 31–50 years: 149.1 [±74.6]mg; 51–92
years: 184.4 [±72.4]mg). Energy drink consumption significantly decreased with age
121 | P a g e
r=−0.297, p=0.008). Tea consumption increased with age (r=0.217, p=0.056) and soft
drink consumption decreased with age (r=−0.221, p=0.052), however these effects were
only trends. Coffee intake and chocolate intake was not correlated with age (Figure
7.2).
Figure 7.2: Percent of caffeine intake from different sources by age groups.
Notes: Total n = 80; 18–30 years: n = 32, 31–50 years: n = 31, 51–92 years: n = 16.
Chocolate includes beverages and food.
122 | P a g e
Impact of Caffeine on Sleep Quality
Sleep onset latency and sleep efficiency were not significantly correlated with caffeine.
Time in bed was significantly related to total caffeine intake (r=−0.229, p=0.041) with TIB
increasing with decreasing caffeine consumption (Figure 7.3). Furthermore, caffeine
consumption was not significantly associated with other sleep factors (subjective sleep
quality, medications needed to sleep, sleep disturbance, and daily dysfunction due to
sleepiness).
Figure 7.3: Scatterplots showing the relationship between total caffeine consumed and sleep variables:
sleep efficiency, time in bed and sleep onset latency.
Notes: [A] represents total caffeine intake versus sleep onset latency, r=0.028; [B] shows total caffeine
intake versus sleep efficiency, r=−0.113; [C] shows total caffeine intake versus time in bed, r=−0.229.
123 | P a g e
Caffeine Intake between Self-Reported Good and Poor Sleepers
To determine if caffeine consumption differed between good and poor sleepers,
caffeine variables were compared between participants who reported a PSQI global
score ≥5 (poor sleep quality) and PSQI <5 (good sleep quality). On average, poor
sleepers reported greater total caffeine consumption (mean ±SD: 192.1 ±122.5mg)
compared to good sleepers (mean ±SD: 130.0 ±62.6mg; mean difference = 62.2mg,
t(78)=−2.73, p=0.008, d=0.64).
Furthermore, caffeine from coffee intake was significantly less in good sleepers
(median=66.5, IQR=83.1) than poor sleepers (median=99.2, IQR=83.3), U=1028.50,
z=2.34, p=0.019, r=0.26. While caffeine from tea (p=0.874), chocolate (p=0.658), soft
drink (p=0.368) and energy drinks (p=0.395) showed no differences found between
groups (Table 7.2).
Table 7.2: A descriptive table showing the means, standard deviations, medians, and interquartile
ranges for all variables for good and poor sleepers and any significant differences.
Good Sleep Quality Poor Sleep Quality

n = 35 n = 45
Mean (SD) Median (IQR) Mean (SD) Median (IQR) p Values
192.1 I
Total caffeine (mg) 130.0 (62.6) 123.2 (58.2) 140.4 (160.6) 0.008
(122.5)
132.2 M
Coffee (mg) 81.7 (68.5) 66.5 (83.1) 99.2 (83.3) 0.019
(115.0)
M
Tea (mg) 35.4 (48.4) 13.6 (54.3) 39.5 (61.5) 13.6 (55.0) 0.874
M
Chocolate (mg) 3.6 (4.0) 1.7 (4.5) 3.5 (4.9) 1.7 (4.5) 0.658
M
Soft Drink (mg) 6.5 (13.2) 0.0 (5.2) 10.9 (29.2) 0.0 (10.4) 0.368
M
Energy drink (mg) 2.9 (15.4) 0.0 (0.0) 6.0 (23.0) 0.0 (0.0) 0.395
C
SOL (mins) 11.4 (7.6) 10.0 (13.0) 24.3 (11.0) 30.0 (15.0) 0.000
M
SE (%) 95.0 (4.6) 94.1 (8.6) 82.9 (8.8) 82.4 (11.1) 0.239
C
TIB (h) 7.9 (0.7) 8.0 (1.3) 8.2 (1.2) 8.0 (1.1) 0.000
Notes: Sleep quality value determined by the PSQI. The table also shows p values comparing good and
poor sleepers using either independent samples t-test, Mann Whitney U test and chi square tests.
Participants with good sleep quality had PSQI global scores of less than 5 and participants with poor sleep
quality had PSQI global scores of greater than 5. Chocolate includes both food and drink.
Abbreviations: SD, standard deviation; PSQI, Pittsburgh Sleep Quality Index; IQR, interquartile range; mg,
milligrams; mins, minutes; SOL, sleep onset latency; SE, sleep efficiency; TIB, time in bed; global score, PSQI
total score; %, percentage; I, Independent samples t-test was used; M, Mann-Whitney U test was used; C,
Chi-square test was used.
124 | P a g e
7.6 Discussion
The current study found that caffeine consumption did not change dramatically across
the lifespan, with only energy drinks changing significantly across age groups.
Furthermore, decreased TIB was associated with increased caffeine consumption, and
people with poor self-reported sleep quality consumed significantly more caffeine than
people with good self-reported sleep quality scores.
The first aim of the study was to identify the sources of caffeine that contribute to an
adult’s total caffeine intake using the newly validated C-FFQ. It was found that the
average consumption of caffeine in this population was 165.1mg per day. National data,
obtained using an intensive 24-hour recall process on two occasions, suggests Australian
adults consume approximately 103–183 mg of caffeine per day (Australian Bureau of
Statistics, 2014b). This is similar to that found in the current study additionally, the
sources of caffeine across age groups were also similar in amounts to the national data
(Australian Bureau of Statistics, 2014b). In the current study the percentage of caffeine
from tea was highest in the 51–92 years age group; in the previous national data, it
steadily rose across the lifespan and was highest in the 71 years and over age group.
Furthermore, the current sample showed that as people age they are significantly less
likely to consume caffeine from energy drinks and consume less soft drinks. Finally, the
percentage of caffeine from chocolate remained steady across the lifespan in both our
sample and the previous national data. There was, however, a difference in percentage
of caffeine from coffee between the current data and the previous national data. The
current study consumed less caffeine from coffee intake in the 31–50 years age bracket
when compared to the national data, with the percentage of caffeine from coffee
remaining equal from 18–50 years and declining in the oldest age group.
The second aim of the study was to determine the relationship between caffeine and
sleep quality (variables from the subscales of the PSQI). The current study showed that
total caffeine consumption had a small negative correlation with TIB. This result is
similar to a previous cross-sectional survey study (Sanchez-Ortuno et al., 2005) however,
in contrast with Sanchez-Ortuno and colleagues (2005) total caffeine consumption was
not correlated with sleep efficiency. The contrast in findings could be explained by the
differences in measuring caffeine. The current study used a validated food frequency
125 | P a g e
questionnaire to estimate total caffeine intake and not a generic “cups per day” or
“number of drinks in the last week” question as in other studies (Brick et al., 2010; Lund
et al., 2010; Sanchez-Ortuno et al., 2005; Zunhammer et al., 2014), nor did it rely on
participants knowing what beverages they consume contain caffeine. Furthermore, the
mean caffeine intake value is greater in the study by Sanchez-Ortuno and colleagues
(2005; mean caffeine intake across the day: 225mg) and perhaps is large enough to
impact sleep compared to the current study which recorded an average of 165.1mg per
day.
The current study showed that total caffeine consumption was not correlated with
subjective sleep quality, medicinal sleep aids, sleep disturbance, or daily dysfunction due
to sleepiness, similar to other studies (Brick et al., 2010; Lund et al., 2010). However,
this is different to experimental laboratory studies that have found caffeine
consumption negatively impacts sleep quality (Hindmarch et al., 2000; Shilo et al., 2002).
This could be explained by the dose of caffeine consumed, as laboratory studies tend to
give higher amounts of caffeine to participants than the habitual amount recorded in the
current study. For example Carrier et al. (2007) gave 100mg three hours before bed
followed by an additional 100 mg one hour before bed time. Additionally, given that the
time of caffeine consumption was not recorded in this study it is possible that it was
consumed earlier in the day and therefore had less of an effect on sleep. Finally, the
results of the current study indicate a global relationship between sleep and caffeine
consumption and it is highly possibly that poor or short sleep drives caffeine
consumption rather than the other way around. It would be beneficial for future studies
to measure the timing of caffeine consumption to further investigate if habitual caffeine
consumption interferes with sleep.
The third aim of the study was to determine the difference in caffeine consumption
between good and poor sleepers as measured by the PSQI global score (a score which
included all the variables to determine an overall score of sleep quality). There was a
significant difference in caffeine consumption between participants who were classified
as good sleepers and those classified as poor sleepers. Good sleepers on average
consumed 67 mg less caffeine per day; approximately one cup of instant coffee.
However other field studies (Brick et al., 2010; Lund et al., 2010; Zunhammer et al.,
126 | P a g e
2014) have found that the PSQI global score was not related to the amount of caffeine
consumed. The contrast in findings could be explained by the differences in the ways
caffeine consumption was measured as discussed above. The strength of the current
study is that it measured all sources of caffeine and examined a broad population with a
wide age range throughout the year, that is, not specifically medical students (Brick et
al., 2010) or students during exam period (Zunhammer et al., 2014).
The current study provides an overview of the average daily caffeine habits of adults
across the lifespan and compares it to the sleep habits. However, some limitations need
to be taken into consideration. Firstly, we did not record the participants’ weight or
height measurements to calculate the body mass index. The relationship between sleep
and weight is well-documented (Cappuccio et al., 2008), with shorter sleep and poorer
sleep quality associated with obesity. Secondly, the general population sample means
that the results may not be generalizable to groups who consume more caffeine such as
shift-workers. Therefore, it is not possible to generalise this information to shift-workers
as they have different sleep and wake patterns which would potentially increase
caffeine use and worsen sleep quality. Furthermore, there are limitations to self-report
sleep compared to objective measures such as actigraphy and Polysomnography. While
objective measures are preferred, observational studies have widely used subjective
measures previously due to cost and logistics. In adults, it has been shown that
subjective measures of sleep are equal to objective measures of sleep (Armitage,
Trivedi, Hoffmann, & Rush, 1997; Zinkhan et al., 2014).
7.7 Conclusions
This study demonstrates the importance of accurately measuring caffeine sources as

patterns of consumption can differ across the lifespan. Additionally, while this study
found higher amounts of caffeine consumed related to decreased time in bed there was
no association between subjective sleep quality and the amount of caffeine consumed.
This could be due to the fact daily caffeine consumption was, on average, low to
moderate or it could be because caffeine was not consumed close to bed time. This
should be the focus of future research. With caffeine being consumed by a wide range
of the population and found in many beverages and foods, it is necessary to understand
127 | P a g e
the full health implications and how it may interfere with our daily behaviours such as
sleep.
128 | P a g e
Chapter 8
The Relationship Between Caffeine, Sleep and
Behaviour in Children
Published in Journal of Clinical Sleep Medicine

Reference:
Watson, E., Banks, S., Coates, A., and Kohler, M. The Relationship between Caffeine,
Sleep and Behavior in Children, J Clin Sleep Med. 2017 (in press).
129 | P a g e
The current chapter identifies common sources of caffeine and the average amount
consumed in children using the caffeine food frequency questionnaire. The caffeine food
frequency questionnaire has been validated in adults, but yet to be validated in children.
Furthermore, this chapter shows a relationship between caffeine consumption and sleep
and behaviour variables. However, the relationship between caffeine consumption and
behaviour appears to be partially mediated by sleep. This is the first study to use
validated measures to capture sleep, caffeine and behaviour and identify a relationship
in children.
8.2 Abstract
Study Objectives: To examine caffeine consumption from various dietary sources in a

cohort of Australian children and the relationship between caffeine consumption, sleep
and daytime behaviour.
Method: Children aged 8-12 years and their parents/guardians completed a battery of
questionnaires. Children completed a caffeine questionnaire while parents completed
questionnaires regarding demographics, sleep and behaviour.
Results: The final sample consisted of 309 children (10.6 ±1.3 years, male=48%) and
corresponding parent reports. On average 10.2 ±17.4mg/day of caffeine was consumed
with a range of 0-151mg/day. Of the children who consumed caffeine (87% of the
sample) the largest contributor was coffee and tea; making up 41% of total caffeine
intake and sodas (soft drinks) contributed to 40% of caffeine intake. Total caffeine
consumption was significantly associated with sleep routine (r=0.152); morning
tiredness (r=0.129); restless sleep (r=0.113); and internalizing behavioural problems
(r=0.128). Using path analysis, caffeine consumption was positively associated with
morning tiredness (β=0.111, p=0.050) which was positively associated with internalizing
behaviours (β=0.432, p<0.001). The addition of sleep routine and restless sleep to the
model led to a complete mediation of caffeine consumption on morning tiredness, as
well as a partial mediation of the association between morning tiredness and internal
behaviours.
130 | P a g e
Conclusions: In 8-12-year-olds, the primary sources of caffeine are coffee/tea and sodas.
Overall mean caffeine consumption is small by adult standards but has an impact on
behaviour and sleep in children. The effect on behaviour is mediated by disrupted sleep,
indicating that caffeine is a contributor to sleep problems and related behaviour in
children.
8.3 Brief summary:
This study is important for clinicians who manage children reporting sleep and/or
behavioural problems. There is limited data available regarding caffeine intake and the
relationship with sleep and behaviour in children. Caffeine intake in 309 children aged
8-12 years was associated with reduced sleep quality, which was in turn related to
internalising behaviour symptoms. A large portion of children reported caffeine use,
with soda, coffee and tea being primary sources. Future research should investigate
timing of caffeine intake in children as this may further influence the impact of caffeine
consumption on child sleep and behaviour.
131 | P a g e
8.4 Introduction
Caffeine is commonly consumed in chocolate, soda, energy drinks, coffee and tea; and
of these products many are marketed toward children and adolescents using enticing
slogans and branding (Bramstedt, 2007; Warner, 2005). The 2007 Australian National
Children’s Nutrition and Physical Activity Survey collected data on 4,487 children aged 2-
18 years and found that children aged 4-13 years consumed 7-47mg of caffeine per day
(Commonwealth Scientific and Industrial Research Organisation [Australia] et al., 2008).
More recently, the Australian Bureau of Statistics (ABS) reported that children aged 4-18
years in 2011-2012 consumed 6-52mg of caffeine per day (Australian Bureau of
Statistics, 2014a). This amount of caffeine consumption is within the suggested
guidelines provided by other countries such as Canada (Health Canada, 2012), however,
even a modest amount of caffeine has been shown to impact sleep in children aged 13
years old (Lodato et al., 2013), and therefore could impact child behaviour, academic
performance and well-being (Owens, Mindell, & Baylor, 2014).
Few studies have intended to explore the relationship between caffeine and sleep in
children under 12 years with majority of studies focusing on adolescents or adults (Clark
& Landolt, 2016). Unlike adults and adolescents, school-aged children (5-12 years) may
not drink caffeinated beverages to purposefully stay awake, but rather the consumption
of caffeine may disturb their sleep leading children to become sleepy during the day
(Calamaro et al., 2012). In school-aged children studies have found that higher caffeine
consumption is associated with shorter total sleep time (TST; Calamaro et al., 2012; Li et
al., 2010; Warzak, Evans, Floress, Gross, & Stoolman, 2011) and a greater frequency of
sleep problems (Kristjansson et al., 2014). However, other studies have found that
caffeine has no effect on TST (Drescher et al., 2011), daytime sleepiness (Calhoun et al.,
2011), or sleep problems (Warzak et al., 2011). Other domains of sleep quality such as
bedtime anxiety, morning tiredness, night arousals, sleep disordered breathing and
bedtime routine have not previously been linked with caffeine consumption. These
studies in children have not directly aimed to determine the association between
caffeine and sleep, and many have used non-validated measures of caffeine.
132 | P a g e
In previous studies a range of methods have been used to collect caffeine data, for
example personal interviews (Warzak et al., 2011), phone interviews (Calamaro et al.,
2012) or a food frequency questionnaire (Drescher et al., 2011). Some of these
approaches may have under reported caffeine consumption as they have limited the
range of products asked about (Calamaro et al., 2012; Drescher, Goodwin, Silva, & Quan,
2011; Kristjansson, Sigfusdottir, Allegrante, & James, 2011), have focused on caffeine
consumption only after 6pm (Li et al., 2010), or not included a clear description of
methods used to measure caffeine intake (Calhoun et al., 2011).
Sleep has a well-documented relationship with behaviour in children with reduced or

disrupted sleep predictive of cognitive and behavioural problems such as decreased
attention, decreased memory, decreased school performance, decreased executive
function, increased hyperactivity, increased aggression, increased sick days, increased
symptoms of depression and increased mood problems (Beebe, 2011; Blunden & Beebe,
2006). Whether caffeine consumption has a similar impact on a child’s behaviour is
unclear. Higher caffeine consumption, specifically coffee and soda, has been shown to
be significantly associated with an increase in depressive symptoms (internalising
behaviours) in children aged 14 years (Fulkerson, Sherwood, Perry, Neumark-Sztainer, &
Story, 2004). Additionally, girls in grades 6-10 who report high caffeine consumption are
1.4 times more likely to complain of headaches, stomach-aches, and backaches (somatic
complaints) compared with those who consumed lower levels of caffeine (Ghandour,
Overpeck, Huang, Kogan, & Scheidt, 2004). Furthermore, increased caffeine
consumption has been shown to be related to increased aggression in children aged 14-
15 years (Kim, 2013; Kristjansson et al., 2011), and strongly related to increased violent
behaviours and conduct disorder (externalising behaviours) in children 10-12 years of
age (Kristjansson, Sigfusdottir, Frost, & James, 2013).
To date, only one study has reported a three-way interaction between aspects of sleep,
caffeine consumption and behaviour in children. Kristjansson et al (2011) conducted
structural equation modelling to investigate the direct effect of caffeine on aggressive
behaviour and the indirect effect with sleepiness and licit substance (cigarettes and
alcohol) use in children aged 14-15 years. The study found that sleepiness was partly
responsible for a relationship between caffeine and anger, suggesting 43-48% of the
133 | P a g e
association between caffeine and anger was due to mediation by sleepiness and licit
substance use. The unique contribution of sleepiness was not reported. Also, no study
has investigated this relationship across a broader range of sleep factors or with the use
of a validated assessment of caffeine consumption, sleep and behaviour in children.
The aim of this study was to determine 1) the amount and source of caffeine in 8-12-
year-old Australian children, 2) the relationship between caffeine consumption and
sleep time and sleep quality, 3) caffeine consumption and behaviour, and 4) in an
attempt to understand the inter-relationship between the variables use structural
equation modelling to determine if sleep acts as a mediator between caffeine
consumption and behaviour.
8.5 Method
8.5.1 Participants
Children were aged between 8-12 years and fluent in both reading and writing of
English. They were excluded if they had diabetes or had been clinically diagnosed with,
or were currently experiencing, an abnormality in diet, behaviour or sleep; for example,
attention deficit hyperactivity disorder, or diagnosed sleep disorders.
8.5.2 Measures
This study required both a parent and child to complete a questionnaire battery
reflecting on the child’s caffeine intake, behaviour and sleep. Questionnaires were
collected on one occasion and took approximately 45 minutes for parents/guardians to
complete and 10 minutes for the child to complete.
Parent Questionnaires
Demographic: Parents or guardians were asked to provide information on postcode,
their child’s sex, date of birth, body mass, and height. Socio-economic status (SES) was
determined by postcode, using the deciles of the Socio-Economic Indexes for Areas
(SEIFA), 2011 (Australian Bureau of Statistics, 2011). A SEIFA score is created using
information variables such as household income, education, occupation, employment
and housing (Australian Bureau of Statistics, 2006). This score is standardized against a
national mean of 1000 with a standard deviation of 100, which is then divided into decile
134 | P a g e
rankings. A lower decile ranking indicates lower SES. Body mass index (BMI) Z-scores
for children and adolescents were calculated using CDC criteria (Kuczmarski et al., 2002).
A BMI Z-score of 0.0 indicates the child is at the 50 th percentile, greater than 1.0 is
classified as overweight, and greater than 2.0 is classified as obese.
Child Behaviour Checklist for ages 6-18 years: The Child Behaviour Checklist (CBCL) was
used to measure each child’s typical behaviour (Achenbach & Rescorla, 2001). The CBCL
consists of 113 questions and asks parents or guardians to consider their child’s
behaviour over the past 6 months and rate each item as 0=not true, 1=somewhat or
sometimes true, or 2=often true or very often true of their child. The CBCL can be used
to assess a range of behaviour domains, which can be further condensed in to broad
domains of internalising behaviours, or a measure of emotional problems
(anxious/depressed, withdrawn/depressed and somatic problems; problems that are
mainly within the self), externalising behaviours is a measure of behavioural problems
(rule breaking behaviour and aggressive behaviour; problems that mainly inflict conflicts
with other people and with their expectations for their child), and total behaviour
problems this a combination of both measures.
The current study used 109 items from the CBCL and excluded four items which were
deemed problematic in this study and not essential for the described aims. The items
were automatically assigned a 0 and included items 59 (plays with own sex parts in
public), 60 (plays with own sex parts too much), 73 (sexual problems), and 96 (thinks
about sex too much). Removed items 59 and 60 were associated with thought problems
(Internalising problems) and items 73 and 96 were associated with rule breaking
behaviour (Externalising behaviour). A total problem score was computed by summing
all 109 items answered. A high score on the scale (a T-score greater than or equal to 70)
indicates a clinical problem. The CBCL is reported to have test-retest reliability across all
scales ranging from r=0.80-0.94), as well as moderate to strong internal consistency
(alpha=0.72-0.97; Achenbach & Rescorla, 2001).
Paediatric Sleep Problem Survey Instrument): The Paediatric Sleep Problem Survey
Instrument (PSPSI) is a validated questionnaire which assesses sleep problems in school-
aged children (Biggs, Kennedy, Martin, van den Heuvel, & Lushington, 2012). It contains
135 | P a g e
a total of 30 items and parents are asked to retrospectively report on their child's sleep
behaviours over the last ‘typical’ seven-day week. Items are rated on a four point Likert
scale of ‘never’, ‘rarely’ (once per week), ‘sometimes’ (2-4 times per week), and ‘usually’
(5-7 times per week). The sleep domains include: sleep routine (a regular and consistent
pattern of activities before bed; 6 items), bedtime anxiety (fear of going to bed; 4 items),
morning tiredness (being less alert in the morning; 4 items), night arousals (amount of
times the child wakes during the night; 4 items), sleep disordered breathing (snoring
during the night; 4 items) and restless sleep (movement during sleep; 4 items). Within
each scale a higher score indicates greater symptoms. The PSPSI is a validated tool in
school-aged children (Biggs et al., 2012), with internal consistency for each scale domain
ranging between 0.63-0.80 (Biggs et al., 2012). Reported T-scores were used in analyses,
with a T-score of 70 or more indicating a score above the 95 th percentile. Finally, an
additional two sleep questions were added asking about the child’s TST and sleep onset
latency (SOL). Categorical data was collected for TST and SOL on Likert scales; TST: 9-11
hours, 8-9 hours, 7-8 hours, 5-7 hours, and less than 5 hours; SOL: less than 15 minutes,
15-30 minutes, 30-45 minutes, 45-60 minutes and over 60 minutes. Test-retest
reliability of the PSPSI is reported as moderate to strong (0.36-0.90), with moderate to
strong internal consistency (alpha: 0.6-0.8; Biggs et al., 2012).
Puberty Category Scores Questionnaire: Puberty category scores (PCS; Carskadon &
Acebo, 1993) were calculated to determine the level of pubertal development of each
child. This questionnaire has five items relating to symptoms of puberty for example
hair growth and pimples, which were assessed on a four point Likert scale (1= not
started, 2= barely started, 3= definitely started, 4= seems complete). For girls the fifth
question asks if they have started menarche which required a ‘yes’, ‘no’, or ‘I don’t
know’ response. A PCS of 2-3 indicates pre-pubertal growth and development; a PCS of
4-11 indicates the progression of puberty; and a PCS of 12 indicates sexual maturation
has been completed. The measure has established validity and reliability (Carskadon &
Acebo, 1993) and parent ratings are strongly correlated with child and paediatrician
ratings of Tanner stages (Carskadon & Acebo, 1993). It was left to each parent as to
whether the parent or the child completed the questionnaire. Internal consistency of
the PCS shows reasonable reliability for both the self (α=0.67-0.70) and parent (α=0.68-
136 | P a g e
0.78) versions. Correlations between PCS Tanner ratings and paediatrician assessments
were considered strong (Carskadon & Acebo, 1993).
Child questionnaire
Caffeine Food Frequency Questionnaire (C-FFQ): The C-FFQ is a self-report questionnaire
designed to assess the average daily caffeine consumption over the previous week. The
C-FFQ was specifically designed to capture caffeine intake from beverages and foods
available in Adelaide, South Australia (location of study) between 2013 and 2015. The
questions on the C-FFQ ask about the different beverages (for example energy drinks,
soft drinks [soda], both hot and cold coffee and tea, and chocolate flavoured milk) and
foods (for example chocolate) consumed in the previous seven-days. Caffeine content
was obtained Nutrient Tables for use in Australia (NUTTAB 2010 - Australian Food
Composition Tables: Food Standards Australia New Zealand, Canberra) and specific food
manufacturer websites. Medications, supplements, and other sources of caffeine that
are not diet-related were not included. The questionnaire requires participants to select
the beverage or foods they consumed based on images of currently available products in
Australia, indicating the specific brand, size, and the number of times over the last week
each was consumed. The C-FFQ has been validated in Australian adults (Watson et al.,
20156 with successful piloting in a small group of children prior to this study. The C-FFQ
calculates caffeine values, in milligrams per day, for total caffeine and the four
subcategories of beverages and foods by calculating total caffeine consumption over the
previous week and dividing by seven. Amount consumed on specific individual days is
not determined. A Bland Altman plot was used to determine validity and indicated fair
agreement overall and for values below 300mg of caffeine per day, the agreement
improved. Test-retest reliability was strong for total caffeine (r=0.90, p<0.001) and for
the four individual categories (r=0.94-1.00, p<0.001). Cohen’s κ showed reasonable
agreement between the C-FFQ1 and C-FFQ2, κ= 0.65; 95%; CI: 0.52, 0.78; p<0.001
(Watson et al., 2016).
8.5.3 Procedure
This was a cross-sectional study conducted in South Australia between October 2013
and July 2015. The study was approved by the Human Research Ethics Committees of
the University of South Australia (Adelaide, Australia; HREC number: 30885), the
137 | P a g e
Department for Education and Child Development (DECD), and the Catholic Education
Office. Written informed consent was obtained from parents/guardians of all children
prior to commencement. As children were 8-12 years they did not provide their own
consent, however they were free to withdraw at any time. Recruitment of participants
was completed through the community and through schools.
Community approach: Community recruitment was through fliers placed in the Adelaide
metropolitan area. Locations included universities, hospitals, public libraries,
community noticeboards, medical centres, and entertainment facilities. Interested
participants either emailed or called for further information. Potential participants were
then sent an information sheet, consent form and all questionnaires with completed
information returned in a supplied reply-paid envelope. Seventy-two questionnaire
packs were distributed amongst the community and 53 complete packs were returned
therefore, a 74% acceptability rate.
School approach: South Australian public and private schools were also approached
from a variety of socio-economic areas. Socio-economic areas for South Australian
public schools were determined using the 2012 Index of Educational Disadvantage
released by the DECD. If an individual school wished to take part in the study, a member
from the research team presented the study to the children and teachers and the school
then distributed information sheets, consent forms and parent questionnaires to all
children aged 8-12 years. On return of questionnaires and consent forms, the
researcher returned to the school to supervise the children during class while they
completed their questionnaires. Questionnaire packs were distributed across 17
schools; however, it is not possible to ascertain the exact number of questionnaire packs
that were distributed across schools.
8.5.4 Statistical analyses

For the C-FFQ, PSPSI, and CBCL a selection of N/A or a non-response resulted in an item
score of zero to underestimate rather than overestimate problems in the sample. All
data were checked for normality, and caffeine variables and some sleep variables (sleep
routine, bedtime anxiety, night arousals and sleep disordered breathing) were found to
be positively skewed. Total caffeine intake was log transformed and sleep routine was
138 | P a g e
inversely transformed to correct to a normal distribution for parametric testing.
However, caffeine intake data from individual beverage and food categories were
skewed such that they could not be transformed into a normal distribution.
Mean, standard deviation, median, interquartile range, and range was calculated to
describe the population and to show where the differences between populations
occurred. Independent samples t-tests were used to determine differences between
groups and sex differences in caffeine consumption. To determine if there were
differences in total caffeine consumption across each response category of TST and SOL,
one-way ANOVA was used.
Spearman correlations were used to assess relationships between caffeine, sleep and
behaviour variables. A path analysis was also undertaken to determine if sleep factors
mediated the relationship between caffeine intake and behaviour. A path analysis
allows reporting on direct and indirect pathways and provides estimates of explained
variance between variables. The path analysis was conducted using Analysis of Moment
Structures (AMOS; version 23.0; Amos Development Corp., Crawford, FL). The path
analysis was guided by the Spearman correlation variables calculated. Bootstrap
maximum likelihood estimation was used (1000 samples). There were three missing
values for SES which was replaced by the sample mean. Modification indices were
examined post hoc to ascertain if any potentially relevant pathways were omitted from
the model. Significance was set at p<0.05.
139 | P a g e
8.6 Results
Study participants:
Figure 8.1: Description of recruitment strategies and participant flow from each source
Majority of participants were recruited through approaching schools and children

completing questionnaires under supervision during class. A total of 309 participants
(Figure 8.1) were included in the final analysis, with an average age of 10.6 ±1.3 years
(mean ±SD) and 47.9% male. Females on average had a puberty score of 5.2 (±2.4) and
males 3.8 (±1.3), this was significantly different t(299)=-6.317, p<0.001. Caffeine
consumption did not differ between boys and girls (p=0.844). The average SES rating
was 6.9 (±2.6) and on average participants were in the healthy BMI range (BMI z-score =
0.22kg/m2 ±1.63). Total caffeine intake was on average 10.2mg (±17.4) per day, which is
equal to a 50g solid milk chocolate bar or 30mL of RedBullTM. When considering only
participants who consumed caffeine, the average increased to 11.8mg (±18.1) per day.
Majority of the participants did not wet the bed (95%); sleep walk (90%), experience
140 | P a g e
hyperhidrosis (97%), or report bruxism (91%). Furthermore, of the sample, 64.7%
achieved 9-11 hours of sleep per night but only 20.7% took less than 15 minutes to fall
asleep. Additional participant descriptive details are reported in (Table 8.1). Highest
level of parental education was also reported indicating all mothers completed high
school, with 49.3% of the mothers completing a University degree and 27.2% completing
a technical and further education (TAFE) degree. In contrast 1.3% of the fathers only
completed primary school and the remaining fathers completed high school with 49.2%
of fathers completing a University degree and 19.7% a TAFE degree.
Table 8.1: Means (standard deviation), medians (interquartile range), and range of values for caffeine
variables, sleep variables and behaviour variables.
Mean (SD) Median (IQR) Range of responses
Caffeine variables (mg per day)
Total caffeine 10.2 (17.4) 3.4 (12.6) 0-150.8
consumption
Coffee & tea 4.2 (11.0) 0.0 (4.4) 0-101.8
Chocolate 1.3 (1.5) 0.8 (1.5) 0-11.1
Sodas (soft drinks) 4.1 (8.0) 0.0 (5.1) 0-45.8
Energy drinks 0.6 (4.5) 0.0 (0.0) 0-57.1
Sleep variables (T-scores)
Bedtime anxiety 49.0 (9.2) 45.0 (6.0) 44-95
Morning tiredness 49.5 (10.2) 46.0 (18.0) 38-83
Night arousals 48.5 (8.7) 47.0 (14.0) 41-88
Sleep routine 49.5 (10.1) 47.0 (12.0) 41-95
Sleep disordered 49.7 (9.8) 44.0 (11.0) 42-97
breathing
Restless sleep 46.0 (8.2) 45.0 (12.0) 37-75
Behaviour variables (T-scores)
Internal behaviour 53.6 (11.2) 54.0 (17.0) 33-86
External behaviour 50.3 (11.1) 50.0 (14.0) 33-82
Total problems 52.1 (11.3) 51.0 (15.5) 24-79
Abbreviations: SD, standard deviation; IQR, interquartile range; mg, milligrams
Notes: Variables are calculated from the Caffeine Food Frequency Questionnaire, Paediatric Sleep
Problem Survey Instrument and Child Behaviour Checklist. Chocolate included both food and drinks
141 | P a g e
Caffeine description
Overall 87% of the sample consumed caffeine during the week before testing. Of the
children who consumed caffeine, the largest contributor to caffeine consumption was
coffee and tea; making up 41% of total caffeine intake. However, only 84 (27.4%)
children reported having coffee and tea over the past week. Sodas (soft drinks)
contributed to 40% of caffeine intake with 116 (37.8%) participants consuming this
beverage type over the last week. Only 8 participants consumed energy drinks,
contributing to 13% of the caffeine intake. Foods and drinks containing chocolate were
consumed by 245 participants (79% of sample), however this contributed to only 6% of
caffeine intake.
Caffeine consumption differed by amount of sleep, F(3,305)=7.55, p<0.001. Post hoc

analyses showed children who slept 9-11 hours consumed significantly less caffeine
compared to children who slept 7-8 hours per night (p<0.001). There were no
differences in caffeine consumption across SOL categories.
Correlations
Age, puberty, and SES were correlated with total caffeine, sleep and behavioural
variables. Increased age, higher puberty scores and higher morning tiredness values
were correlated with increased total caffeine consumption (r=0.166, p=0.004; r=0.582,
p<0.001; r=0.126, p=0.027, respectively). Higher SES was correlated with decreased
total caffeine consumption (r=-0.162, p=0.004) and better internalising (r=-0.119,
p=0.048), externalising (r=-0.159, p=0.007), and total behavioural problems (r=0.167,
p=0.005). The results indicated that BMI Z-score was not correlated with any variables.
Higher caffeine consumption was significantly associated with higher scores in sleep
routine (r=0.152, p=0.007), morning tiredness (r=0.129, p=0.023), and restless sleep
(r=0.113, p=0.047). Caffeine consumption was not associated with night arousals
(p=0.092) and sleep disordered breathing (p=0.694). Regarding daytime behaviour,
higher total caffeine consumption was associated with worse internalising behavioural
problems (r=0.128, p=0.024), but not externalising (p=0.244) or total problems (p=0.052;
Table 8.2).
142 | P a g e
143 | P a g e
Path analysis
A path analysis investigated the impact of caffeine intake on internalising behaviours

through its impact on sleep. Initially a model that included only a direct path from
caffeine consumption to morning tiredness, and morning tiredness to internalising
behaviour was evaluated (Figure 8.2). As expected, total caffeine intake was positively
associated with morning tiredness (β=0.111, p=0.050) which was positively associated
with internalising behaviour (β=0.432, p<.001).
When the additional sleep variables associated with caffeine consumption (sleep routine
[inverse] and restless sleep) were included in the model the direct relationship between
caffeine consumption and morning tiredness was no longer significant (β=0.013,
p=0.794; Figure 8.3). However significant paths from caffeine consumption to restless
sleep (β=0.149, p=0.008) and sleep routine [inverse] (β = -0.162, p=0.004) emerged.
Both sleep routine and restless sleep were significantly related to morning tiredness (β=-
0.378, p<0.001; and β=0.245, p<0.001, respectively) and internalising behaviour (β=-
0.151, p=0.007; and β=0.141, p=0.008 respectively; Table 8.3), indicating mediation. It is
important to note that the relationship between morning tiredness and internalising
behaviours also decreased when sleep routine and restless sleep were added to the
model, indicating partial mediation for this aspect of the overall model. Measure of fit
was determined by Schreiber, Nora, Stage, Barlow, & King (2006) and the path analysis
showed good overall fit (Table 8.4).
144 | P a g e
Table 8.2: Spearman correlation values for all relationships between variables
SR BA MT NA SDB RS Int Ext TBP

TCC .152** 0.045 .129* 0.096 0.023 .113* .128* 0.066 0.111
SR - .240** .403** .309** 0.106 .246** .305** .283** .360**
BA - .113* .153** -0.099 0.042 0.095 -0.032 0.046
MT - .315** .164** .337** .393** .373** .429**
NA - .257** .353** .315** .192** .282**
SDB - .413** .266** .188** .253**
RS - .298** .271** .354**
Int - .633** .854**
Ext - .873**
TBP -
*p<0.05, ** p<0.01, *** p<0.001

Abbreviations: TCC, total caffeine consumption; SR, sleep routine; NA, night arousals; SDB, sleep
disordered breathing; RS, restless sleep; BA, bedtime anxiety; MT, morning tiredness; Int, internalising; Ext,
externalising; TBP, total behavioural problems.
Figure 8.2: Direct pathway analysis with Beta (β) values between the controlled variables (age and
socio-economic status) and between total caffeine consumption, morning tiredness and internalising
behaviours.
*p<0.05, ** p<0.01, *** p<0.001
145 | P a g e
Figure 8.3: Final Model. Indirect pathway analysis with Beta (β) values between the controlled variables
(age and socio-economic status) and between total caffeine consumption, sleep routine, restless sleep,
morning tiredness and internalising behaviours.
A straight arrow represents a hypothesized effect.
*p<0.05, ** p<0.01, *** p<0.001
Table 8.3: A Pathway analysis summary, with standardized and unstandardized regression weights for
the model
Standardized Unstandardized Standard Critical 95% Confidence

coefficients (β) coefficients Error Ratio Intervals
Age → Total Caffeine Consumption (log) 0.147** 0.062 0.024 2.639 0.013, 0.107
SES → Total Caffeine Consumption (log) -0.156** -0.033 0.012 -2.801 -0.057, -0.010
Total Caffeine Consumption (log) → -0.162** -0.001 0.000 -2.890 -0.002, 0.000
Sleep Routine (Inverse)
Total Caffeine Consumption (log) → 0.149** 2.255 0.852 2.645 0.464, 3.952
Restless Sleep
Total Caffeine Consumption (log) → 0.013 0.247 0.945 0.261 -1.795, -2.225
Morning Tiredness
Sleep Routine (Inverse) → Morning -0.378*** -1131.178 151.301 -7.476 -1446.565, -
Tiredness 805.311
Restless Sleep → Morning Tiredness 0.245*** 0.304 0.063 4.851 0.192, -0.432
Age → Morning Tiredness 0.119* 0.943 0.391 2.411 0.202, 1.639
Sleep Routine (Inverse) → Internal -0.151** -493.573 183.672 -2.687 -839.84, -93.660
Behaviours
Morning Tiredness → Internal 0.318*** 0.348 0.063 5.487 0.233, 0.471
Behaviours
Restless Sleep → Internal Behaviours 0.141** 0.192 0.073 2.633 0.069, 0.327
*p<0.05, ** p<0.01, *** p<0.001
146 | P a g e
Table 8.4: Summary of the Model fit indices
Fit measure Initial model Final model ‘Acceptable’ fit criteria

2 2
χ 17.408 17.440 Smaller χ = better fit
p value (df) 0.004 (5) 0.042 (9) p>0.05, positive df
2
Normed χ 3.482 1.938 <2
RMSEA 0.090 0.055 <0.06
90% CI (p close) 0.046-0.137 (0.064) 0.010-0.094 (0.365) 0-0.08, (>0.05)
CFI 0.873 0.961 >0.95
Hoelter’s 0.05 and 0.01 critical 196-267 299-383 >200
N
NFI 0.839 0.927 >0.95
Abbreviations: CFI, comparative fit index; CI, confidence interval; df, degrees of freedom; RMSEA, root
2
mean square error of approximation; χ , chi square; NFI, normed fit index.
Acceptable fit criteria taken from Schreiber et al (2006).
8.7 Discussion
Overall this study found that majority of children in this cohort were consuming caffeine,
and consumption ranged between 0-151mg per day, which is equivalent to a 750g of
milk chocolate or 500mL of RedBuillTM. Whilst coffee and tea contributed the greatest
amount of caffeine per serve, the most commonly consumed caffeine source was
chocolate. Caffeine consumption did not differ between sexes, but there were
significant associations between caffeine consumption and age, SES, and puberty status.
This is similar to results from previous studies who have needed to control for SES and
age on caffeine consumption and adjusted for each in statistical tests (James,
Kristjansson, & Sigfusdottir, 2011; Kristjansson et al., 2011; Li et al., 2014; Li et al., 2010;
Orbeta, Overpeck, Ramcharran, Kogan, & Ledsky, 2006). Furthermore, caffeine
consumption was significantly associated with morning tiredness, sleep routine and
restless sleep, as well as problematic internalising daytime behaviour. Modelling of the
associations between these factors indicated that sleep variables completely mediated
the relationship between caffeine and behaviour.
The first aim of the study was to determine the average caffeine intake and the sources
of the caffeine consumption. Of the children who consumed caffeine, 41% of caffeine
came from coffee and tea, 40% from sodas (soft drinks), 13% from chocolate food and
147 | P a g e
drinks and 6% from energy drinks. This was similar to previous Australian national data
which showed that for children aged 9-13 years 9% of caffeine intake was derived from
chocolate, 38% from sodas (soft drinks), 5% from coffee, 19% from tea, and only 1%
from energy drinks (Australian Bureau of Statistics, 2014a). Icelandic data (Kristjansson
et al., 2011a) looking at 7,348 14-15-year-olds reported 76.3% of their sample consumed
caffeine (as a beverage) and of the four beverages surveyed (soft drink, energy drink,
coffee and tea) the most consumed was cola drinks, followed by energy drinks. When
considering beverages only in our sample sodas (soft drinks) were also the most
consumed, however energy drinks were the least consumed.
The second aim was to determine if there was a relationship between caffeine
consumption, sleep and daytime behaviour. Caffeine consumption was significantly
related to sleep routine, morning tiredness and restless sleep in children, with increasing
caffeine correlated with increasing sleep problems. There was also significant between
group differences in TST, but not SOL. Less TST as caffeine consumption increased is
consistent with other literature in children (Calamaro, Yang, Ratcliffe, & Chasens, 2012;
Li et al., 2010; Warzak et al., 2011); while similarly no relationship with SOL was found in
a sample of adolescents (Astill, Verhoeven, Vijzelaar, & Someren, 2013). Regarding
broader sleep problems, sleep routine and restless sleep, specifically in relation to
caffeine intake has not been investigated previously in any age group. Li and colleagues
(2014) found an increase in the number of nocturnal awakenings when children aged 5-
12 years consumed caffeine after 6pm. While Pollak and Bright (2003) found that
increased total caffeine consumption (including beverages, condiments and
medications) increased wake after sleep onset in children aged 12-14 years. Finally,
similar to our results, Orbeta et al (2006) showed that caffeine consumption increased
the risk of morning tiredness by 1.8 times in older children and adolescents in grades 6-
10, however caffeine intake comprised of only soda and coffee.
Finally, caffeine was significantly associated with internalising behaviour. Internalising

behaviour is defined as an over-control of emotions, which includes social withdrawal,
demand for attention, feelings of worthlessness or inferiority, and dependency
(Achenbach & Edelbrock, 1978; McCulloch, Wiggins, Joshi, & Sachdev, 2000). Symptoms
commonly classified as internalising type include anxiety, depression and
148 | P a g e
psychosomatization. Previous studies have reported similar findings with increases in
somatic complaints amongst caffeine consumers in grades 6-10 (Ghandour et al., 2004).
Furthermore, Fulkerson et al (2004) indicated that increased caffeine consumption,
specifically coffee and sodas, was significantly associated with an increase in depressive
symptoms (Fulkerson et al., 2004).
No relationship was found with caffeine and externalising problems in our sample, which
is in contrast to Kim (2013) and Kristjansson et al (2011) who found that aggression in
children was significantly related to total caffeine intake. The differing results could be
due to different cultural approaches to anger, as the two previous studies were in
Iceland (Kristjansson et al., 2011) and Korea (Kim, 2013) whilst the present study was
conducted in Australia (Maxwell, Sukhodolsky, Chow, & Wong, 2005; Ramírez, Fujihara,
& Goozen, 2001). Furthermore, the way in which caffeine was measured varied across
studies. The current study utilized a C-FFQ which analysed caffeine intake over the last
seven days and included caffeine from both beverages and foods. The C-FFQ provides
the child with a list of all caffeinated products and sizes to assist in reporting and relying
less on memory and previous knowledge. Kim (2013) asked participants whether they
drank caffeine daily, with a simple ‘yes’ or ‘no’ response. Kristjansson et al (2011)
collected more detail and asked participants to report their daily caffeine consumption
(beverages only) as the number of glasses or cups of coffee, tea, cola drinks, or energy
drinks that contain caffeine. There were seven response options from ‘‘never’’ through
to ‘‘six glasses/cups or more’’, which therefore relied on the child knowing which
beverages contained caffeine; data from this study was reported as the mean Likert
scale response.
The final aim of this study was to determine if sleep mediates the relationship between
caffeine and behaviour. Sleep routine is determined by measuring the consistency of
bedtime and sleep patterns (Biggs et al., 2012) and can be defined as behavioural
practices that promote good sleep quality, adequate sleep duration and full daytime
alertness (American Sleep Disorders Association, Diagnostic Classification Steering
Committee, Thorpy, & Diagnostic Classification Steering Committee, 1990). Previously
good sleep routine has been shown to have a positive impact on sleep quality
(LeBourgeois, Giannotti, Cortesi, Wolfson, & Harsh, 2005). This model indicated that
149 | P a g e
sleep routine and restless sleep completely mediate the relationship between caffeine
consumption and morning tiredness but only partially mediated the relationship
between morning tiredness and internalising behaviours. This is important because it
gives further insight into the importance of sleep and further understanding of the
impact caffeine may have on the relationship between sleep and behaviour in children.
Overall, relatively little is known about the short and long term health effects of caffeine
in the paediatric population and this study extends our knowledge to how caffeine plays
a role in children’s day to day sleep and behaviour. The clinical implications extend to a
growing need to consider caffeine intake for children presenting with sleep and/or
behaviour issues. The C-FFQ (Watson et al., 2016) used in this study provides a simple
and effective means to estimate caffeine consumption in children in the clinical setting.
Further study is required using objective measures of sleep and caffeine to further
understand the mechanistic relationship between caffeine and sleep.
When interpreting these results, it is important to consider other ingredients of popular

caffeinated beverages and chocolate such as sugar, guarana, food additives and
theobromine, and their additional effects on sleep and behaviour. Higher intakes of
added sugar have also been associated with shorter sleep duration (Kjeldsen et al., 2014;
Hjorth et al., 2014) in children aged 8-11 years. Further, research looking specifically at
an association between sugar sweetened beverages and sleep in children shows
increased sugar sweetened beverage consumption is associated with decreased sleep
duration (Al-Hazzaa et al., 2014; Franckle et al., 2015; Hjorth et al., 2014) and most
consistently in males (Al-Haifi et al., 2016; Collison et al., 2010; Hitze et al., 2009). The
studies indicate further research on the short and longer term impact of sugar on sleep
and sleepiness is required in children. Guarana contains large amounts of caffeine and
could provide further caffeinated impacts on sleep and behaviour that has not been
measured in this study. The typical beverages that contain guarana were not consumed
by many of this sample of participants and therefore unlikely to contribute significantly
to the results.
The foods and beverages that contain caffeine also contain a variety of food additives
and other compounds. Food additives have been found to have a relationship with
150 | P a g e
sleep disturbances previously. Rowe (1988) reported that food colours increased sleep
disturbance and Kaplan, McNicol, Conte, & Moghadam (1989) found that by removing
food additives from hyperactive children’s diet there is a reduction in sleep disturbance,
and shorter SOL. However, it is difficult to associate these changes with food additives
specifically as many other aspects of the diet were changed. Finally, the chemical
compound theobromine is primarily found in chocolate and has shown mixed results on
its impact on sleep. Grandner, Jackson, Gerstner, and Knutson (2014) reported
theobromine to increase daytime sleepiness (Grandner et al., 2014). However,
theobromine was consumed more by people sleeping for 7-8 hours than those sleeping
more than 9 hours, with longer sleep time a potential indication of increased sleepiness
(Grandner, et al, 2014).
The strengths of this study include a sufficiently large sample size, naturalistic setting
and strong associations between valid and reliable measures. However, there are
limitations that should be considered. One consideration is that the current study
utilizes self-report and parental report surveys. Although previous research and survey
validation shows that self-reported caffeine use and sleep behaviour can provide reliable
estimates it would be ideal to obtain an objective measure of both caffeine intake and
sleep to substantiate the findings. Secondly, the timing of caffeine intake was not
reported. This is of importance as the half-life of caffeine is estimated to be four to five
hours based on reports in adults (Kaplan et al., 1997) and therefore, theoretically,
children consuming caffeine after school could be at increased risk for disturbed sleep
compared to children having caffeine during or before school. Adult studies have
investigated the impact of caffeine intake in the morning (Landolt, Werth, Borbély, &
Dijk, 1995) and evening (Drapeau et al., 2006), and while both have shown that caffeine
intake impacts sleep negatively, evening intake has the largest effect. Similar studies
have not been conducted in children and it would be beneficial to be aware of the
impact of caffeine consumption at different times of the day has on sleep and
behaviour. Thirdly, this study did not capture other factors such as physical activity that
may influence sleepiness and energy intake in children.
Overall, this study has shown that caffeine intake in children aged 8-12 years is
associated with reduced sleep quality, which in turn increases internalising behaviour
151 | P a g e
symptoms. This is especially important with such high numbers of 8-12 year olds
reported to consume caffeine. In our sample, it appears energy drinks do not contribute
largely to the caffeine intake, but it is important to be aware of the high numbers of
children consuming soft drink, coffee and tea. This is likely to be an important
consideration for clinicians dealing with children presenting with sleep and/or
behavioural problems, as caffeine consumption could be playing a role in the presented
problems. Furthermore, parents may be purchasing caffeinated drinks such as energy
drinks, iced tea, or caffeinated sodas (soft drinks) without realizing the impact on sleep
and behaviour.
152 | P a g e
Chapter 9
Total Dietary Sugar Consumption Does Not
Influence Sleep or Behaviour in Australian
Children
153 | P a g e
The current chapter shows the average amount of sugar being consumed by Australian
children. Furthermore, this chapter shows no association between sugar intake and
sleep variables and behaviour, which was further supported with exploratory analysis.
This novel study is the first study to recently analyse sugar intake and behaviour, and
discuss total sugar intake and sleep quality in children.
9.2 Abstract
Objective: Sugar consumption has been associated with disruption to sleep. This study
aimed to compare sugar intake in Australian children with current guidelines and
determine if total sugar consumption exacerbates the well-established relationship
between sleep and behaviour.
Method: In a cross-sectional study, children aged 8-12 years, and their

parents/guardians completed a battery of questionnaires. Children completed the
Australian Child and Adolescent Eating Survey Food Frequency Questionnaire (FFQ), and
parents completed a demographic questionnaire, the Paediatric Sleep Problem Survey
Instrument (PSPSI), and the Child Behaviour Checklist (CBCL). Total sugar (g) and
percentage of energy from sugar were extracted from the FFQ. Internalising,
externalising and total behavioural problems were determined from the CBCL, while
child sleep quality across a number of domains was extracted from the PSPSI.
Results: Final sample included 287 children (boys 48.8%, age: 10.7±1.3 years). Their
average sugar intake was 134.9±71.7g per day, and only 55 (19%) participants did not
exceed the recommended sugar intake limit. Correlations indicated that total sugar
intake (as a percent of total energy intake) was not significantly associated with any
sleep or behavioural domains (r range= -0.07-0.08; p range= -0.07-0.08). Logistic
regressions, controlling for demographic variables, further showed that the addition of
sugar consumption as a percentage of energy did not significantly contribute to
prediction of sleep behaviour problems (p range= 0.16-0.80).
154 | P a g e
Conclusion: Whilst a high proportion of children consumed above the recommended
amount of daily total sugar, total sugar consumption was not related to behavioural or
sleep problems, nor affected the relationship between these variables.
155 | P a g e
9.3 Introduction
Disrupted sleep in children is frequently associated with cognitive and behavioural

problems such as decreased attention, memory, school performance, executive
function, and increased hyperactivity, aggression, sick days, depressive symptoms and
mood problems (Beebe, 2011; Blunden & Beebe, 2006). There are many potential
factors that influence the relationship between sleep and behaviour, one of which is diet
(Garaulet et al., 2011; Pelsser, Frankena, Buitelaar, & Rommelse, 2010; Pelsser et al.,
2011; Peuhkuri et al., 2012). A dietary component suggested to modulate the
relationship between sleep and behaviour is sugar (Blunden, Milte, & Sinn, 2011).
Sugars are simple carbohydrates, in contrast to complex carbohydrates such as starches

and fibre, and can be classified as monosaccharides (glucose, fructose, and galactose)
and disaccharides (maltose, sucrose, and lactose). Natural sugars (in milk, fruits,
vegetables and grains) account for about half of the sugar intake in typical diets in
Australia and New Zealand, with the remainder consisting of sugars added during the
food production process (Whitney et al., 2011). The term total sugar refers to all sugars
consumed in the diet that are both added and naturally occurring. Moderate amounts
of sugars can be used to modify flavour profiles in foods and are an important source of
energy, however excess dietary sugar in children can increase the risk for obesity (Chen
et al., 2008; Malik et al., 2013; Van Cauter & Knutson, 2008; Vartanian et al., 2007),
dental caries (Moynihan & Kelly, 2013), and in adults Type-II diabetes and metabolic
syndrome (Hu & Malik, 2010; Malik et al., 2010).
There is currently minimal information regarding the impact of total sugar consumption
on children’s sleep and behaviour, which is important to consider given that children are
consistently reported to consume more than the recommended amounts of sugar
(Australian Bureau of Statistics, 2014b; Commonwealth Scientific and Industrial
Research Organisation [Australia] et al., 2008; Cook et al., 2001). Commonly, there is a
perception that high total sugar intake impairs sleep and behaviour in children. The first
major study to assess this relationship was undertaken by Prinz et al. (1980). The study
found a significant positive correlation between the amount of sugar consumed and
levels of destructive-aggressive and restless behaviour in a group of hyperactive
children. Amongst control children higher activity was associated with greater sugar
156 | P a g e
intake. Experimental work since has largely refuted the notion that sugar consumption
impacts behaviour (Bellisle, 2004; Milich et al., 1986). However, while experimental
work allows for determination of causation, it often does not give an accurate depiction
of habitual diet, and does not assist in explaining ongoing behavioural issues.
Few studies have examined the effect of habitual total sugar intake on behaviour in
school-aged children. One of the few studies measuring habitual total sugar and
behaviour is by Wolraich, Stumbo, Milich, Chenard, and Schultz (1986) who compared
the diets of hyperactive and control children aged 7-12 years measured by a three-day
food diary completed by parents. There were no significant differences between diets
of the two groups of children. However, a greater proportion of the parents with
hyperactive children reported they were restricting or had tried to restrict total sugar
intake and at the time of the study data on sugar content of food was not readily
available, making it difficult to accurately calculate total sugar value.
Very few studies have looked at the impact of total sugar consumption on sleep in
children, due to the difficulty in calculating total sugar intake previously. Previously total
sugar intake was represented by intake of carbohydrate and sugar sweetened beverages
and foods. Overall, cross-sectional studies have found mixed results regarding the
relationship between carbohydrate intake and sleep duration in children. Increased
carbohydrate consumption was found to be associated with less total sleep time (TST) in
two studies (Al-Disi et al., 2010; Firouzi et al., 2014), whilst two others have reported no
association between carbohydrate intake and TST (Awad et al., 2013; He et al., 2015).
One study showed that greater carbohydrate consumption was associated with greater
TST (Weiss et al., 2010). Furthermore, the research focusing on foods and beverages
containing high amounts of sugar suggest a negative interaction with sleep duration.
With studies showing that higher soft drink consumption is associated with reduced
sleep duration in adolescents (Al-Haifi et al., 2016) and children (Franckle et al., 2015;
Hjorth et al., 2014; Kjeldsen et al., 2014; Moreira et al., 2010), though for some studies
this effect was only found in boys (Al-Haifi et al., 2016). Only two studies have shown
longer sleep duration to be associated with higher frequencies of sugar sweetened
beverage consumption (Al-Hazzaa, 2013; Garaulet et al., 2011).
157 | P a g e
Only three studies have investigated actual sugar intake and sleep, with two studies
specifically investigating added sugar (Hjorth et al., 2014; Kjeldsen et al., 2014). The
studies investigating added sugar looked at associations with sleep duration in
overweight/obese children aged 8-11 years. Added sugar intake was assessed using
validated dietary assessment software for children and sleep duration was assessed with
actigraphy. Both studies found that reduced sleep duration was associated with higher
intakes of added sugar. The third study utilized a sample of overweight children aged
two to nine years, where parents completed a 24-hour dietary recall for their child and
additional information was asked regarding their child’s hours of nocturnal sleep
(Hunsberger et al., 2015). The study found that short sleep was associated with less
total sugar intake at midday but greater intake in the evening. Currently, no study has
investigated the association of total sugar with other sleep variables, such as sleep
quality or behavioural sleep problems.
Overall, little is known about the implications of habitually high total sugar intake and its
impact on behaviour or sleep. Results to date are largely mixed with limited study of
general behaviour and sugar intake. This study aimed to 1) determine the amount of
total sugar consumed by Australian children aged 8-12 years, and 2) examine if higher
total sugar intake is associated with worse sleep and behaviour.
9.4 Method
9.4.1 Participants
Children recruited to the study were aged between 8-12 years and fluent in both reading
and writing of English. Children were excluded if they had diabetes or had been clinically
diagnosed with sleep or behaviour problems, for example: attention deficit hyperactivity
disorder, or diagnosed sleep disorder. A power analysis was conducted using G*Power
(v 3.1.9.2, Universität Kiel, Germany) to determine the number of participants needed.
Previous research done by Kjeldsen et al (2014) on added sugar consumption and sleep
in children showed a sample of 30 children is needed for an effect size f=0.74 with 80%
power and alpha at 0.05
158 | P a g e
9.4.2 Measures
This study required both parent and child to complete a questionnaire battery assessing
the child’s dietary intake, behaviour, and sleep. Questionnaires were completed on one
occasion and took approximately 45 minutes for parents/guardians and children to
complete respectively. Parents completed questionnaires assessing child demographic,
daytime behaviour and sleep characteristics. Children completed a food frequency
questionnaire, and depending on the family and the child’s age either the parent or child
completed a questionnaire assessing pubertal status. Depending on the family and the
child’s age, either the parent or child completed the puberty category questionnaire
with the decision left to the individual family.
Parent Questionnaires
Demographic: Parents or guardians were asked to provide information on postcode,
their child’s sex, date of birth, and body mass and height from which body mass index
was calculated.
Child Behaviour Checklist for ages 6-18 years: The Child Behaviour Checklist (CBCL) was
used to measure each child’s typical behaviour (Achenbach & Rescorla, 2001). The CBCL
consists of 113 questions and asks parents or guardians to consider their child’s
behaviour over the past 6 months; rating each item as 0=not true, 1=somewhat or
sometimes true, or 2= often true or very often true of their child. The CBCL has 8
syndrome scales which can be grouped into internalizing problems, externalising
problems and total problems. Internalizing behavioural problems consists of
anxious/depressed, withdrawn/depressed and somatic problems (problems that are
mainly within the self), and externalising behavioural problems consists of rule breaking
behaviour and aggressive behaviour (problems that mainly inflict conflicts with other
people and with their expectations for their child).
The current study excluded four items which were deemed problematic in this study and
not essential for the described aims. The items were automatically assigned a 0 and
included items 59 (plays with own sex parts in public), 60 (plays with own sex parts too
much), 73 (sexual problems), and 96 (thinks about sex too much). Removed items 59
and 60 were associated with Internalizing problems and items 73 and 96 were
159 | P a g e
associated with Externalising behaviour. Therefore, the current study used 109 items
from the CBCL; by summing all 109 items answered a total problem score was
computed. A high score on the scale (a T-score greater than or equal to 70) indicates a
clinical problem (Achenbach & Rescorla, 2001). The CBCL tested reliability using test-
retest reliability, the scores were very high for all scales ranging from r=0.80-0.94;
furthermore, internal consistency was measured with alpha coefficients showed scores
ranging between 0.72-0.97.
Paediatric Sleep Problem Survey Instrument (PSPSI): The Paediatric Sleep Problem
Survey Instrument (PSPSI) is a validated questionnaire which assesses sleep problems in
school-aged children (Biggs et al., 2012). It contains a total of 30 items which parents
are asked to recall their child’s sleep behaviours over the last typical school week. Items
are rated on a four point Likert scale of ‘never’, ‘rarely’ (once per week), ‘sometimes’ (2-
4 times per week), and ‘usually’ (5-7 times per week). The sleep domains include: sleep
routine (6 items), bedtime anxiety (4 items), morning tiredness (4 items), night arousals
(4 items), sleep disordered breathing (4 items) and restless sleep (4 items). A higher
score for each scale indicates worse symptoms. The questionnaire has been validated
for use in Australian school-aged children (Biggs et al., 2012), with internal consistency
for each scale domain ranging between 0.63-0.80 (Biggs et al., 2012). Reported t-scores
were used in analyses, with a t-score of 70 or more indicating a score above the 95 th
percentile. A spearman’s Rho correlational analysis was conducted to determine the
test-retest reliability of the PSPSI with scores ranging 0.36–0.90; furthermore, internal
consistency was moderate to strong with ranges of the Cronbach’s alpha 0.6-0.8.
Finally, an additional two sleep questions were added asking about the child’s total sleep
time (TST) and sleep onset latency (SOL). Categorical data was collected for TST and SOL
on Likert scales; TST: 9-11 hours, 8-9 hours, 7-8 hours, 5-7 hours, and less than 5 hours;
SOL: less than 15 minutes, 15-30 minutes, 30-45 minutes, 45-60 minutes and over 60
minutes.
Puberty Category Scores Questionnaire: Puberty Category Scores (PCS; Carskadon &
Acebo, 1993) were calculated to determine the level of pubertal development of each
child. This questionnaire has five items relating to symptoms of puberty for example
hair growth and pimples, which were assessed on a four point Likert scale (1= not
160 | P a g e
started, 2= barely started, 3= definitely started, 4= seems complete). For girls, the fifth
question asks if they have started menarche which required a ‘yes’, ‘no’, or ‘I don’t
know’ response. A PCS of 2-3 indicates pre-pubertal growth and development; a PCS of
4-11 indicates the progression of puberty; and a PCS of 12 indicates sexual maturation
has been completed. Reliability of the PCS was tested using internal consistency with
Cronbach’s coefficient demonstrating reasonable reliability for both the child (α=0.67-
0.70) and parent (α=0.68-0.78) versions of the puberty scales. Validity was determined
by strong correlations between the child/parent Tanner stage rating and paediatrician
ratings (Carskadon & Acebo, 1993).
Child questionnaire
Australian Child and Adolescent Eating Survey Food Frequency Questionnaire (ACAES-
FFQ): The Australian Child and Adolescent Eating Survey Food Frequency Questionnaire
(ACAES-FFQ; Watson, Collins, Sibbritt, Dibley, & Garg, 2009) is the first food frequency
questionnaire (FFQ) developed specifically for Australian children aged 9-16 years. The
ACAES-FFQ is a 120-item FFQ with 15 supplementary questions regarding age, use of
vitamin supplements, food behaviours and sedentary behaviours. It is designed as a
self-administered tool to collect information about the dietary intake of children over
the previous 6 months, however can be completed by a parent or guardian. An
individual response for each food, or food type, is required, with frequency options
ranging from, ‘never’ to ‘4 or more times per day,’ but varied depending on the food.
The ACAES-FFQ is semi quantitative with a standard portion size provided for each food
item and determined using ‘natural’ serving size i.e. one slice of bread. In the absence of
a natural serving size, portion sizes were derived from the 1995 National Nutrition
Survey (unpublished data that the authors of the FFQ purchased). The ACAES-FFQ
provides a value for total sugar intake in grams as well as percent of energy derived from
confectionary intake and sweet drink separately. Validity and reproducibility has
previously been tested and found to be moderate (Watson et al., 2009). Validity of the
ACAES-FFQ was compared against a food record with Spearman's correlation
coefficients ranging from 0.09-0.35; and the weighted kappa values for the nutrient
intake data from the ACAES-FFQ time two and the food records ranging from 0.09-0.36.
Spearman correlations for test-retest reproducibility ranged from 0.34-0.53; and test-
161 | P a g e
retest results of the nutrient intake data from time one and time two showed weighted
kappa values ranging from 0.36-0.54 (Watson et al., 2009).
9.4.3 Procedure:
This study was cross-sectional and conducted in South Australia between October 2013
and July 2015. The study was approved by the Human Research Ethics Committees of
the University of South Australia (Adelaide, Australia; HREC number: 30885), the
Department for Education and Child Development, and the Catholic Education Office.
Written informed consent was obtained from parents/guardians of all children prior to
commencement. As children were 8-12 years they did not provide their own consent,
however they were free to withdraw at any time. To recruit participants the study was
advertised in the community and through schools in South Australia.
9.4.4 Statistical analysis:

Statistical analyses were conducted using SPSS version 22.0 for Windows (SPSS, Inc.,
Chicago, IL). All p values reported are 2 tailed, with significance determined at p=0.05,
unless otherwise stated. Data are presented as mean (SD) unless otherwise stated.
Initially data were cleaned so that all total energy intake values that were less than
2,090 kilojoules (kJ) or greater than 20,900 kJ for an individual record were removed as
they were deemed implausible (n=22; Rockett et al., 1997; Watson et al., 2009). All
sleep variables were non-normally distributed and could not be transformed and
therefore a median split was completed on sleep variables (bedtime anxiety, sleep
routine, night time arousals, sleep disordered breathing, restless sleep) to split in to two
groups; poor sleep quality and good sleep quality. All behaviour variables were normally
distributed.
Socio-economic status (SES) was determined by postcode, using the deciles of the Socio-
Economic Indexes for Areas (SEIFA), 2011 (Australian Bureau Of Statistics, 2011). A
SEIFA score is created using information variables such as household income, education,
occupation, employment and housing (Australian Bureau Of Statistics, 2006). This score
is standardized against a national mean of 1000 with a standard deviation of 100, which
is then divided into decile rankings. A lower decile ranking indicates lower SES. Body
mass index (BMI) Z-scores for children and adolescents were calculated using CDC
162 | P a g e
criteria (Kuczmarski et al., 2002). A BMI Z-score of 0.0 indicates the child is at the 50th
percentile, greater than 1.0 is classified as overweight, and greater than 2.0 is classified
as obese. Sugar as a percentage of energy was determined by multiplying total sugar
intake (g) by number of kilojoules in each gram of sugar (17kJ) and dividing by total
energy intake and expressed as a percentage. Healthy amounts of sugar were equal to
less than 20% of total energy intake, compared to those who consume excess amounts
of sugar or greater than 20% of total energy intake using guidelines set out by National
Health and Medical Research Council 2003 dietary guidelines for children and
adolescents in Australia (Binns et al., 2003).
To determine associations between diet and sleep variables, Spearman correlations

were conducted between all variables, and due to the multiple comparisons a Sidak
correction (Šidák, 1967) was used such that significance was set at p=0.0006. When
identifying differences between groups, independent t-tests were done on normally
distributed variables, to determine differences between categorical groups a chi square
test was completed and to determine if there were any differences in non-normal
variables a Mann Whitney U-test was completed.
The risk of sleep problems due to sugar intake, behavioural problems and demographics
was determined using logistic regression. Sleep variables were entered as the
categorical dependent variable, and behaviour variables, sugar intake as a percentage of
energy, and other cofounders were entered as predictors. Hierarchical logistic
regression was performed to identify variables that were significantly related to the
odds of having sleep problems. The results are reported as odds ratios with 95 percent
confidence intervals. The first step of the hierarchical logistic regressions included
behaviour and any demographics that needed to be controlled for (as indicated in
correlations) and step two included the addition of sugar as a percentage of energy.
9.5 Results
In total, 309 children completed the study, however, 22 questionnaires were excluded
due to implausible energy intakes (either below 2,090kJ or above 20,900kJ; Rockett et
al., 1997; Watson et al., 2009) leaving a final sample for analysis of 287 children (48.8%
male) with an average age of 10.7 (1.3) years. The full recruitment process is outlined in
163 | P a g e
Figure 9.1. On average 66.2% of the participants slept 9-11 hours per night, while 20.6%
of participants fell asleep in less than 15 minutes, and 53% fell asleep between 15-30
minutes. Of the total sample 5.2% wet the bed, 9.1% sleep walked during their sleep,
3.1% sweated at night, and 8.7% ground their teeth, at least once in the previous week.
The average puberty score for the boys was 3.85 (1.36) and for the girls 5.30 (2.45), with
girls significantly more advanced than the boys, t(277)=-6.085, p<0.001. Average sugar
intake for all children was 134.9g (71.7g) per day, representing 26.0% (7.0%) of their
total energy intake (Table 9.1). Fifty-five (19%) children met the guidelines for moderate
sugar consumption of less than 20% of energy from total sugar whereas 232 (81%)
children did not. Girls had a higher intake of sugar (as a percent of total energy intake)
compared with boys (boys: 25.0% SD: 6.5; girls: 26.9% SD: 7.3), t(285)=-2.272, p=0.02.
Figure 9.1: Description of recruitment strategies and participant flow from each source
164 | P a g e
Table 9.1: Means (standard deviation), medians (interquartile range), and range of values for sugar
variables, sleep variables and behaviour variables.
Mean (SD) Median (IQR) Reported Range

Bedtime Routine 49.1 (9.8) 44.0 (12.0) 41.0 – 95.0
Bedtime Anxiety 48.5 (8.5) 45.0 (4.0) 44.0 – 95.0
Morning Tiredness 49.3 (10.2) 46.0 (18.0) 38.0 – 83.0
Night Arousals 48.2 (8.4) 47.0 (14.0) 41.0- 81.0
Sleep Disordered Breathing 49.3 (9.1) 44.0 (11.0) 42.0 – 83.0
Restless Sleep 45.8 (8.3) 45.0 (12.0) 37.0 – 75.0
Internal Problems 53.0 (11.0) 54.0 (16.0) 33.0 – 86.0
External Problems 49.6 (10.5) 49.0 (12.0) 33.0 – 77.0
Total Behavioural Problems 51.2 (10.9) 51.0 (15.0) 24.0 – 79.0
Total Sugar (g) 134.9 (71.7) 124.9 (87.8) 21.2 – 498.0
Sugar (%E) 26.0 (7.0) 25.8 (7.3) 6.0 - 51.5.0
Sweet drinks (%E) 4.1 (4.9) 2.0 (5.0) 0.0-32.0
Confectionary (%E) 7.7 (6.7) 5.0 (8.0) 0.0-44.0
SES category 6.5 (2.7) 7.0 (5.0) 1.0 – 10.0
BMI-Z score 0.2 (1.66) -0.2 (1.1) -2.7 - 12.9
Puberty Category Score 4.6 (2.1) 4.0 (2.0) 3.0 - 12.0
Abbreviations: SD, standard deviation; SES, socio-economic score; g, grams; %, percentage; BMI-Z, body
mass index; n, number of participants; IQR, interquartile range; %E, percent of energy.
Notes: Values calculated from the Australian Child and Adolescent Food Frequency Questionnaire,
Paediatric Sleep Problem Survey Instrument and Child Behaviour Checklist.
Spearman correlations (Table 9.2) indicated that sugar intake (as a percent of total
energy intake) had no association with any outcome variable (p range = -0.07-0.08).
Furthermore, age, SES and BMI-Z scores were not related with the main outcome
variables. The correlations did however indicate a strong relationship between sleep
and behavioural problems, aside from bedtime anxiety which was not associated with
any behavioural variables.
165 | P a g e
Table 9.2: Spearman correlation values for all relationships between variables
SES BMI_Z PCS SR BA MT NA SDB RS Int Ext TBP %SUG

AGE -0.14 0.04 0.61* 0.04 -0.09 0.14 -0.08 0.05 0.06 0.02 -0.01 0.01 -0.02
SES 1 0.08 -0.22* -0.01 0.05 0.02 .12 -0.07 -0.09 -0.09 -.13 -.14 -0.07
BMI_Z 1 0.06 0.06 -0.08 0.01 -0.05 0.07 0.05 -0.00 -0.01 0.02 -0.03
PCS 1 0.13 -0.14 0.16 0.07 0.25* 0.20 0.07 0.03 0.02 -0.03
SR 1 0.20* .40* .31* 0.09 .25* 0.29* .26* .36* 0.03
BA 1 0.08 .12 -.12 0.04 0.08 -0.05 0.04 -0.02
MT 1 .31* .16 .35* 0.40* .38* .46* -0.01
NA 1 .27* .34* 0.32* .19 .29* 0.08
SDB 1 .40* 0.25* .17 .23* 0.04
RS 1 0.30* .26* .35* -0.02
Int 1 .60* .85* -0.02
Ext 1 .86* 0.00
TBP 1 -0.01
Abbreviations: SES, Socio-economic status; BM-Z, body mass index; PCS, puberty category scale; SR, sleep routine; BA, bedtime anxiety; MT, morning tiredness; NA, night
arousals; SDB, sleep disordered breathing; RS, restless sleep; Int, internal behavioural problems; Ext, external behavioural problems; TBP, total behavioural problems; %SUG,
sugar intake as a percentage of energy intake.
*p<0.0006 Sidak correction applied
166 | P a g e
Children were grouped as above or below the median score for each sleep variable
(sleep routine, sleep disordered breathing, restless sleep, and night time arousals).
There was no difference between children scoring above or below the median score for
sleep routine, sleep disordered breathing, and restless sleep on measures of gender
(p=0.959, 0.170, 0.115 respectively), age (p=0.405, 0.350, 0.134), SES (p=0.825, 0.256,
0.316) and BMl-Z score (p= 0.976, 0.087, 0.35). For night time arousals, children above
the median score reported lower SES category compared with those below the median,
t(285)=-1.988, p=0.05.
Logistic regressions were complete to further ascertain the effects of behaviour and
sugar on sleep problems, controlling for demographic variables shown in previous
analyses to influence the dependent variable. Only sleep variables which showed a
relationship with behaviour were included (that is restless sleep, sleep disordered
breathing, sleep routine and night time arousals). The addition of behaviour (internal,
external and total behavioural problems) was significantly associated with more intense
sleep problems in each instance. For night arousals SES was also included as a risk
factor. Initial models showed that behaviour (and SES for night arousals), significantly
predicted sleep problems. However, the addition of sugar consumption as a percentage
of energy did not significantly contribute to prediction of sleep behaviour problems in
any domain, p-values ranging between 0.16-0.80.
Further, the population was split into high sugar intake (defined as above the guidelines;
i.e. >20% of energy intake) and low sugar intake (defined as below the guidelines; i.e.
<20% of energy intake) and further exploratory analyses were conducted. There were
no differences in age, sex, SES, BMI-Z scores, behavioural variables, and sleep variables
between these groups, p values range from 0.12-0.99.
Since the current study could not separate added sugar from natural sugar final
exploratory analyses were conducted to determine if commonly consumed non-core
foods with high added sugar content, namely confectionary and sweet drinks, were
related to total sleep time, sleep quality domains and behavioural variables. Overall,
only one child in the sample consumed a diet that has no energy derived from sweet
drinks and/or confectionary (Table 9.1). The percent of energy derived from these foods
did not differ between boys and girls for either sweet drinks (boys: 3.8%E [4.4], girls:
167 | P a g e
4.3%E [5.3]; p= 0.325) or confectionary (boys: 7.8%E [7.0], girls: 7.6%E [6.5]; p= 0.793).
No sleep or behavioural variables were associated with either percentage of energy
derived from confectionary or sweet drinks (see Table 9.3 for Spearman coefficients and
significance values) and no difference between hours of TST (p=0.409, p=0.841).
Table 9.3: Correlations between sleep and behavioural variables and percentage of energy derived from
confectionary or sweet drinks
%E Sweet Drink %E Confectionary

Sleep Routine r 0.017 -0.069
p 0.774 0.246
Bedtime Anxiety r -0.038 0.03
p 0.517 0.619
Morning Tiredness r 0.067 0.002
p 0.259 0.972
Night-time Arousals r 0.102 0.038
p 0.085 0.522
Sleep Disordered Breathing r 0.059 0.093
p 0.322 0.117
Restless Sleep r -0.008 -0.09
p 0.898 0.128
Internalizing problems r 0.055 -0.01
p 0.352 0.872
Externalising problems r 0.07 -0.006
p 0.234 0.917
Total Problems r 0.046 0.001
p 0.44 0.985
Abbreviations: %E, percent of energy
9.6 Discussion
This study found that majority (81%) of children aged 8-12 years exceeded the 2007
dietary recommendations for total sugar intake (less than 20% of daily energy intake).
As previous research has shown, sleep was strongly associated with daytime behaviour
(internalising, externalising and total behaviour; Beebe, 2011; Blunden & Beebe, 2006),
further indicating the importance of sleep for daytime functioning in child populations.
This study demonstrated for the first time that diets containing a high proportion of
total sugar do not exacerbate the relationship between parental reported sleep and
behaviour. The power analysis suggested that a sample size of 30 was needed for an
effect size f=0.74 with power at 80% and alpha at 0.05, indicating that if this relationship
was present it would have been found in this sample.
168 | P a g e
The first aim of the study was to determine the average amount of total sugar consumed
in this cohort of children aged 8-12 years. The 2003 Dietary Guidelines for Children and
Adolescents in Australia recommend consuming only moderate amount of total sugars,
and for most Australians consumption of up to 20% of energy as sugars is compatible
with a healthy diet (Binns et al., 2003). The current study found that sugar intake as a
percent of energy was approximately 26% (boys: 25.0%, girls: 26.9%) and similar to
previous findings of Australian children. Data from the 2011 National Nutrition Survey
showed that the mean contribution of total sugars to energy intake was 22.6% for
children aged 9-13 years, equating to 120.4g per day (boys: 125.4g, girls: 115.3g)
(Australian Bureau of Statistics, 2014b). Consistent with this finding, another large scale
Australian-based survey in 2007 showed that total sugar intake in 9-13-year-old children
accounted for 23.7% of their daily energy intake equal to 135.4g per day (boys: 145.9g,
girls: 124.4g) (Commonwealth Scientific and Industrial Research Organisation [Australia]
et al., 2008). In contrast, older Australian data from 1995 showed in a sample of
children aged 10-15 years that boys consumed 174g per day and girls 137g per day of
total sugar (Cook et al., 2001). Taken together, these data suggest a reduction in average
total sugar consumption over time, however our study indicates that many Australian
children continue to consume above the recommended amount.
The second aim of this study was to determine whether total sugar consumption
impacted the relationship between sleep and behaviour. These findings suggest that
habitually high total sugar diets do not impact sleep quality, or parental reported
behaviour, and do not change the association between the two. This supports previous
early findings which suggest that sugar intake in children does not influence behavioural
difficulties (Bellisle, 2004; Milich et al., 1986; Wolraich et al., 1986). Previous studies
which investigated the association between sleep and habitual sugar intake found that
reduced sleep duration was associated with higher sugar intake (Hjorth et al., 2014;
Kjeldsen et al., 2014), and that timing of total sugar intake differed with sleep duration,
suggesting a negative influence of total sugar on sleep in children (Hunsberger et al.,
2015). Our results did not identify any significant relationship between sleep quality
variables and total sugar intake. However, it is important to note that this current study
is the first to look at sleep quality domains as opposed to just sleep duration.
169 | P a g e
Further research would benefit from separating added sugar from natural sugar and
determining the amounts of each that are being consumed, as the recent WHO
guidelines recommend no more than 10% of added sugar should be part of a healthy
diet (World Health Organization, 2015). We were unable to differentiate between total
sugar intake and added sugar intake, but the study did look at associations between
non-core foods containing a large amount of added sugar (confectionary and sweet
drinks) as a percent of energy and behavioural and sleep variables and found no
association. This result contrasts with other studies which have looked at sugar
sweetened beverages and found an association with reduced sleep duration (Al-Haifi et
al., 2016; Franckle et al., 2015; Hjorth et al., 2014; Kjeldsen et al., 2014; Moreira et al.,
2010). The discrepancy in findings could be due to the present study looking at
confectionary and sweet drink as a percent of energy rather than sugar amount or
product volume, or due to the current study utilising both validated sleep and sugar
measures. Furthermore, there are many other factors that could influence sleep, for
example, caffeine has also been associated with sleep changes (Mindell et al., 2009).
There are no directly comparable behavioural studies, but our findings are similar to
those that report sugar has no influence on behaviour (Bellisle, 2004; Milich et al., 1986).
However, it would be interesting in future analyses to see if specifically added sugar in
the diet has any relation to sleep problems and or behavioural problems in children.
This study found strong correlations between sleep variables and behavioural variables
as expected based on previous literature. Two predominant mechanisms have been
proposed to augment the relationship between reduced/disrupted sleep and behaviour
(Sadeh, 2007). Firstly, insufficient sleep prevents or reduces the brain activity required
for brain maturation, affect regulation, memory consolidation, and learning (Frank &
Benington, 2006; Peigneux, Laureys, Delbeuck, & Maquet, 2001; Stickgold, 2001; Robert
Stickgold, Hobson, Fosse, & Fosse, 2001; Tarokh, Saletin, & Carskadon, 2016; Wilhelm,
Stuiber, Ludtke, & Wilhelm, 2014). Secondly, reduced/disrupted sleep leads to
increased daytime sleepiness and reduced alertness which can compromise daytime
functioning of specific brain areas such as the prefrontal cortex (Gruber, Cassoff,
Frenette, Wiebe, & Carrier, 2012; Muzur, Pace-Schott, & Hobson, 2002). This in turn
170 | P a g e
leads to many areas of compromised cognitive function and behaviour regulation
(Sadeh, 2007).
The strength of this study is that it is the first to look at the combined associations
between habitual sleep, total dietary sugar and behaviour in children. Further strengths
include a naturalistic setting and the use of valid and reliable measures. However, there
are limitations that should be considered. One consideration is that the current study
utilizes self-report and parental report surveys, it would be ideal to obtain an objective
measure of both sugar intake and sleep to substantiate the findings. Secondly, this
study did not capture other factors such as physical activity that may influence
sleepiness and energy intake in children.
Overall, this study has shown that while a large proportion of Australian children
included in this study consume more than the recommended amounts of total sugar
daily, their total sugar intake is not strongly associated with behavioural or sleep
problems, nor does it affect the relationship between sleep and behaviour. While a diet
high in sugar poses significant health concerns (Hu & Malik, 2010; Milich et al., 1986;
Moynihan & Kelly, 2013; Vartanian et al., 2007), this study suggests that poor
behavioural and sleep outcomes are not amongst them.
171 | P a g e
Chapter 10
General Discussion
172 | P a g e
10.1 Overview
This research aimed to investigate patterns of caffeine consumption in a sample of

healthy Australian adults and children, and to assess the impact of that consumption on
sleep. To achieve this, a new and novel measure of caffeine intake was developed and
administered to both adults and children. The sample of children further completed a
broad food frequency questionnaire and a survey of daytime behaviour so further
interactions of caffeine, sugar, sleep and behaviour could be explored. Overall, the
findings of the studies in this thesis suggest that higher caffeine intake in adults and
children is associated with worsened aspects of sleep. Higher caffeine consumption also
appears to be associated with worse behaviour in children, though this was partially
mediated by sleep factors. Interestingly, the survey in children indicated that whilst
sugar intake was high, total sugar consumption was not associated with sleep or
behaviour (Table 10.1).
173 | P a g e
Table 10.1: An overview of the findings from studies one and two
Study Population Outcome Chapter Specific Findings
Reproducibility Strong reproducibility

1 Adults 6
Validity Moderate to fair validity
TIB ↓Caffeine consumption = more TIB
SOL → No association between caffeine and SOL

1 Adults 7
SE → No association between caffeine and SE
Sleep quality ↑Caffeine consumption amongst poor sleepers
TST ↑Caffeine consumption = less TST
↑Caffeine consumption = worse sleep routine,

morning tiredness and restless sleep
Sleep quality → No association between caffeine consumption
and night arousals, bedtime anxiety and sleep
disordered breathing
↑ Caffeine consumption = worse internalising

2 Children 8 behaviours
Behaviour → No association between caffeine consumption
and externalising behaviours and total behavioural
problems
Sleep mediates the

relationship between Sleep partially mediates the relationship between
caffeine and caffeine and behaviour
behaviour
→ No association between sugar intake as a

percent of energy and night arousals, bedtime
Sleep quality
anxiety, sleep disordered breathing, sleep routine,
morning tiredness and restless sleep
→ No association between sugar intake as a

2 Children Behaviour 9 percent of energy and internalising, externalising,
and total behaviours
Sugar strengthened
the relationship Sugar did not impact the relationship between
between sleep and sleep and behaviour
behavioural problems
Abbreviations: →, no association; ↓, increase; ↑, decrease; SE, sleep efficiency; SOL, sleep onset latency;
TIB, time in bed; TST, total sleep time.
174 | P a g e
10.2 Caffeine food frequency questionnaire
The development of an Australian-based caffeine food frequency questionnaire (C-FFQ),

which captures the average amount of caffeine consumed over a week from different
sources such as coffee and tea (hot and cold), soft drinks (soda), chocolate (food and
beverage), and energy drinks, is a novel contribution to the field. This is an important
development for research given the emerging reports of the impact of caffeine on daily
function (Childs & de Wit, 2006; Clark & Landolt, 2016; Smith, 2002), the increasing
number of caffeine products available, and the increased accessibility of caffeine-
containing products to children, a group in which the impact of caffeine is relatively
unknown. Caffeine content in food and beverage products varies considerably across
nations (for example, Sunkist in the United States of America [USA] has approximately
11mg of caffeine per 100ml [Dr Pepper Snapple Group, 2016] compared to in Australia
where it has 0mg), and as such there is a need for the development of reporting tools
specific to each country. The benefits of using a food frequency questionnaire approach
is that it can quickly gather valid data on a large group of participants at a lower cost
compared with food diaries and 24-hour recall (Erdman Jr et al., 2012). There have been
other caffeine-specific food frequency questionnaires designed in Europe and USA
(Boylan et al., 2008; Bühler et al., 2013; Ferraroni et al., 2004; Rudolph et al., 2012;
Schliep et al., 2013), however only two of these (Boylan et al., 2008; Rudolph et al.,
2012) considered all sources of possible caffeine consumption, like the questionnaire
developed for this study.
Intakes in children and adults

In adults, the C-FFQ revealed a wide range of daily caffeine intakes (41.6-726.6mg per
day) and further showed that the primary sources of this consumption were from coffee
and tea, with smaller amounts from soft drink, energy drinks and chocolate. In Europe
and North America, national recommendations state that less than 400mg of caffeine
should be consumed per day to avoid adverse effects (European Food Safety Authority,
2015; Food and Drug Administration, 2013; Health Canada, 2012). On average the
Australian adult sample included in this thesis appears to be within the healthy range of
consumption; however, 2% of adults in this sample reported consuming above the
recommended daily limit. This finding suggests that in a proportion of adults, caffeine
175 | P a g e
intake may be a contributing factor for adverse effects, such as cardiovascular effects,
effects on bone status and calcium balance, changes in adult behaviour, increased
incidence of cancer or effects on male fertility (Nawrot et al., 2003).
The current thesis showed that, of the children who participated, the majority reported
consuming some caffeine on a weekly basis, with caffeine intake ranging from 0-
150.8mg per day (approximately 13% reported no caffeine intake). The major caffeine
sources predominantly featured were coffee and tea, chocolate, and soft drinks. Of
note, energy drink consumption was not a major contributing factor to caffeine intake in
children, with only 3% of children reporting to consume these beverages. In Canada, the
national recommendations suggest that a maximum daily caffeine intake of 2.5mg per
kilogram of body weight should be consumed in children (Health Canada, 2012). Thus
the daily average and range of caffeine consumption in the majority of Australian
children included in this study appears relatively healthy.
10.3 Impact of caffeine
Sleep in adults and children

In adults, increased caffeine consumption was associated with reduced time in bed (TIB)
and greater caffeine was consumed in poor sleepers, as determined by validated
questionnaires. This result is similar to un-validated measures of this association (Baker
et al., 2009; Hollander et al., 2001; Huntley & Juliano, 2012; Lohsoonthorn et al., 2013;
Mniszek, 1988) where increased caffeine consumption was associated with poorer sleep
quality. Unlike some previous reports, this study did not find any association between
caffeine consumption and either sleep efficiency or sleep onset latency. This may be
related to the greater precision of measure used in this study, but could also be due to
differences in the time of day in which caffeine is consumed. Limitations of the studies
are that the timing of caffeine consumption was not reported and the decision
processes adults may have used to determine timing of caffeine consumption was not
captured. Nonetheless, the results presented in this thesis provide sufficient
justification to extend this work and consider changes in the associations with sleep
factors by consumption at different times of the day or hours from attempts to sleep.
176 | P a g e
Of greater novelty to the field, the study also showed that increased caffeine
consumption was associated with reduced total sleep time (TST) and poor sleep quality
in children, specifically in regards to sleep routine, morning tiredness, and restless sleep.
This result is important, given the apparently healthy levels of caffeine consumption in
this Australian sample. The relationship with reduced TST is consistent with a limited
number of prior studies (Calamaro et al., 2012; Li et al., 2010; Mindell et al., 2009),
however previously there has been a lack of consistency amongst studies as to the
aspects of sleep impacted by caffeine in children (Kristjansson et al., 2014; Warzak et al.,
2011; Quach et al., 2012; Li et al., 2014). The use of valid and broader ranging
assessment of sleep problems has helped clarify the association between caffeine and
sleep in this younger age group.
Behaviour in children
Consistent with previous reports, the current study shows increased caffeine is
associated with more problematic behaviour in children. The current study found that
increased caffeine is associated with increased internalising behaviours such as
depression and anxiety, which is consistent with previous findings (Azagba et al., 2014;
Fulkerson et al., 2004; Guilleminault et al., 1976; Richards & Smith, 2015). For the first
time, the association between caffeine consumption and poor internalising behaviour in
children was shown to be partially mediated by the impact of caffeine on sleep. The
direct impact of sleep on behaviour in children is well-documented (Beebe, 2011;
Blunden & Beebe, 2006), and the impact of caffeine on behaviour, shown to be in part
due the disturbance on sleep, is of particular relevance to clinical assessment and
education. It follows that concern over a child’s behaviour warrants the combined
consideration of both caffeine intake and sleep health. Similarly, advice on caffeine
allowances for young children also needs to consider the impact on sleep due to the
subsequent effect this is likely to have on child behaviour.
Despite the overall amounts of daily caffeine consumption being relatively low in
children, an impact on daytime and nocturnal function is evident from the results. The
high proportion of children consuming caffeine is significant, and reasons for the
prevalent consumption should be considered by future work. One possibility could be a
lack of labelling on products that contain caffeine, such as chocolate and coffee
177 | P a g e
flavoured drinks. Irrespective, the results presented in this thesis support an argument
for clear product labelling for all caffeinated products and particularly those marketed to
children. To further support such endeavours, national caffeine guidelines should be
developed which are likely to help families and the general public understand how much
caffeine is unhealthy, and the impact that caffeine can have on both adults and children.
10.4 Relationship between caffeine and sugar
There is a growing concern over the impact of an increasingly sugar-fuelled diet on

health and wellbeing with much of the focus to date surrounding obesity (Malik et al.,
2013) and dental disease (Burt & Pai, 2001). Products that are high in caffeine tend to
also be high in sugar (Food Standards Australia New Zealand, 2015). This is further
supported by Figure 10.1 which shows the relationship between caffeine and sugar
consumption amongst children in this study. Initial analyses indicated a correlation
coefficient of r=0.236 (p<0.001), which remained relatively unchanged after removal of a
single outlier (caffeine consumption: 150.8mg per day, sugar consumption: 349.14g per
day) (r=0.228, p<0.001; Figure 10.1). This suggests that, in addition to caffeine, sugar
consumption is another important factor to consider in relation to child sleep and
behaviour.
Figure 10.1: A scatterplot showing the relationship between caffeine consumption (mg) and sugar
consumption (g), r=0.228, p=0.001
178 | P a g e
10.5 Sugar intake in children
Excess dietary sugar in children can be harmful, and has been shown to increase the risk
of obesity (Chen et al., 2008; Malik et al., 2013; Van Cauter & Knutson, 2008; Vartanian
et al., 2007), and dental caries (Moynihan & Kelly, 2013). Furthermore, in adulthood,
increased sugar has been shown to be related to type II diabetes and metabolic
syndrome (Hu & Malik, 2010; Malik et al., 2010). Therefore, the reduction of sugar
intake in Australian children over the last twenty years is encouraging; however, sugar
intake is still greater than international recommendations for children (Binns et al.,
2003). The current results found that average sugar intake for all children represented
26.0% ±7.0% of their total energy intake on average, and 80% of Australian children are
still consuming above the recommended daily sugar intake. An important point when
interpreting these findings is that the current results do not differentiate between added
sugar and natural sugar, and thus do not allow a careful comparison against the WHO
guidelines (World Health Organisation, 2015). Irrespective, it is clear that further work is
required to reduce sugar consumption in Australian children.
10.6 Impact of sugar
Sleep in children
Despite the excessive sugar consumption, the current study found no direct association
between total sugar intake and sleep variables including TST and sleep quality in
children. This is in contrast to previous studies which reported higher added sugar was
associated with lower TST (Hjorth et al., 2014; Kjeldsen et al., 2014); and more-recently
Hunsberger et al. (2015) found that timing of total sugar intake made a difference to
TST, specifically, short sleep (less than 10 hours) was associated with less sugar intake at
midday but higher intakes in the evening. A limitation of the current study is that timing
of food and beverage intake was not reported (which is typical of food frequency
questionnaires generally). An investigation of variations in sugar intake across the day in
relation to sleep would be a natural extension to the current study. Future work could
also consider the impact of sugar consumption in children with sleep problems, who
may be more vulnerable to subtle effects of sugar on sleep processes. Furthermore, the
effects of other additives in foods that are high in sugar such as colours and stimulants
179 | P a g e
could also be considered. Irrespective of these limitations, this was the first study to
look at sleep quality variables and total sugar intake in children with no association
found, indicating that in healthy children habitual sugar intake may not impact sleep
significantly.
Behaviour in children
Previously, there has been a paucity of research regarding total overall sugar intake and
behaviour in children, largely due to the difficulty in accessing sugar amounts in
products and measuring sugar intake. Previous studies have mostly measured sugar
intake experimentally and investigated the acute effects of a large dose of sugar, with
the results showing no change in behaviour in children. This novel cross-sectional study
gathered total sugar intake in grams and found that habitual sugar is not associated with
behaviour. This finding supports a previous cross-sectional study by Wolraich et al.
(1986) who used educated estimates to determine sugar intake in food products and
found no difference in sugar intake between hyperactive and non-hyperactive children.
The current study also supports previous experimental work, which showed that sugar
did not impact behaviours in children (Bellisle, 2004; Milich et al., 1986). Overall, it is
acknowledged that there are numerous health implications associated with high sugar
intake (Chen et al., 2008; Malik et al., 2013; Moynihan & Kelly, 2013; Van Cauter &
Knutson, 2008; Vartanian et al., 2007). However, the present findings suggest that
habitual sugar intake may not be a primary concern for problematic sleep and behaviour
in healthy children contrary to popular self-help books and a variety of media (Appleton,
1985; Appleton & Jacobs, 2009; Maxted, 2015; Rufus, 2005). Perhaps this stigma needs
to be acknowledged in medical and psychological communities as we strive for an
empirically informed health management system.
10.7 Future research
Results from the studies reported in the current thesis characterise the role of caffeine
and sugar in the relationship between and sleep and behaviour in healthy children and
the effect of caffeine on sleep for adults. A number of limitations and questions for
future studies are also raised, which will further improve our understanding of the
relationship between diet, sleep, and behaviour.
180 | P a g e
Firstly, although development of the C-FFQ was successful at gaining data on habitual
caffeine intake in both children and adults, further adaptions could be made to provide a
more detailed understanding of the relationship with both sleep and behaviour. For
example, how children receive caffeinated food and beverages will assist in targeted
education, i.e. do they purchase them with their own money, or do their parents provide
them directly? Furthermore, understanding the motivations for consumption, while
possibly beyond the scope of a food frequency questionnaire adaption, will also help
target education and marketing recommendations. For example, is it the marketing
appeal, product taste, or (as is likely in adults) is it due to the stimulant properties, and
are these properties explicitly known or only implicitly experienced? In addition, as
caffeine has a half-life of four to five hours (Kaplan et al., 1997) the timing of caffeine
intake is important when considering impact on sleep. The Sleep Health Foundation
recommends, in children, having as little of caffeine as possible, and even then, not
consuming caffeine after lunch time (Sleep Health Foundation, 2011) and in adults to
avoid caffeine for at least four hours before bed (Sleep Health Foundation, 2013).
Intuitively, caffeine intake in the morning is likely to have much less impact on sleep
compared with caffeine consumed in the mid- to late-afternoon in children and adults.
Secondly, broadly for all self-report approaches there is the risk of bias in both parent
and child responses. For example, public information about the importance of sleep for
wellbeing or the dangers of excessive sugar consumption can lead to social desirability
bias in responses (Herbert, 1995). While diet can be particularly difficult to observe
objectively in an economical manner for larger samples, the use of motion tracking
devices such as actigraphy to quantify sleep objectively are feasible with adequate
funding. Such monitoring can also provide a richer data, providing detailed assessment
of sleep quality as well as sleep timing and duration (de Souza et al., 2003; Marino et al.,
2013).
Thirdly, many of the products containing caffeine and/or high sugar often also contain a
range of additive and other ingredients that may independently influence sleep or
behaviour (Grandner et al., 2014; Rowe, 1988). Future laboratory studies need to be
conducted to separate such components that exist in energy drinks, soft drinks, and
chocolate, that is not caffeine and sugar, but may play a role on sleep and behaviour
181 | P a g e
disruption in children. Laboratory studies allow for cause and effect relationships to be
observed and can utilise more pure forms of caffeine and sugar and more objective
measures of sleep and behaviour. While sugar does not appear to impact sleep and
behaviour in a cross-sectional field environment, research from a laboratory study has
never been published. It might be of interest to determine if changes in sugar intake
impact sleep architecture, as sleep stages have been associated with different cognitive
functions such as memory (Genzel, Kroes, Dresler, & Battaglia, 2014). Furthermore,
laboratory studies allow a precise and objective measure of timing to accurately
determine impact of timing of caffeine and sugar consumption on sleep. Timing of sugar
intake may be more influential on sleep than total sugar consumption, as Hunsberger et
al. (2015) showed an association between later sugar consumption and shorter TST.
Fourthly, repeating study two but including a sample of children with sleep and
behavioural difficulties could identify existing relationships between sugar and caffeine
with sleep or behaviour that are not otherwise evident in the healthy population. While
the intentions of this thesis were to determine effects of dietary caffeine and sugar on a
healthy sample of children, and in doing so establish a typical pattern of association, it is
still necessary to understand the impact sugar and caffeine may have on children who
are at increased risk of disruption to sleep or daytime behaviour. Previous research in a
sample of children with attention deficit hyperactivity disorder (ADHD) showed
increased sugar intake was associated with greater sleep disruption (Blunden et al.,
2011). While these were preliminary findings, it does suggest that, compared to healthy
children, children with behavioural problems may be have a greater predisposition for
dietary factors to influence the relationship between sleep and behaviour.
Finally, the current results do not make the distinction between added and natural sugar
consumption. At the time of the studies reported in this thesis there was no facility to
identify and separate added sugars from natural sugars in Australian foods and
beverages, and consequently total sugar content was used in analyses. As shown
throughout this thesis, there is a large variation in results regarding sugar consumption
and its impact on sleep. To help understand previous research separating added sugars
(those that are processed and added to non-core foods such as confectionary and sweet
drinks) from natural sugars (those that are not processed and in core foods such as milk
182 | P a g e
and fruit) will help characterise the effects of different sugars on the human body.
Determining the different impacts different sugars may have on sleep variables such as
sleep quality and TST may show further support for healthy eating.
10.8 Concluding remarks
This thesis describes the design and validation of a caffeine food frequency
questionnaire that can quickly capture caffeine intake in Australian diets. Using this C-
FFQ it was shown that caffeine negatively impacts sleep in both children and adults and
behaviour in children, however sleep may partially mediate the association between
caffeine and behaviour. Further results from the studies showed that while sugar intake
has many negative health concerns it potentially does not impact sleep and behaviour in
children. Everyone needs to both sleep and eat, therefore it is important that we
understand the relationship these two behaviours may have on each other to ensure
optimal health and wellbeing.
183 | P a g e
References
Achenbach, T. M., & Edelbrock, C. S. (1978). The classification of child psychopathology:
a review and analysis of empirical efforts. Psychological Bulletin, 85, 1275.
Achenbach, T. M., & Rescorla, L. (2001). ASEBA school-age forms & profiles: Aseba
Burlington.
Achermann, P. (2004). The two-process model of sleep regulation revisited. Aviation,
Space, and Environmental Medicine, 75(Suppl 1), A37-A43.
Aepli, A., Kurth, S., Tesler, N., Jenni, O. G., & Huber, R. (2015). Caffeine consuming
children and adolescents show altered sleep behavior and deep sleep. Brain
Sciences, 5(4), 441-455.
Åkerstedt, T., & Nilsson, P. (2003). Sleep as restitution: an introduction. Journal of
Internal Medicine, 254(1), 6-12.
Al-Disi, D., Al-Daghri, N., Khanam, L., Al-Othman, A., Al-Saif, M., Sabico, S., & Chrousos,
G. (2010). Subjective sleep duration and quality influence diet composition and
circulating adipocytokines and ghrelin levels in teen-age girls. Endocrine Journal,
57(10), 915-923.
Al-Haifi, A. A., AlMajed, H. T., Al-Hazzaa, H. M., Musaiger, A. O., Arab, M. A., & Hasan, R.
A. (2016). Relative Contribution of Obesity, Sedentary Behaviors and Dietary
Habits to Sleep Duration Among Kuwaiti Adolescents. Global Journal of Health
Science, 8(1), 107-117. doi:10.5539/gjhs.v8n1p107
Al-Hazzaa, H. (2013). Physical activity and lifestyle habits of Arab adolescents:
Preliminary findings from the Arab teens lifestyle study. Annals of Nutrition and
Metabolism, 63, 52-53. doi:http://dx.doi.org/10.1159/000354245
Al-Hazzaa, H., Musaiger, A., Abahussain, N., Al-Sobayel, H., & Qahwaji, D. (2014).
Lifestyle correlates of self-reported sleep duration among Saudi adolescents: A
multicentre school-based cross-sectional study. Child: Care, Health and
Development, 40(4), 533-542. doi:10.1111/cch.12051
Alfano, C., Ginsburg, G., & Kingery, J. (2007). Sleep-related problems among children and
adolescents with anxiety disorders. Journal of the American Academy of Child &
Adolescent Psychiatry, 46(2), 224-232.
Alfano, C., Zakem, A., Costa, N., Taylor, L., & Weems, C. (2009). Sleep problems and their
relation to cognitive factors, anxiety, and depressive symptoms in children and
adolescents. Depression and Anxiety, 26(6), 503-512.
Ali, N., Pitson, D., & Stradling, J. (1993). Snoring, sleep disturbance, and behaviour in 4-5
year olds. Archives of Disease in Childhood, 68(3), 360.
American Psychiatric Association. (2013). DSM 5: American Psychiatric Association.
American Sleep Disorders Association, Diagnostic Classification Steering Committee,
Thorpy, M. J., & Diagnostic Classification Steering Committee. (1990). The
international classification of sleep disorders: diagnostic and coding manual:
American Sleep Disorders Association.
Angus, R. G., Heslegrave, R. J., & Myles, W. S. (1985). Effects of prolonged sleep
deprivation, with and without chronic physical exercise, on mood and
performance. Psychophysiology, 22(3), 276-282.
Appleton, N. (1985). Lick the sugar habit: Penguin.
Appleton, N., & Jacobs, G. (2009). Suicide by Sugar: A Startling Look at Our# 1 National
Addiction: Square One Publishers.
184 | P a g e
Armitage, R., Trivedi, M., Hoffmann, R., & Rush, A. J. (1997). Relationship between
objective and subjective sleep measures in depressed patients and healthy
controls. Depression and Anxiety, 5(2), 97-102.
Arnaud, M. (1999). Caffeine/chemistry and physiological effects. Encyclopedia of Human
Nutrition, 206-214.
Astill, R. G., Verhoeven, D., Vijzelaar, R. L., & Someren, E. J. (2013). Chronic stress
undermines the compensatory sleep efficiency increase in response to sleep
restriction in adolescents. Journal of Sleep Research, 22(4), 373-379.
doi:10.1111/jsr.12032
Attwood, A., Higgs, S., & Terry, P. (2007). Differential responsiveness to caffeine and
perceived effects of caffeine in moderate and high regular caffeine consumers.
Psychopharmacology, 190(4), 469-477.
Australian Bureau Of Statistics. (2006). Socio-Economic indices for Areas (SEIFA)-
Technical Paper. Australian Bureau of Statistics. Canberra.
Australian Bureau Of Statistics. (2011). Postal area (POA) index of relative socio-
economic advantage and disadvantage, 2011. Canberra: Australian Bureau of
Statistics.
Australian Bureau of Statistics. (2014a, 15 October 2015). Australian Health Survey:
Nutrition First Results - Foods and Nutrients, 2011-12 Proportion of Caffeine
from food groups
Australian Bureau of Statistics. (2014b, 15 October 2015). Australian Health Survey:
Nutrition First Results - Foods and Nutrients, 2011-12 Foods Consumed: Non-
alcoholic beverages. Retrieved from
http://www.abs.gov.au/ausstats/abs@.nsf/Lookup/by%20Subject/4364.0.55.007
~2011-12~Main%20Features~Non-alcoholic%20beverages~701
Australian Bureau of Statistics. (2014c). Nutrition First Results – Food and Nutrients,
2011-12. (4364.0.55.007). Canberra.
Australian Bureau of Statistics. (2015a). Australian Health Survey: Usual Nutrient Intakes,
2011-12 Table 4: Essential minerals (and caffeine) Canberra.
Australian Bureau of Statistics. (2015b). National Health Survey: Mental health and co-
existing physical health conditions, 2014-15 — Australia. data cube: SuperTABLE,
cat. no. 4329.0.00.004. Retrieved from
http://www.abs.gov.au/AUSSTATS/abs@.nsf/DetailsPage/4329.0.00.0042014%2
0-%2015?OpenDocument
Awad, K. M., Drescher, A. A., Malhotra, A., & Quan, S. F. (2013). Effects of exercise and
nutritional intake on sleep architecture in adolescents. Sleep and Breathing,
17(1), 117-124. doi:10.1007/s11325-012-0658-6
Ayas, N. T., White, D. P., Manson, J. E., Stampfer, M. J., Speizer, F. E., Malhotra, A., & Hu,
F. B. (2003). A prospective study of sleep duration and coronary heart disease in
women. Archives of Internal Medicine, 163(2), 205-209.
Azagba, S., Langille, D., & Asbridge, M. (2014). An emerging adolescent health risk:
caffeinated energy drink consumption patterns among high school students.
Preventive Medicine, 62, 54-59.
doi:http://dx.doi.org/10.1016/j.ypmed.2014.01.019
Backhaus, J., Junghanns, K., Broocks, A., Riemann, D., & Hohagen, F. (2002). Test–retest
reliability and validity of the Pittsburgh Sleep Quality Index in primary insomnia.
Journal of Psychosomatic Research, 53(3), 737-740.
185 | P a g e
Baker, F. C., Wolfson, A. R., & Lee, K. A. (2009). Association of sociodemographic,
lifestyle, and health factors with sleep quality and daytime sleepiness in women:
Findings from the 2007 National Sleep Foundation "Sleep in America Poll.".
Journal of Women's Health, 18(6), 841-849. doi:10.1089/jwh.2008.0986
Baker, F. C., Maloney, S., & Driver, H. S. (1999). A comparison of subjective estimates of sleep with objective
polysomnographic data in healthy men and women. Journal of Psychosomatic Research, 47(4), 335-
341.
Balkin, T. J., Bliese, P. D., Belenky, G., Sing, H., Thorne, D. R., Thomas, M., . . . Wesensten,
N. J. (2004). Comparative utility of instruments for monitoring sleepiness‐related
performance decrements in the operational environment. Journal of Sleep
Research, 13(3), 219-227.
Barone, J., & Roberts, H. (1996). Caffeine consumption. Food and chemical toxicology,
34(1), 119-129.
Barrett-Connor, E. (1991). Nutrition epidemiology: how do we know what they ate? The
American Journal of Clinical Nutrition, 54(1), 182S-187S.
Basheer, R., Strecker, R. E., Thakkar, M. M., & McCarley, R. W. (2004). Adenosine and
sleep–wake regulation. Progress in Neurobiology, 73(6), 379-396.
Bates, J., Viken, R., Alexander, D., Beyers, J., & Stockton, L. (2002). Sleep and adjustment
in preschool children: Sleep diary reports by mothers relate to behavior reports
by teachers. Child Development, 73(1), 62-75.
Bathalon, G. P., Tucker, K. L., Hays, N. P., Vinken, A. G., Greenberg, A. S., McCrory, M. A.,
& Roberts, S. B. (2000). Psychological measures of eating behavior and the
accuracy of 3 common dietary assessment methods in healthy postmenopausal
women. The American Journal of Clinical Nutrition, 71(3), 739-745.
Beaton, G. H., Milner, J., Corey, P., McGuire, V., Cousins, M., Stewart, E., . . . Kassim, N.
(1979). Sources of variance in 24-hour dietary recall data: implications for
nutrition study design and interpretation. American Journal of Clinical Nutrition
(USA).
Beebe, D. W. (2011). Cognitive, behavioral, and functional consequences of inadequate
sleep in children and adolescents. Pediatric Clinics of North America, 58(3), 649-
665.
Beebe, D. W., Simon, S., Summer, S., Hemmer, S., Strotman, D., & Dolan, L. M. (2013).
Dietary intake following experimentally restricted sleep in adolescents. SLEEP,
36(6), 827-834.
Belenky, G., Wesensten, N. J., Thorne, D. R., Thomas, M. L., Sing, H. C., Redmond, D. P., .
. . Balkin, T. J. (2003). Patterns of performance degradation and restoration
during sleep restriction and subsequent recovery: A sleep dose‐response study.
Journal of Sleep Research, 12(1), 1-12.
Bellisle, F. (2004). Effects of diet on behaviour and cognition in children. British Journal
of Nutrition, 92(Suppl.2), S227-S232.
Bernstein, G. A., Carroll, M. E., Thuras, P. D., Cosgrove, K. P., & Roth, M. E. (2002). Caffeine dependence in
teenagers. Drug and Alcohol Dependence, 66(1), 1-6.
Biggs, S., Kennedy, J., Martin, A., van den Heuvel, C., & Lushington, K. (2012).
Psychometric properties of an omnibus sleep problems questionnaire for school-
aged children. Sleep Medicine, 13, 390-395.
Biggs, S. N., & Dollman, J. (2007). Association between sleep, BMI and waist girth in
children and adolescents: a retrospective analysis. Acta Paediatrica, 96(12),
1839-1840.
186 | P a g e
Bin, Y. S., Marshall, N. S., & Glozier, N. (2012). Secular trends in adult sleep duration: a
systematic review. Sleep Medicine Reviews, 16(3), 223-230.
Bin, Y. S., Marshall, N. S., & Glozier, N. (2013). Sleeping at the limits: the changing
prevalence of short and long sleep durations in 10 countries. American Journal of
Epidemiology, 308.
Binns, C., Davidson, G., & Forbes, D. (2003). Dietary guidelines for children and
adolescents in Australia. Canberra, Australia, National Health and Medical
Research Council, 21-43.
Bjorness, T. E., & Greene, R. W. (2009). Adenosine and sleep. Current
Neuropharmacology, 7(3), 238-245.
Black, A. E., Prentice, A. M., Goldberg, G. R., Jebb, S. A., Bingham, S. A., Livingstone, M. B.
E., & Coward, A. (1993). Measurements of total energy expenditure provide
insights into the validity of dietary measurements of energy intake. Journal of the
American Dietetic Association, 93(5), 572-579.
Blagrove, M., & Akehurst, L. (2001). Personality and the modulation of effects of sleep
loss on mood and cognition. Personality and Individual Differences, 30(5), 819-
828.
Bland, J. M., & Altman, D. G. (2010). Statistical methods for assessing agreement
between two methods of clinical measurement. International Journal of Nursing
Studies, 47(8), 931-936.
Blunden, S., & Beebe, D. (2006). The contribution of intermittent hypoxia, sleep debt
and sleep disruption to daytime performance deficits in children: consideration
of respiratory and non-respiratory sleep disorders. Sleep Medicine Reviews,
10(2), 109-118.
Blunden, S., Lushington, K., Kennedy, D., Martin, J., & Dawson, D. (2000). Behavior and
neurocognitive performance in children aged 5-10 years who snore compared to
controls. Journal of Clinical and Experimental Neuropsychology, 22(5), 554-568.
Blunden, S., Milte, C. M., & Sinn, N. (2011). Diet and sleep in children with attention
deficit hyperactivity disorder: Preliminary data in Australian children. Journal of
Child Health Care, 15(1), 14-24.
Boergers, J., Gable, C. J., & Owens, J. A. (2014). Later school start time Is associated with
improved sleep and daytime functioning in adolescents. Journal of
Developmental and Behavioral Pediatrics, 35(1), 11-17.
doi:http://dx.doi.org/10.1097/DBP.0000000000000018
Borbély, A., & Achermann, P. (1999). Sleep homeostasis and models of sleep regulation.
Journal of Biological Rhythms, 14(6), 557-568.
Borbély, A. A. (1982). A two process model of sleep regulation. Human Neurobiology,
1(3), 195-204.
Borbély, A. A., Daan, S., Wirz‐Justice, A., & Deboer, T. (2016). The two‐process model of
sleep regulation: a reappraisal. Journal of Sleep Research, 25(2),
Born, J., & Wilhelm, I. (2012). System consolidation of memory during sleep.
Psychological Research, 76(2), 192-203.
Bostick, R. M., Potter, J. D., Kushi, L. H., Sellers, T. A., Steinmetz, K. A., McKenzie, D. R., . .
. Folsom, A. R. (1994). Sugar, meat, and fat intake, and non-dietary risk factors
for colon cancer incidence in Iowa women (United States). Cancer Causes and
Control, 5(1), 38-52.
187 | P a g e
Boylan, S. M., Cade, J. E., Kirk, S. F., Greenwood, D. C., White, K. L., Shires, S., . . . Hay, A.
W. (2008). Assessing caffeine exposure in pregnant women. British Journal of
Nutrition, 100(04), 875-882.
Bramstedt, K. A. (2007). Caffeine use by children: the quest for enhancement. Substance
Use and Misuse, 42, 1237-1251.
Brendel, D. H., Reynolds, C., Jennings, J., Hoch, C., Monk, T., Berman, S., . . . Kupfer, D.
(1990). Sleep stage physiology, mood, and vigilance responses to total sleep
deprivation in healthy 80‐year‐olds and 20‐year‐olds. Psychophysiology, 27(6),
677-685.
Brice, C., & Smith, A. (2001). The effects of caffeine on simulated driving, subjective
alertness and sustained attention. Human Psychopharmacology: Clinical and
Experimental, 16(7), 523-531.
Brice, C. F., & Smith, A. P. (2002). Effects of caffeine on mood and performance: a study
of realistic consumption. Psychopharmacology, 164(2), 188-192.
Brick, C. A., Seely, D. L., & Palermo, T. M. (2010). Association between sleep hygiene and
sleep quality in medical students. Behavioral Sleep Medicine, 8(2), 113-121.
doi:http://dx.doi.org/10.1080/15402001003622925
Bruni, O., Ottaviano, S., Guidetti, V., Romoli, M., Innocenzi, M., Cortesi, F., & Giannotti,
F. (1997). The Sleep Disturbance Scale for Children (SDSC). Construction and
validation of an instrument to evaluate sleep disturbances in childhood and
adolescence. Journal of Sleep Research, 5(4), 251-261.
Brunner, E., Juneja, M., & Marmot, M. (2001). Dietary assessment in Whitehall II:
comparison of 7 d diet diary and food-frequency questionnaire and validity
against biomarkers. British Journal of Nutrition, 86(3), 405-414.
Bühler, E., Lachenmeier, D., Schle-gel, K., & Winkler, G. (2013). Development of a tool to
assess the caf-feine intake among teenagers and young adults. Ernahrungs
Umschau, 61(4), 58-63.
Burrell, S. (2016). New ABS data reveals how much sugar Australians really consume.
news.com.au. Retrieved from
http://www.news.com.au/lifestyle/health/diet/new-abs-data-reveals-how-
much-sugar-australians-really-consume/news-
story/979263910569a4c55bb0051551bdce1a
Burrows, T., Truby, H., Morgan, P., Callister, R., Davies, P., & Collins, C. E. (2013). A
comparison and validation of child versus parent reporting of children's energy
intake using food frequency questionnaires versus food records: Who's an
accurate reporter? Clinical Nutrition, 32(4), 613-618.
Burt, B. A., & Pai, S. (2001). Sugar consumption and caries risk: a systematic review.
Journal of Dental Education, 65(10), 1017-1023.
Buysse, D. J., Hall, M. L., Strollo, P. J., Kamarck, T. W., Owens, J., Lee, L.,.Matthews, K. A.
(2008). Relationships between the Pittsburgh Sleep Quality Index (PSQI), Epworth
Sleepiness Scale (ESS), and clinical/polysomnographic measures in a community
sample. Journal of Clinical Sleep Medicine, 4(6), 563-571.
Buysse, D. J., Reynolds, C. F., Monk, T. H., Berman, S. R., & Kupfer, D. J. (1989). The
Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and
research. Psychiatry Research, 28(2), 193-213.
188 | P a g e
Buysse, D. J., Reynolds, C. F., Monk, T. H., & Hoch, C. C. (1991). Quantification of
subjective sleep quality in healthy elderly men and women using the Pittsburgh
Sleep Quality Index (PSQI). SLEEP, 14(4), 331-338.
Cade, J., Burley, V., Warm, D., Thompson, R., & Margetts, B. (2004). Food-frequency
questionnaires: a review of their design, validation and utilisation. Nutrition
Research Reviews, 17(1), 5-22.
Calamaro, C. J., Mason, T. B., & Ratcliffe, S. J. (2009). Adolescents living the 24/7
lifestyle: Effects of caffeine and technology on sleep duration and daytime
functioning. Pediatrics, 123(6), 1005-1010.
doi:http://dx.doi.org/10.1542/peds.2008-3641
Calamaro, C. J., Yang, K., Ratcliffe, S., & Chasens, E. R. (2012). Wired at a young age: The
effect of caffeine and technology on sleep duration and body mass index in
school-aged children. Journal of Pediatric Health Care, 26, 276-282.
doi:http://dx.doi.org/10.1016/j.pedhc.2010.12.002
Calhoun, S. L., Vgontzas, A. N., Fernandez-Mendoza, J., Mayes, S. D., Tsaoussoglou, M.,
Basta, M., & Bixler, E. O. (2011). Prevalence and risk factors of excessive daytime
sleepiness in a community sample of young children: the role of obesity, asthma,
anxiety/depression, and sleep. SLEEP, 34(4), 503-507.
Campbell, S. S., & Murphy, P. J. (2007). Delayed sleep phase disorder in temporal
isolation. SLEEP, 30(9), 1225.
Cappuccio, F. P., D'Elia, L., Strazzullo, P., & Miller, M. A. (2010). Quantity and Quality of
Sleep and Incidence of Type 2 Diabetes A systematic review and meta-analysis.
Diabetes Care, 33(2), 414-420.
Cappuccio, F. P., Taggart, F. M., Kandala, N., Currie, A., Peile, E., Stranges, S., & Miller, M.
A. (2008). Meta-analysis of short sleep duration and obesity in children and
adults. SLEEP, 31(5), 619.
Carpenter, J. S., & Andrykowski, M. A. (1998). Psychometric evaluation of the Pittsburgh
sleep quality index. Journal of Psychosomatic Research, 45(1), 5-13.
Carrier, J., Fernandez-Bolanos, M., Robillard, R., Dumont, M., Paquet, J., Selmaoui, B., &
Filipini, D. (2007). Effects of caffeine are more marked on daytime recovery sleep
than on nocturnal sleep. Neuropsychopharmacology, 32(4), 964-972.
Carskadon, M. A., & Acebo, C. (1993). A self-administered rating scale for pubertal
development. Journal of Adolescent Health, 14(3), 190-195.
Carskadon, M. A., & Dement, W. C. (2005). Normal human sleep: an overview. In M.
Kryger, T. Roth, & W. Dement (Eds.), Principles and practice of sleep medicine.
4th ed. (pp. 13-23). Philadelphia: Elsevier Saunders.
Casey, P. H., Goolsby, S. L., Lensing, S. Y., Perloff, B. P., & Bogle, M. L. (1999). The use of
telephone interview methodology to obtain 24-hour dietary recalls. Journal of
the American Dietetic Association, 99(11), 1406-1411.
Chan, E. Y., Ng, D. K., Chan, C.H., Kwok, K.l., Chow, P.Y., Cheung, J. M., & Leung, S.Y.
(2009). Modified Epworth Sleepiness Scale in Chinese children with obstructive
sleep apnea: a retrospective study. Sleep and Breathing, 13(1), 59-63.
Chee, M. W., Tan, J. C., Zheng, H., Parimal, S., Weissman, D. H., Zagorodnov, V., &
Dinges, D. F. (2008). Lapsing during sleep deprivation is associated with
distributed changes in brain activation. The Journal of Neuroscience, 28(21),
5519-5528.
189 | P a g e
Chen, M.Y., Wang, E. K., & Jeng, Y.J. (2006). Adequate sleep among adolescents is
positively associated with health status and health-related behaviors. BMC Public
Health, 6(1), 1.
Chen, X., Beydoun, M. A., & Wang, Y. (2008). Is sleep duration associated with childhood
obesity? A systematic review and meta‐analysis. Obesity, 16(2), 265-274.
Chervin, R. D., Hedger, K., Dillon, J. E., & Pituch, K. J. (2000). Pediatric sleep
questionnaire (PSQ): validity and reliability of scales for sleep-disordered
breathing, snoring, sleepiness, and behavioral problems. Sleep Medicine, 1(1),
21-32.
Childs, E., & de Wit, H. (2006). Subjective, behavioral, and physiological effects of acute
caffeine in light, nondependent caffeine users. Psychopharmacology, 185(4),
514-523.
Chinawa, J. M., Chukwu, B. F., & Obu, H. A. (2014). Sleep practices among medical
students in Pediatrics Department of University of Nigeria Teaching Hospital,
Ituku/Ozalla, Enugu, Nigeria. Nigerian Journal of Clinical Practice, 17(2), 232-236.
doi:http://dx.doi.org/10.4103/1119-3077.127565
Choo, W.-C., Lee, W.-W., Venkatraman, V., Sheu, F.-S., & Chee, M. W. (2005).
Dissociation of cortical regions modulated by both working memory load and
sleep deprivation and by sleep deprivation alone. Neuroimage, 25(2), 579-587.
Chuah, Y. L., Venkatraman, V., Dinges, D. F., & Chee, M. W. (2006). The neural basis of
interindividual variability in inhibitory efficiency after sleep deprivation. The
Journal of Neuroscience, 26(27), 7156-7162.
Clark, I., & Landolt, H. P. (2016). Coffee, caffeine, and sleep: A systematic review of
epidemiological studies and randomized controlled trials. Sleep medicine reviews.
1-9
Clark, L. A., Watson, D., & Leeka, J. (1989). Diurnal variation in the positive affects.
Motivation and Emotion, 13(3), 205-234.
Cohen, S., Doyle, W. J., Alper, C. M., Janicki-Deverts, D., & Turner, R. B. (2009). Sleep
habits and susceptibility to the common cold. Archives of Internal Medicine,
169(1), 62-67.
Cole, J. C., Motivala, S. J., Buysse, D. J., Oxman, M. N., Levin, M. J., & Irwin, M. R. (2006).
Validation of a 3-factor scoring model for the Pittsburgh sleep quality index in
older adults. SLEEP, 29(1), 112.
Collins, C. E., Boggess, M. M., Watson, J. F., Guest, M., Duncanson, K., Pezdirc, K., . . .
Burrows, T. L. (2014). Reproducibility and comparative validity of a food
frequency questionnaire for Australian adults. Clinical Nutrition, 33(5), 906-914.
Collins, L. M., Graham, J. W., Rousculp, S. S., & Hansen, W. B. (1997). Heavy caffeine use and the beginning of
the substance use onset process: An illustration of latent transition analysis. In K. J. W. Bryant,
Michael W, Stephen G (Ed.), The science of prevention: Methodological advances from alcohol and
substance abuse research (pp. 79-99). Washington DC, US: American Psychological Association.
Collison, K. S., Zaidi, M. Z., Subhani, S. N., Al-Rubeaan, K., Shoukri, M., & Al-Mohanna, F.
A. (2010). Sugar-sweetened carbonated beverage consumption correlates with
BMI, waist circumference, and poor dietary choices in school children. BMC
Public Health, 10, 234. doi:http://dx.doi.org/10.1186/1471-2458-10-234
Commonwealth Scientific and Industrial Research Organisation (Australia), Preventative
Health National Research Flagship, & The University of South Australia. (2008).
2007 Australian national children's nutrition and physical activity survey: main
findings: Department of Health and Ageing.
190 | P a g e
Conway, J. M., Ingwersen, L. A., Vinyard, B. T., & Moshfegh, A. J. (2003). Effectiveness of
the US Department of Agriculture 5-step multiple-pass method in assessing food
intake in obese and nonobese women. The American Journal of Clinical Nutrition,
77(5), 1171-1178.
Cook, T., Rutishauser, I. H., & Seelig, M. (2001). Comparable data on food and nutrient
intake and physical measurements from the 1983, 1985 and 1995 national
nutrition surveys: Health and Aged Care.
Crews, H., Olivier IV, L., & Wilson, L. (2001). Urinary biomarkers for assessing dietary
exposure to caffeine. Food Additives and Contaminants, 18(12), 1075-1087.
Currie, S. R. (2006). Sleep dysfunction. Clinician’s handbook of adult behavioral
assessment, 401-430.
Czeisler, C. A., Duffy, J. F., Shanahan, T. L., Brown, E. N., Mitchell, J. F., Rimmer, D. W., . . .
Emens, J. S. (1999). Stability, precision, and near-24-hour period of the human
circadian pacemaker. Science, 284(5423), 2177-2181.
Davis, J. M., Zhao, Z., Stock, H. S., Mehl, K. A., Buggy, J., & Hand, G. A. (2003). Central
nervous system effects of caffeine and adenosine on fatigue. American Journal of
Physiology-Regulatory, Integrative and Comparative Physiology, 284(2), 399-404.
de Mejia, E. G., & Ramirez-Mares, M. V. (2014). Impact of caffeine and coffee on our
health. Trends in Endocrinology and Metabolism, 25(10), 489-492.
de Souza, L., Benedito-Silva, A. A., Pires, M. N., Poyares, D., Tufik, S., & Calil, H. M.
(2003). Further validation of actigraphy for sleep studies. SLEEP, 26(1), 81-85.
DeFronzo, R., & Ferrannini, E. (2001). Regulation of intermediary metabolism during
fasting and feeding. Endocrinology, 4, 737-755.
Dewald, J. F., Meijer, A. M., Oort, F. J., Kerkhof, G. A., & Bögels, S. M. (2010). The
influence of sleep quality, sleep duration and sleepiness on school performance
in children and adolescents: a meta-analytic review. Sleep Medicine Reviews,
14(3), 179-189.
Ding, M., Bhupathiraju, S. N., Chen, M., van Dam, R. M., & Hu, F. B. (2014). Caffeinated
and decaffeinated coffee consumption and risk of type 2 diabetes: a systematic
review and a dose-response meta-analysis. Diabetes Care, 37(2), 569-586.
Dinges, D. F., Pack, F., Williams, K., Gillen, K. A., Powell, J. W., Ott, G. E., . . . Pack, A. I.
(1997). Cumulative sleepiness, mood disturbance and psychomotor vigilance
performance decrements during aweek of sleep restricted to 4-5 hours per night.
Sleep: Journal of Sleep Research and Sleep Medicine.
Doi, Y., Inoue, Y., Minowa, M., Uchiyama, M., & Okawa, M. (2003). Periodic leg
movements during sleep in Japanese community-dwelling adults based on the
assessments of their bed partners. Journal of Epidemiology / Japan
Epidemiological Association, 13(5), 259-265.
Dollman, J., Ridley, K., Olds, T., & Lowe, E. (2007). Trends in the duration of school‐day
sleep among 10‐to 15‐year‐old South Australians between 1985 and 2004. Acta
Paediatrica, 96(7), 1011-1014.
Donaldson, M. S. (2004). Nutrition and cancer: a review of the evidence for an anti-
cancer diet. Nutrition Journal, 3(1), 1.
Dorrian, J., Paterson, J., Dawson, D., Pincombe, J., Grech, C., & Rogers, A. E. (2011).
Sleep, stress and compensatory behaviors in Australian nurses and midwives.
Revista de Saude Publica, 45(5), 922-930.
191 | P a g e
Dr Pepper Snapple Group. (2016). Sunkist Orange Soda, Nutritional Information.
Retrieved from
http://www.dpsgproductfacts.com/en/product/SUNKIST_ORANGE_20
Drapeau, C., Hamel‐Hebert, I., Robillard, R., Selmaoui, B., Filipini, D., & Carrier, J. (2006).
Challenging sleep in aging: the effects of 200 mg of caffeine during the evening in
young and middle‐aged moderate caffeine consumers. Journal of Sleep Research,
15(2), 133-141.
Drescher, A. A., Goodwin, J. L., Silva, G. E., & Quan, S. F. (2011). Caffeine and screen time
in adolescence: associations with short sleep and obesity. Journal of Clinical Sleep
Medicine, 7(4), 337-342. doi:http://dx.doi.org/10.5664/JCSM.1182
Drummond, S. P., Brown, G. G., Stricker, J. L., Buxton, R. B., Wong, E. C., & Gillin, J. C.
(1999). Sleep deprivation‐induced reduction in cortical functional response to
serial subtraction. Neuroreport, 10(18), 3745-3748.
Dunn, S., Datta, A., Kallis, S., Law, E., Myers, C. E., & Whelan, K. (2011). Validation of a
food frequency questionnaire to measure intakes of inulin and oligofructose.
European Journal of Clinical Nutrition, 65(3), 402-408.
Echeverri, D., Montes, F. R., Cabrera, M., Galán, A., & Prieto, A. (2010). Caffeine's
vascular mechanisms of action. International journal of vascular medicine, 2010.
Eisenmann, J., Ekkekakis, P., & Holmes, M. (2006). Sleep duration and overweight among
Australian children and adolescents. Acta Paediatrica, 95(8), 956-963.
Ellenbogen, J. M., Payne, J. D., & Stickgold, R. (2006). The role of sleep in declarative
memory consolidation: passive, permissive, active or none? Current Opinion in
Neurobiology, 16(6), 716-722.
Ellis, B. W., Johns, M. W., Lancaster, R., Raptopoulos, P., Angelopoulos, N., & Priest, R. G.
(1981). The St. Mary’s Hospital sleep questionnaire: a study of reliability. SLEEP,
4(1), 93-97.
Erdman Jr, J. W., MacDonald, I. A., & Zeisel, S. H. (2012). Present knowledge in nutrition:
John Wiley & Sons.
Erkkola, M., Karppinen, M., Javanainen, J., Räsänen, L., Knip, M., & Virtanen, S. M.
(2001). Validity and reproducibility of a food frequency questionnaire for
pregnant Finnish women. American Journal of Epidemiology, 154(5), 466-476.
European Food Safety Authority. (2015). Scientific opinion on the safety of caffeine.
EFSA Journal, 13(5), 120. doi:10.2903/j.efsa.2015.4102
Evans, S. M., & Griffiths, R. R. (1992). Caffeine tolerance and choice in humans.
Psychopharmacology, 108(1-2), 51-59.
Fair, D. A., Cohen, A. L., Dosenbach, N. U., Church, J. A., Miezin, F. M., Barch, D. M., . . .
Schlaggar, B. L. (2008). The maturing architecture of the brain's default network.
Proceedings of the National Academy of Sciences, 105(10), 4028-4032.
Ferraroni, M., Tavani, A., Decarli, A., Franceschi, S., Parpinel, M., Negri, E., & La Vecchia,
C. (2004). Reproducibility and validity of coffee and tea consumption in Italy.
Ferreira, A., Clemente, V., Gozal, D., Gomes, A., Pissarra, C., Cesar, H., . . . Azevedo, M.
(2000). Snoring in Portuguese primary school children. Pediatrics, 106(5), e64.
Fiese, B., Everhart, R., & Wildenger, L. (2009). Wheezing, sleeping, and worrying: The
hidden risks of asthma and obesity in school-age children. Psychology in the
Schools, 46(8), 728-738.
192 | P a g e
Firouzi, S., Poh, B. K., Ismail, M. N., & Sadeghilar, A. (2014). Sleep habits, food intake, and
physical activity levels in normal and overweight and obese Malaysian children.
Obesity Research and Clinical Practice, 8(1), e70-78.
doi:http://dx.doi.org/10.1016/j.orcp.2012.12.001
Flores, G. B., Andrade, F., & Lima, D. R. (2000). Can coffee help fighting the drug
problem? Preliminary results of a Brazilian youth drug study. Acta
Pharmacologica Sinica, 21(12), 1059-1070.
Food and Drug Administration. (2013). Retrieved from
http://www.fda.gov/ForConsumers/ConsumerUpdates/ucm350570.htm
Food Standards Australia New Zealand. (2015). NUTTAB 2010 Online Searchable
Database. Retrieved from
http://www.foodstandards.gov.au/science/monitoringnutrients/nutrientables/n
uttab/Pages/default.aspx
Franckle, R. L., Falbe, J., Gortmaker, S., Ganter, C., Taveras, E. M., Land, T., & Davison, K.
K. (2015). Insufficient sleep among elementary and middle school students is
linked with elevated soda consumption and other unhealthy dietary behaviors.
Preventive Medicine, 74, 36-41.
doi:http://dx.doi.org/10.1016/j.ypmed.2015.02.007
Frank, M. G., & Benington, J. H. (2006). The role of sleep in memory consolidation and
brain plasticity: dream or reality? The Neuroscientist, 12(6), 477-488.
Fredholm, B. B., Bättig, K., Holmén, J., Nehlig, A., & Zvartau, E. E. (1999). Actions of
caffeine in the brain with special reference to factors that contribute to its
widespread use. Pharmacological Reviews, 51(1), 83-133.
Fulkerson, J. A., Sherwood, N. E., Perry, C. L., Neumark-Sztainer, D., & Story, M. (2004).
Depressive symptoms and adolescent eating and health behaviors: a
multifaceted view in a population-based sample. Preventive Medicine, 38(6), 865-
875.
Gameau, D. (2016). Damon Gameau, on how to cut your sugar consumption in half.
news.com.au. Retrieved from
http://www.news.com.au/lifestyle/health/diet/damon-gameau-on-how-to-cut-
your-sugar-consumption-in-half/news-
story/3617ff4de783b714e21acd2e5faaf5fd
Gangwisch, J. E., Malaspina, D., Boden-Albala, B., & Heymsfield, S. B. (2005). Inadequate
sleep as a risk factor for obesity: analyses of the NHANES I. SLEEP, 28(10), 1289.
Garaulet, M., Ortega, F., Ruiz, J., Rey-Lopez, J., Beghin, L., Manios, Y., . . . Moreno, L.
(2011). Short sleep duration is associated with increased obesity markers in
European adolescents: Effect of physical activity and dietary habits. The HELENA
study. International Journal of Obesity, 35(10), 1308-1317.
doi:10.1038/ijo.2011.149
Garcia-Garcia, F., Juarez-Aguilar, E., Santiago-Garcia, J., & Cardinali, D. P. (2014). Ghrelin
and its interactions with growth hormone, leptin and orexins: Implications for
the sleep–wake cycle and metabolism. Sleep Medicine Reviews, 18(1), 89-97.
Genzel, L., Kroes, M. C., Dresler, M., & Battaglia, F. P. (2014). Light sleep versus slow
wave sleep in memory consolidation: a question of global versus local processes?
Trends in Neurosciences, 37(1), 10-19.
Ghandour, R. M., Overpeck, M. D., Huang, Z. J., Kogan, M. D., & Scheidt, P. C. (2004).
Headache, stomachache, backache, and morning fatigue among adolescent girls
193 | P a g e
in the United States: associations with behavioral, sociodemographic, and
environmental factors. Archives of Pediatrics and Adolescent Medicine, 158(8),
797-803.
Giannotti, F., Cortesi, F., Sebastiani, T., & Ottaviano, S. (2002). Circadian preference,
sleep and daytime behaviour in adolescence. Journal of Sleep Research, 11(3),
191-199.
Gibson, E. S., Powles, A. C., Thabane, L., O'Brien, S., Molnar, D. S., Trajanovic, N.,
Chilcott-Tanser, L. (2006). "Sleepiness" is serious in adolescence: two surveys of
3235 Canadian students. BMC Public Health, 6, 116.
Gibson, T. M., Ferrucci, L. M., Tangrea, J. A., & Schatzkin, A. (2010). Epidemiological and
Clinical Studies of Nutrition. Paper presented at the Seminars in oncology.
Goel, N., Rao, H., Durmer, J. S., & Dinges, D. F. (2009). Neurocognitive Consequences of
Sleep Deprivation. Paper presented at the Seminars in neurology.
Gottlieb, D. J., Redline, S., Nieto, F. J., Baldwin, C. M., Newman, A. B., Resnick, H. E., &
Punjabi, N. M. (2006). Association of usual sleep duration with hypertension: The
Sleep Heart Health Study. SLEEP, 29(8), 1009-1014.
Gozal, D. (1998). Sleep-disordered breathing and school performance in children.
Pediatrics, 102(3), 616.
Gozal, D., & Pope Jr, D. (2001). Snoring during early childhood and academic
performance at ages thirteen to fourteen years. Pediatrics, 107(6), 1394.
Grandner, M. A., Jackson, N., Gerstner, J. R., & Knutson, K. L. (2013). Dietary nutrients
associated with short and long sleep duration. Data from a nationally
representative sample. Appetite, 64, 71-80.
Grandner, M. A., Jackson, N., Gerstner, J. R., & Knutson, K. L. (2014). Sleep symptoms
associated with intake of specific dietary nutrients. Journal of Sleep Research, 23,
22-34.
Grandner, M. A., Kripke, D. F., Naidoo, N., & Langer, R. D. (2010). Relationships among
dietary nutrients and subjective sleep, objective sleep, and napping in women.
Sleep Medicine, 11(2), 180-184.
Grant, D. M., Tang, B. K., & Kalow, W. (1983). Variability in caffeine metabolism. Clinical
Pharmacology and Therapeutics, 33(5), 591-602.
Gregory, A. M., & Sadeh, A. (2012). Sleep, emotional and behavioral difficulties in
children and adolescents. Sleep Medicine Reviews, 16(2), 129-136.
Griefahn, B. (2002). Sleep disturbances related to environmental noise. Noise and
Health, 4(15), 57.
Grosso, G., Micek, A., Godos, J., Sciacca, S., Pajak, A., Martínez-González, M. A., . . .
Galvano, F. (2016). Coffee consumption and risk of all-cause, cardiovascular, and
cancer mortality in smokers and non-smokers: a dose-response meta-analysis:
Springer.
Grosso, L. M., & Bracken, M. B. (2005). Caffeine metabolism, genetics, and perinatal
outcomes: a review of exposure assessment considerations during pregnancy.
Annals of Epidemiology, 15(6), 460-466.
Gruber, R., Cassoff, J., Frenette, S., Wiebe, S., & Carrier, J. (2012). Impact of sleep
extension and restriction on children’s emotional lability and impulsivity.
Pediatrics, 130(5), 1155-1161.
194 | P a g e
Grunstein, R. (2005). Endocrine disorders. In M. Kryger, T. Roth, & W. Dement (Eds.),
Principles and practice of sleep medicine. 4th ed. (pp. 1237-1245). Philadelphia:
Elsevier Saunders.
Grunstein, R. R., & Sullivan, C. E. (1988). Sleep apnea and hypothyroidism: mechanisms
and management. The American Journal of Medicine, 85(6), 775-779.
Guilleminault, C., Eldridge, F., Simmons, F., & Dement, W. (1976). Sleep apnea in eight
children. Pediatrics, 58(1), 23.
Gupta, N., Mueller, W., Chan, W., & Meininger, J. (2002). Is obesity associated with poor
sleep quality in adolescents? American Journal of Human Biology, 14(6), 762-768.
Harrison, Y., & Horne, J. A. (2000). Sleep loss and temporal memory. The Quarterly
Journal of Experimental Psychology: Section A, 53(1), 271-279.
Hartmann, E. (1983). Effects of L-tryptophan on sleepiness and on sleep. Journal of
Psychiatric Research, 17(2), 107-113.
Haskell, E. H., Palca, J. W., Walker, J. M., Berger, R. J., & Heller, H. C. (1981). The effects
of high and low ambient temperatures on human sleep stages.
Electroencephalography and Clinical Neurophysiology, 51(5), 494-501.
Hastings, M., O’Neill, J. S., & Maywood, E. S. (2007). Circadian clocks: regulators of
endocrine and metabolic rhythms. Journal of Endocrinology, 195(2), 187-198.
He, F., Bixler, E. O., Berg, A., Kawasawa, Y. I., Vgontzas, A. N., Fernandez-Mendoza, J., . . .
Liao, D. (2015). Habitual sleep variability, not sleep duration, is associated with
caloric intake in adolescents. Sleep Medicine, 16(7), 856-861.
doi:http://dx.doi.org/10.1016/j.sleep.2015.03.004
Health Canada. (2012). Food and nutrition: Caffeine in food. Retrieved from
http://www.hc-sc.gc.ca/fn-an/securit/addit/caf/food-caf-aliments-eng.php
Heath, G., Roach, G. D., Dorrian, J., Ferguson, S. A., Darwent, D., & Sargent, C. (2012).
The effect of sleep restriction on snacking behaviour during a week of simulated
shiftwork. Accident Analysis and Prevention, 45, 62-67.
Heaton, K., & Griffin, R. (2015). The Effects of Caffeine Use on Driving Safety Among
Truck Drivers Who Are Habitual Caffeine Users. Workplace health and safety,
63(8), 333-341. doi:http://dx.doi.org/10.1177/2165079915579561
Hebert, J. R., Clemow, L., Pbert, L., Ockene, I. S., & Ockene, J. K. (1995). Social desirability
bias in dietary self-report may compromise the validity of dietary intake
measures. International Journal of Epidemiology, 24(2), 389-398.
Herman, M. A., & Kahn, B. B. (2006). Glucose transport and sensing in the maintenance
of glucose homeostasis and metabolic harmony. The Journal of Clinical
Investigation, 116(7), 1767-1775.
Herscovitch, J., & Broughton, R. (1981). Performance deficits following short-term partial
sleep deprivation and subsequent recovery oversleeping. Canadian Journal of
Psychology/Revue Canadienne de Psychologie, 35(4), 309.
Hicks, R. A., Hicks, G. J., Reyes, J. R., & Cheers, Y. (1983). Daily caffeine use and the sleep
of college students. Bulletin of the Psychonomic Society, 21(1), 24-25.
Hill, R., & Davies, P. (2001). The validity of self-reported energy intake as determined
using the doubly labelled water technique. British Journal of Nutrition, 85(04),
415-430.
Hindmarch, I., Rigney, U., Stanley, N., Quinlan, P., Rycroft, J., & Lane, J. (2000). A
naturalistic investigation of the effects of day-long consumption of tea, coffee
195 | P a g e
and water on alertness, sleep onset and sleep quality. Psychopharmacology,
149(3), 203-216.
Hirshkowitz, M., Whiton, K., Albert, S. M., Alessi, C., Bruni, O., DonCarlos, L., . . . Katz, E.
S. (2015). National Sleep Foundation’s updated sleep duration
recommendations: final report. Sleep Health, 1(4), 233-243.
Hitze, B., Bosy-Westphal, A., Bielfeldt, F., Settler, U., Plachta-Danielzik, S., Pfeuffer, M., . .
. Muller, M. J. (2009). Determinants and impact of sleep duration in children and
adolescents: data of the Kiel Obesity Prevention Study. European Journal of
Clinical Nutrition, 63(6), 739-746. doi:http://dx.doi.org/10.1038/ejcn.2008.41
Hjorth, M., Quist, J., Andersen, R., Michaelsen, K., Tetens, I., Astrup, A., . . . Sjodin, A.
(2014). Change in sleep duration and proposed dietary risk factors for obesity in
Danish school children. Pediatric Obesity, 9(6), 156-159.
doi:http://dx.doi.org/10.1111/ijpo.264
Hoddes, E., Zarcone, V., Smythe, H., Phillips, R., & Dement, W. (1973). Quantification of
sleepiness: a new approach. Psychophysiology, 10(4), 431-436.
Hodge, A., Patterson, A. J., Brown, W. J., Ireland, P., & Giles, G. (2000). The Anti Cancer
Council of Victoria FFQ: relative validity of nutrient intakes compared with
weighed food records in young to middle-aged women in a study of iron
supplementation. Australian and New Zealand Journal of Public Health, 24(6),
576 -583.
Hogenkamp, P. S., Nilsson, E., Nilsson, V. C., Chapman, C. D., Vogel, H., Lundberg, L. S., . .
. Broman, J.-E. (2013). Acute sleep deprivation increases portion size and affects
food choice in young men. Psychoneuroendocrinology, 38(9), 1668-1674.
Hollander, L. E., Freeman, E. W., Sammel, M. D., Berlin, J. A., Grisso, J. A., & Battistini, M.
(2001). Sleep quality, estradiol levels, and behavioral factors in late reproductive
age women. Obstetrics and Gynecology, 98(3), 391-397.
Horne, J. (1988). Why we sleep: the functions of sleep in humans and other mammals:
Oxford University Press.
Hu, F. B., & Malik, V. S. (2010). Sugar-sweetened beverages and risk of obesity and type
2 diabetes: epidemiologic evidence. Physiology & behavior, 100(1), 47-54.
Huang, Y. S., Wang, C. H., & Guilleminault, C. (2010). An epidemiologic study of sleep
problems among adolescents in North Taiwan. Sleep Medicine, 11(10), 1035-
1042. doi:http://dx.doi.org/10.1016/j.sleep.2010.04.009
Hunsberger, M., Mehlig, K., Bornhorst, C., Hebestreit, A., Moreno, L., Veidebaum, T., . . .
Lissner, L. (2015). Dietary carbohydrate and nocturnal sleep duration in relation
to children's BMI: Findings from the idefics study in eight european countries.
Nutrients, 7(12), 10223-10236. doi:http://dx.doi.org/10.3390/nu7125529
Huntley, E. D., & Juliano, L. M. (2012). Caffeine Expectancy Questionnaire (CaffEQ):
Construction, psychometric properties, and associations with caffeine use,
caffeine dependence, and other related variables. Psychological Assessment,
24(3), 592-607. doi:http://dx.doi.org/10.1037/a0026417
Imaki, M., Hatanaka, Y., Ogawa, Y., Yoshida, Y., & Tanada, S. (2002). An epidemiological
study on relationship between the hours of sleep and life style factors in
Japanese factory workers. Journal of Physiological Anthropology and Applied
Human Science, 21(2), 115-120.
196 | P a g e
Insana, S. P., Hall, M., Buysse, D. J., & Germain, A. (2013). Validation of the Pittsburgh
Sleep Quality Index Addendum for Posttraumatic Stress Disorder (PSQI‐A) in US
Male Military Veterans. Journal of Traumatic Stress, 26(2), 192-200.
Irish, L. A., Kline, C. E., Rothenberger, S. D., Krafty, R. T., Buysse, D. J., Kravitz, H. M., . . .
Hall, M. H. (2014). A 24-hour approach to the study of health behaviors:
Temporal relationships between waking health behaviors and sleep. Annals of
Behavioral Medicine, 47(2), 189-197. doi:http://dx.doi.org/10.1007/s12160-013-
9533-3
Ivanenko, A., Barnes, M., Crabtree, V., & Gozal, D. (2004). Psychiatric symptoms in
children with insomnia referred to a pediatric sleep medicine center. Sleep
Medicine, 5(3), 253-259.
James, J. E. (2004). Critical review of dietary caffeine and blood pressure: a relationship
that should be taken more seriously. Psychosomatic Medicine, 66(1), 63-71.
James, J. E., Kristjansson, A. L., & Sigfusdottir, I. D. (2011). Adolescent substance use,
sleep, and academic achievement: evidence of harm due to caffeine. Journal of
Adolescence, 34, 665-673. doi:10.1016/j.adolescence.2010.09.006
James, J. E., Kristjansson, A. L., & Sigfusdottir, I. D. (2015). A gender-specific analysis of
adolescent dietary caffeine, alcohol consumption, anger, and violent behavior.
Substance Use and Misuse, 50(2), 257-267.
doi:http://dx.doi.org/10.3109/10826084.2014.977394
James, J. E., & Rogers, P. J. (2005). Effects of caffeine on performance and mood:
withdrawal reversal is the most plausible explanation. Psychopharmacology,
182(1), 1-8.
Jenni, O. G., & Carskadon, M. A. (2007). Sleep behavior and sleep regulation from
infancy through adolescence: normative aspects. Sleep Medicine Clinics, 2(3),
321-329.
Jennings, J., Muldoon, M., & Hall, M. (2007). Self-reported sleep quality is associated
with the metabolic syndrome. SLEEP, 30, 219-223.
Johns, M. W. (1991). A new method for measuring daytime sleepiness: the Epworth
sleepiness scale. SLEEP, 14(6), 540-545.
Johnson, R. K., Goran, M. I., & Poehlman, E. T. (1994). Correlates of over-and
underreporting of energy intake in healthy older men and women. The American
Journal of Clinical Nutrition, 59(6), 1286-1290.
Kant, A. K., & Graubard, B. I. (2014). Association of self-reported sleep duration with
eating behaviors of American adults: NHANES 2005-2010. American Journal of
Clinical Nutrition, 100(3), 938-947.
doi:http://dx.doi.org/10.3945/ajcn.114.085191
Kaplan, B. J., McNicol, J., Conte, R. A., & Moghadam, H. (1989). Dietary replacement in
preschool-aged hyperactive boys. Pediatrics, 83, 7-17.
Kaplan, G. B., Greenblatt, D. J., Ehrenberg, B. L., Goddard, J. E., Cotreau, M. M., Harmatz,
J. S., & Shader, R. I. (1997). Dose‐dependent pharmacokinetics and psychomotor
effects of caffeine in humans. The Journal of Clinical Pharmacology, 37(8), 693-
703.
Kazarian, S. S., Howe, M. G., & Csapo, K. G. (1979). Development of the sleep behavior
self-rating scale. Behavior Therapy, 10(3), 412-417.
197 | P a g e
Kim, K. (2013). Association between Internet overuse and aggression in Korean
adolescents. Pediatrics International, 55(6), 703-709.
doi:http://dx.doi.org/10.1111/ped.12171
Kim, S., DeRoo, L. A., & Sandler, D. P. (2011). Eating patterns and nutritional
characteristics associated with sleep duration. Public Health Nutrition, 14(05),
889-895.
Kjeldsen, J., Hjorth, M., Andersen, R., Michaelsen, K., Tetens, I., Astrup, A., Sjodin, A.
(2014). Short sleep duration and large variability in sleep duration are
independently associated with dietary risk factors in obesity in Danish school
children. International Journal of Obesity, 38(1), 32-39.
doi:http://dx.doi.org/10.1038/ijo.2013.147
Kluge, M., Gazea, M., Schüssler, P., Genzel, L., Dresler, M., Kleyer, S., Steiger, A. (2010).
Ghrelin increases slow wave sleep and stage 2 sleep and decreases stage 1 sleep
and REM sleep in elderly men but does not affect sleep in elderly women.
Psychoneuroendocrinology, 35(2), 297-304.
Knutson, K. L., & Van Cauter, E. (2008). Associations between sleep loss and increased
risk of obesity and diabetes. Annals of the New York Academy of Sciences,
1129(1), 287-304.
Koopman, C., Nouriani, B., Erickson, V., Anupindi, R., Butler, L. D., Bachmann, M. H.,
Spiegel, D. (2002). Sleep disturbances in women with metastatic breast cancer.
The Breast Journal, 8(6), 362-370.
Krall, E. A., Dwyer, J. T., & Coleman, K. A. (1988). Factors influencing accuracy of dietary
recall. Nutrition Research, 8(7), 829-841.
Krebs-Smith, S., Graubard, B., Kahle, L., Subar, A., Cleveland, L., & Ballard-Barbash, R.
(2000). Low energy reporters vs others: a comparison of reported food intakes.
Kripke, D. F., Garfinkel, L., Wingard, D. L., Klauber, M. R., & Marler, M. R. (2002).
Mortality associated with sleep duration and insomnia. Archives of General
Psychiatry, 59(2), 131-136.
Kristal, A. R., Peters, U., & Potter, J. D. (2005). Is it time to abandon the food frequency
questionnaire? Cancer Epidemiology Biomarkers and Prevention, 14(12), 2826-
2828.
Kristjansson, A. L., Sigfusdottir, I. D., Allegrante, J. P., & James, J. E. (2011). Adolescent
caffeine consumption, daytime sleepiness, and anger. Journal of Caffeine
Research, 1, 75-82. doi:http://dx.doi.org/10.1089/jcr.2011.0008
Kristjansson, A. L., Sigfusdottir, I. D., Frost, S. S., & James, J. E. (2013). Adolescent
caffeine consumption and self-reported violence and conduct disorder. Journal of
Youth and Adolescence, 42(7), 1053-1062. doi:http://dx.doi.org/10.1007/s10964-
013-9917-5
Kristjansson, A. L., Sigfusdottir, I. D., Mann, M. J., & James, J. E. (2014). Caffeinated
sugar-sweetened beverages and common physical complaints in Icelandic
children aged 10-12 years. Preventive Medicine, 58, 40-44.
Kroke, A., Klipstein-Grobusch, K., Voss, S., Möseneder, J., Thielecke, F., Noack, R., &
Boeing, H. (1999). Validation of a self-administered food-frequency questionnaire
administered in the European Prospective Investigation into Cancer and Nutrition
(EPIC) Study: comparison of energy, protein, and macronutrient intakes
198 | P a g e
estimated with the doubly labeled water, urinary nitrogen, and repeated 24-h
dietary recall methods. The American Journal of Clinical Nutrition, 70(4), 439-447.
Kuczmarski, R. J., Ogden, C. L., Guo, S. S., Grummer-Strawn, L. M., Flegal, K. M., Mei, Z., .
. . Johnson, C. L. (2002). 2000 CDC Growth Charts for the United States: methods
and development. Vital and health statistics. Series 11, Data from the national
health survey(246), 1-190.
Lack, L., & Lushington, K. (1996). The rhythms of human sleep propensity and core body
temperature. Journal of Sleep Research, 5(1), 1.
Lacks, P., & Rotert, M. (1986). Knowledge and practice of sleep hygiene techniques in
insomniacs and good sleepers. Behaviour Research and Therapy, 24(3), 365-368.
Lambert, O. (2016). New research finds sugar addiction is like drug abuse. news.com.au.
Retrieved from http://www.news.com.au/lifestyle/health/health-problems/new-
research-finds-sugar-addiction-is-like-drug-abuse/news-
story/e67611eed4bba66a94e7d945669c22b2
Landolt, H. P., Werth, E., Borbély, A. A., & Dijk, D.-J. (1995). Caffeine intake (200 mg) in
the morning affects human sleep and EEG power spectra at night. Brain
Research, 675(1), 67-74.
Lanius, R. A. (2011). Default mode alterations in posttraumatic stress disorder related to
early-life trauma: a developmental perspective. Journal of Psychiatry &
Neuroscience: JPN, 36(1), 56.
Lavigne, G., McMillan, D., & Zucconi, M. (2005). Pain and sleep. In M. Kryger, T. Roth, &
W. Dement (Eds.), Principles and practice of sleep medicine. 4th ed. (pp. 1246-
1255). Philadelphia: Elsevier Saunders.
LeBourgeois, M. K., Giannotti, F., Cortesi, F., Wolfson, A. R., & Harsh, J. (2005). The
relationship between reported sleep quality and sleep hygiene in Italian and
American adolescents. Pediatrics, 115, 257-265.
Leproult, R., & Van Cauter, E. (2009). Role of sleep and sleep loss in hormonal release
and metabolism Pediatric Neuroendocrinology (Vol. 17, pp. 11-21): Karger
Publishers.
Levinson, P. D. (1995). Weight gain in women was associated with type 2 diabetes
mellitus. ACP Journal Club, 123(2), 50-50.
Lewandowski, A. S., Toliver-Sokol, M., & Palermo, T. M. (2011). Evidence-based review
of subjective pediatric sleep measures. Journal of Pediatric Psychology, 36(7),
780-793.
Li, L., Ren, J., Shi, L., Jin, X., Yan, C., Jiang, F., . . . Li, S. (2014). Frequent nocturnal
awakening in children: Prevalence, risk factors, and associations with subjective
sleep perception and daytime sleepiness. BMC Psychiatry, 14 (1) (204).
doi:http://dx.doi.org/10.1186/1471-244X-14-204
Li, S., Zhu, S., Jin, X., Yan, C., Wu, S., Jiang, F., & Shen, X. (2010a). Risk factors associated
with short sleep duration among Chinese school-aged children. Sleep Medicine,
11, 907-916. doi:http://dx.doi.org/10.1016/j.sleep.2010.03.018
Li, S., Zhu, S., Jin, X., Yan, C., Wu, S., Jiang, F., & Shen, X. (2010b). Risk factors associated
with short sleep duration among Chinese school-aged children. Sleep Medicine,
11(9), 907-916. doi:http://dx.doi.org/10.1016/j.sleep.2010.03.018
Lichtman, S. W., Pisarska, K., Berman, E. R., Pestone, M., Dowling, H., Offenbacher, E., . .
. Heymsfield, S. B. (1992). Discrepancy between self-reported and actual caloric
199 | P a g e
intake and exercise in obese subjects. New England Journal of Medicine, 327(27),
1893-1898.
Lieberman, H., Wurtman, R., Emde, G., Roberts, C., & Coviella, I. (1987). The effects of
low doses of caffeine on human performance and mood. Psychopharmacology,
92(3), 308-312.
Liebman, B. F., & Moyer, G. (1980). The case against sugar. Nutrition action-Center for
Science in the Public Interest (USA).
Lien, N., Jacobs, D. R., & Klepp, K.-I. (2002). Exploring predictors of eating behaviour
among adolescents by gender and socio-economic status. Public Health
Nutrition, 5(05), 671-681.
Lim, J., & Dinges, D. F. (2008). Sleep deprivation and vigilant attention. Annals of the
New York Academy of Sciences, 1129(1), 305-322.
Liu, X., Forbes, E., Ryan, N., Rofey, D., Hannon, T., & Dahl, R. (2008). Rapid eye
movement sleep in relation to overweight in children and adolescents. Archives
of General Psychiatry, 65(8), 924.
Livingstone, M., Prentice, A., Strain, J., Coward, W., Black, A., Barker, M., . . . Whitehead,
R. (1990). Accuracy of weighed dietary records in studies of diet and health.
British Medical Journal, 300(6726), 708-712.
Lodato, F., Araujo, J., Barros, H., Lopes, C., Agodi, A., Barchitta, M., & Ramos, E. (2013).
Caffeine intake reduces sleep duration in adolescents. Nutrition Research, 33(9),
726-732. doi:http://dx.doi.org/10.1016/j.nutres.2013.06.005
Lohsoonthorn, V., Khidir, H., Casillas, G., Lertmaharit, S., Tadesse, M. G., Pensuksan, W.
C., . . . Williams, M. A. (2013). Sleep quality and sleep patterns in relation to
consumption of energy drinks, caffeinated beverages, and other stimulants
among Thai college students. Sleep and Breathing, 17(3), 1017-1028.
doi:http://dx.doi.org/10.1007/s11325-012-0792-1
Bryant Ludden, A., & Wolfson, A. R. (2010). Understanding adolescent caffeine use: connecting use patterns
with expectancies, reasons, and sleep. Health Education & Behavior, 37(3), 330-342.
Luking, K. R., Luby, J. L., & Barch, D. M. (2014). Kids, candy, brain and behavior: Age
differences in responses to candy gains and losses. Developmental Cognitive
Neuroscience, 9, 82-92.
Lund, H. G., Reider, B. D., Whiting, A. B., & Prichard, J. R. (2010). Sleep patterns and
predictors of disturbed sleep in a large population of college students. Journal of
Adolescent Health, 46(2), 124-132.
Malik, V. S., Pan, A., Willett, W. C., & Hu, F. B. (2013). Sugar-sweetened beverages and
weight gain in children and adults: a systematic review and meta-analysis. The
American Journal of Clinical Nutrition, 98(4), 1084-1102.
Malik, V. S., Popkin, B. M., Bray, G. A., Després, J.-P., Willett, W. C., & Hu, F. B. (2010).
Sugar-sweetened beverages and risk of metabolic syndrome and type 2 diabetes
a meta-analysis. Diabetes Care, 33(11), 2477-2483.
Malinauskas, B. M., Aeby, V. G., Overton, R. F., Carpenter-Aeby, T., & Barber-Heidal, K.
(2007). A survey of energy drink consumption patterns among college students.
Nutrition Journal, 6, 35.
Mamun, A., Lawlor, D., Cramb, S., O'Callaghan, M., Williams, G., & Najman, J. (2007). Do
childhood sleeping problems predict obesity in young adulthood? Evidence from
a prospective birth cohort study. American Journal of Epidemiology, 166(12),
1368.
200 | P a g e
Marino, M., Li, Y., Rueschman, M. N., Winkelman, J., Ellenbogen, J., Solet, J., . . . Buxton,
O. M. (2013). Measuring sleep: accuracy, sensitivity, and specificity of wrist
actigraphy compared to polysomnography. SLEEP, 36(11), 1747.
Markwald, R. R., Melanson, E. L., Smith, M. R., Higgins, J., Perreault, L., Eckel, R. H., &
Wright, K. P. (2013). Impact of insufficient sleep on total daily energy
expenditure, food intake, and weight gain. Proceedings of the National Academy
of Sciences, 110(14), 5695-5700.
Marshall, L., & Born, J. (2007). The contribution of sleep to hippocampus-dependent
memory consolidation. Trends in Cognitive Sciences, 11(10), 442-450.
Marventano, S., Salomone, F., Godos, J., Pluchinotta, F., Del Rio, D., Mistretta, A., &
Grosso, G. (2016). Coffee and tea consumption in relation with non-alcoholic
fatty liver and metabolic syndrome: A systematic review and meta-analysis of
observational studies. Clinical Nutrition.
Mastin, D., Peszka, J., Poling, T., Phillips, R., & Duke, J. (2005). Personality as a predictor
of the objective and subjective impact of sleep deprivation. Personality and
Individual Differences, 39(8), 1471-1482.
Matricciani, L., Blunden, S., Rigney, G., Williams, M. T., & Olds, T. S. (2013). Children’s
sleep needs: is there sufficient evidence to recommend optimal sleep for
children. SLEEP, 36(4), 527-534.
Maxted, A. (2015). Did banning sugar turn my kids into sweet little angels? The
Telegraph. Retrieved from
http://www.telegraph.co.uk/foodanddrink/foodanddrinkadvice/11375030/Did-
banning-sugar-turn-my-kids-into-sweet-little-angels.html
Maxwell, J. P., Sukhodolsky, D. G., Chow, C. C., & Wong, C. F. (2005). Anger rumination in
Hong Kong and Great Britain: Validation of the scale and a cross-cultural
comparison. Personality and Individual Differences, 39, 1147-1157.
McCulloch, A., Wiggins, R. D., Joshi, H. E., & Sachdev, D. (2000). Internalising and
externalising children's behaviour problems in Britain and the US: relationships
to family resources1. Children and Society, 14, 368-383.
Meijer, A. M. (2008). Chronic sleep reduction, functioning at school and school
achievement in preadolescents. Journal of Sleep Research, 17(4), 395-405.
Meijer, A. M., Habekothé, H. T., & Van Den Wittenboer, G. L. (2000). Time in bed, quality
of sleep and school functioning of children. Journal of Sleep Research, 9(2), 145.
Melendres, C. S., Lutz, J. M., Rubin, E. D., & Marcus, C. L. (2004). Daytime sleepiness and
hyperactivity in children with suspected sleep-disordered breathing. Pediatrics,
114(3), 768-775.
Mikulincer, M., Babkoff, H., Caspy, T., & Sing, H. (1989). The effects of 72 hours of sleep
loss on psychological variables. British Journal of Psychology, 80(2), 145-162.
Miles, S., & Scaife, V. (2003). Optimistic bias and food. Nutrition Research Reviews,
16(01), 3-19.
Milich, R., Wolraich, M., & Lindgren, S. (1986). Sugar and hyperactivity: A critical review
of empirical findings. Clinical Psychology Review, 6(6), 493-513.
Mindell, J. A., Meltzer, L. J., Carskadon, M. A., & Chervin, R. D. (2009). Developmental
aspects of sleep hygiene: findings from the 2004 National Sleep Foundation Sleep
in America Poll. Sleep Medicine, 10(7), 771-779.
doi:http://dx.doi.org/10.1016/j.sleep.2008.07.016
201 | P a g e
Mniszek, D. H. (1988). Brighton sleep survey: a study of sleep in 20-45-year olds. The
Journal of International Medical Research, 16(1), 61-65.
Molag, M. L., de Vries, J. H., Ocké, M. C., Dagnelie, P. C., van den Brandt, P. A., Jansen,
M. C., . . . van't Veer, P. (2007). Design characteristics of food frequency
questionnaires in relation to their validity. American Journal of Epidemiology,
166(12), 1468-1478.
Moreira, P., Santos, S., Padrao, P., Cordeiro, T., Bessa, M., Valente, H., . . . Moreira, A.
(2010). Food patterns according to sociodemographics, physical activity, sleeping
and obesity in Portuguese children. International Journal of Environmental
Research and Public Health, 7(3), 1121-1138.
doi:http://dx.doi.org/10.3390/ijerph7031121
Morris, G. O., Williams, H. L., & Lubin, A. (1960). Misperception and disorientation during
sleep deprivation. AMA Archives of General Psychiatry, 2(3), 247-254.
Morselli, L., Leproult, R., Balbo, M., & Spiegel, K. (2010). Role of sleep duration in the
regulation of glucose metabolism and appetite. Best Practice and Research
Clinical Endocrinology and Metabolism, 24(5), 687-702.
Moshfegh, A. J., Rhodes, D. G., Baer, D. J., Murayi, T., Clemens, J. C., Rumpler, W. V., . . .
Ingwersen, L. A. (2008). The US Department of Agriculture Automated Multiple-
Pass Method reduces bias in the collection of energy intakes. The American
Journal of Clinical Nutrition, 88(2), 324-332.
Mossavar-Rahmani, Y., Jung, M., Patel, S. R., Sotres-Alvarez, D., Arens, R., Ramos, A., . . .
Van Horn, L. (2015). Eating behavior by sleep duration in the Hispanic
Community Health Study/Study of Latinos. Appetite, 95, 275-284.
doi:http://dx.doi.org/10.1016/j.appet.2015.07.014
Moynihan, P., & Kelly, S. (2013). Effect on caries of restricting sugars intake systematic
review to inform WHO guidelines. Journal of Dental Research,
0022034513508954.
Muzet, A. (2007). Environmental noise, sleep and health. Sleep Medicine Reviews, 11(2),
135-142.
Muzet, A., Libert, J.-P., & Candas, V. (1984). Ambient temperature and human sleep.
Cellular and Molecular Life Sciences, 40(5), 425-429.
Muzur, A., Pace-Schott, E. F., & Hobson, J. A. (2002). The prefrontal cortex in sleep.
Trends in Cognitive Sciences, 6(11), 475-481.
National Health and Medical Research Council. (2013). Eat for health: Australian Dietary
Guidelines. Retrieved from https://www.eatforhealth.gov.au/guidelines
Nawrot, P., Jordan, S., Eastwood, J., Rotstein, J., Hugenholtz, A., & Feeley, M. (2003).
Effects of caffeine on human health. Food Additives and Contaminants, 20(1), 1-
30.
Newbrun, E. (1982). Sugar and dental caries: a review of human studies. Science,
217(4558), 418-423.
Noland, H., Price, J., Dake, J., & Telljohann, S. (2009). Adolescents’ sleep behaviors and
perceptions of sleep. Journal of School Health, 79(5), 224-230.
Ohayon, M. M., Carskadon, M. A., Guilleminault, C., & Vitiello, M. V. (2004). Meta-analysis of quantitative
sleep parameters from childhood to old age in healthy individuals: developing normative sleep
values across the human lifespan. SLEEP, 27, 1255-1274.
Okura, K., Lavigne, G. J., Huynh, N., Manzini, C., Fillipini, D., & Montplaisir, J. Y. (2008).
Comparison of sleep variables between chronic widespread musculoskeletal
202 | P a g e
pain, insomnia, periodic leg movements syndrome and control subjects in a
clinical sleep medicine practice. Sleep Medicine, 9(4), 352-361.
Orbeta, R. L., Overpeck, M. D., Ramcharran, D., Kogan, M. D., & Ledsky, R. (2006). High
caffeine intake in adolescents: associations with difficulty sleeping and feeling
tired in the morning. Journal of Adolescent Health, 38(4), 451-453.
Owens, J. A., Mindell, J., & Baylor, A. (2014). Effect of energy drink and caffeinated
beverage consumption on sleep, mood, and performance in children and
adolescents. Nutrition Reviews, 72, 65-71. doi:10.1111/nure.12150
Owens, J. A., Spirito, A., & McGuinn, M. (2000). The Children's Sleep Habits
Questionnaire (CSHQ): psychometric properties of a survey instrument for
school-aged children. SLEEP, 23(8), 1043-1052.
Paciencia, I., Barros, H., Araujo, J., & Ramos, E. (2013). Association between sleep
duration and blood pressure in adolescents. Hypertension Research - Clinical &
Experimental, 36(8), 747-752. doi:10.1038/hr.2013.36
Pallanti, S., Bernardi, S., & Quercioli, L. (2006). The Shorter PROMIS Questionnaire and the Internet Addiction
Scale in the assessment of multiple addictions in a high-school population: prevalence and related
disability. CNS Spectrums, 11(12), 966-974.
Park, S., Sherry, B., Foti, K., & Blanck, H. M. (2012). Self-reported academic grades and
other correlates of sugar-sweetened soda intake among US adolescents. Journal
of the Academy of Nutrition and Dietetics, 112(1), 125-131.
doi:http://dx.doi.org/10.1016/j.jada.2011.08.045
Partinen, M., & Hublin, C. (2005). Epidemiology of sleep disorders. In M. Kryger, T. Roth,
& W. Dement (Eds.), Principles and Practice of Sleep Medicine. (pp. 694-716).
Philadelphia: Elsevier/Saunders.
Paruthi, S., Brooks, L. J., D'Ambrosio, C., Hall, W. A., Kotagal, S., Lloyd, R. M., . . . Quan, S. F. (2016). Consensus
statement of the American Academy of Sleep Medicine on the recommended amount of sleep for
healthy children: methodology and discussion. Journal of Clinical Sleep Medicine.
Paterson, J., Dorrian, J., Ferguson, S., Jay, S., Lamond, N., Murphy, P., . . . Dawson, D.
(2011). Changes in structural aspects of mood during 39–66h of sleep loss using
matched controls. Applied Ergonomics, 42(2), 196-201.
Peigneux, P., Laureys, S., Delbeuck, X., & Maquet, P. (2001). Sleeping brain, learning
brain. The role of sleep for memory systems. Neuroreport, 12(18), A111-A124.
Pelsser, L. M., Frankena, K., Buitelaar, J. K., & Rommelse, N. N. (2010). Effects of food on
physical and sleep complaints in children with ADHD: a randomised controlled
pilot study. European Journal of Pediatrics, 169(9), 1129-1138.
Pelsser, L. M., Frankena, K., Toorman, J., Savelkoul, H. F., Dubois, A. E., Pereira, R. R., . . .
Buitelaar, J. K. (2011). Effects of a restricted elimination diet on the behaviour of
children with attention-deficit hyperactivity disorder (INCA study): a randomised
controlled trial. The Lancet, 377(9764), 494-503.
Penetar, D., McCann, U., Thorne, D., Kamimori, G., Galinski, C., Sing, H., . . . Belenky, G.
(1993). Caffeine reversal of sleep deprivation effects on alertness and mood.
Psychopharmacology, 112(2-3), 359-365.
Peterson, M., & Benca, R. (2008). Sleep in mood disorders. Sleep Medicine Clinics, 3(2),
231-249.
Peuhkuri, K., Sihvola, N., & Korpela, R. (2012). Diet promotes sleep duration and quality.
Nutrition Research, 32(5), 309-319.
203 | P a g e
Philip, P., Ghorayeb, I., Stoohs, R., Menny, J., Dabadie, P., Bioulac, B., & Guilleminault, C.
(1996). Determinants of sleepiness in automobile drivers. Journal of
Psychosomatic Research, 41(3), 279-288.
Philip, P., Taillard, J., Guilleminault, C., Salva, M. Q., Bioulac, B., & Ohayon, M. (1999).
Long distance driving and self-induced sleep deprivation among automobile
drivers. SLEEP, 22, 475-484.
Pogash, C. (2003). Myth of the 'Twinkie defense' / The verdict in the Dan White case
wasn't based on his ingestion of junk food. SFGate. Retrieved from
http://www.sfgate.com/health/article/Myth-of-the-Twinkie-defense-The-
verdict-in-2511152.php
Pollak, C. P., & Bright, D. (2003). Caffeine consumption and weekly sleep patterns in US
seventh-, eighth-, and ninth-graders. Pediatrics, 111, 42-46.
Porkka-Heiskanen, T., Strecker, R. E., Thakkar, M., Bjørkum, A. A., Greene, R. W., &
McCarley, R. W. (1997). Adenosine: a mediator of the sleep-inducing effects of
prolonged wakefulness. Science, 276(5316), 1265-1268.
Prinz, R. J., Roberts, W. A., & Hantman, E. (1980). Dietary correlates of hyperactive
behavior in children. Journal of Consulting and Clinical Psychology, 48(6), 760.
Quach, J., Hiscock, H., & Wake, M. (2012). Sleep problems and mental health in primary
school new entrants: cross-sectional community-based study. Journal of
Paediatrics and Child Health, 48(12), 1076-1081.
doi:http://dx.doi.org/10.1111/j.1440-1754.2012.02466.x
Ramírez, J. M., Fujihara, T., & Goozen, S. v. (2001). Cultural and gender differences in
anger and aggression: A comparison between Japanese, Dutch, and Spanish
students. The Journal of Social Psychology, 141, 119-121.
Raper, N., Perloff, B., Ingwersen, L., Steinfeldt, L., & Anand, J. (2004). An overview of
USDA's dietary intake data system. Journal of Food Composition and Analysis,
17(3), 545-555.
Rebro, S. M., Patterson, R. E., Kristal, A. R., & Cheney, C. L. (1998). The effect of keeping
food records on eating patterns. Journal of the American Dietetic Association,
98(10), 1163-1165.
Rechtschaffen, A., & Kales, A. (1968). A manual of standardized terminology, techniques
and scoring system for sleep stages of human subjects.
Regestein, Q., Natarajan, V., Pavlova, M., Kawasaki, S., Gleason, R., & Koff, E. (2010).
Sleep debt and depression in female college students. Psychiatry Research,
176(1), 34-39. doi:http://dx.doi.org/10.1016/j.psychres.2008.11.006
Reid, A., & Baker, F. C. (2008). Perceived sleep quality and sleepiness in South African
university students. South African Journal of Psychology, 38(2), 287-303.
doi:10.1177/008124630803800203
Rétey, J. V., Adam, M., Gottselig, J. M., Khatami, R., Dürr, R., Achermann, P., & Landolt,
H.-P. (2006). Adenosinergic mechanisms contribute to individual differences in
sleep deprivation-induced changes in neurobehavioral function and brain
rhythmic activity. The Journal of Neuroscience, 26(41), 10472-10479.
Riboli, E., Hunt, K., Slimani, N., Ferrari, P., Norat, T., Fahey, M., . . . Vignat, J. (2002).
European Prospective Investigation into Cancer and Nutrition (EPIC): study
populations and data collection. Public Health Nutrition, 5(6b), 1113-1124.
204 | P a g e
Richards, G., & Smith, A. (2015). Caffeine consumption and self-assessed stress, anxiety,
and depression in secondary school children. Journal of Psychopharmacology,
29(12), 1236-1247. doi:http://dx.doi.org/10.1177/0269881115612404
Robertson, D., Frölich, J. C., Carr, R. K., Watson, J. T., Hollifield, J. W., Shand, D. G., &
Oates, J. A. (1978). Effects of caffeine on plasma renin activity, catecholamines
and blood pressure. New England Journal of Medicine, 298(4), 181-186.
Rockett, H. R., Breitenbach, M., Frazier, A. L., Witschi, J., Wolf, A. M., Field, A. E., &
Colditz, G. A. (1997). Validation of a youth/adolescent food frequency
questionnaire. Preventive Medicine, 26(6), 808-816.
Rodgers, C., Paterson, D., Cunningham, D., & Noble, E. (1995). Sleep deprivation: Effects
on work capacity, self-paced walking, contractile properties and perceived
exertion. Sleep: Journal of Sleep Research & Sleep Medicine.
Roehrs, T., & Roth, T. (2001). Sleep, sleepiness, sleep disorders and alcohol use and
abuse. Sleep Medicine Reviews, 5(4), 287-297.
Rogers, P. J. (2007). Caffeine, mood and mental performance in everyday life. Nutrition
Bulletin, 32(s1), 84-89.
Rosen, C., Storfer-Isser, A., Taylor, H., Kirchner, H., Emancipator, J., & Redline, S. (2004).
Increased behavioral morbidity in school-aged children with sleep-disordered
breathing. Pediatrics, 114(6), 1640.
Rowe, K. S. (1988). Synthetic food colourings and ‘hyperactivity’: A double-blind
crossover study. Journal of Paediatrics and Child Health, 24, 143-147.
doi:10.1111/j.1440-1754.1988.tb00307.x
Rudolph, E., Faerbinger, A., & Koenig, J. (2014). Caffeine intake from all sources in
adolescents and young adults in Austria. European Journal of Clinical Nutrition,
68(7), 793-798.
Rudolph, E., Färbinger, A., & König, J. (2012). Determination of the caffeine contents of
various food items within the Austrian market and validation of a caffeine
assessment tool (CAT). Food Additives & Contaminants: Part A, 29(12), 1849-
1860.
Rufus, E. B. (2005). Sugar Addiiction: A Step-by-step Guide to Overcoming Sugar
Addiction: AuthorHouse.
Ryan-Harshman, M., & Aldoori, W. (2007). Diet and colorectal cancer Review of the
evidence. Canadian Family Physician, 53(11), 1913-1920.
Sadeh, A. (2007). Consequences of sleep loss or sleep disruption in children. Sleep
Medicine Clinics, 2(3), 513-520.
Sadeh, A., Gruber, R., & Raviv, A. (2002). Sleep, Neurobehavioral Functioning, and
Behavior Problems in School Age Children. Child Development, 73(2), 405-417.
Sanchez-Ortuno, M., Moore, N., Taillard, J., Valtat, C., Leger, D., Bioulac, B., & Philip, P.
(2005). Sleep duration and caffeine consumption in a French middle-aged
working population. Sleep Medicine, 6(3), 247-251.
Satel, S. (2006). Is caffeine addictive?—A review of the literature. The American Journal of Drug and Alcohol
Abuse, 32(4), 493-502.
Schliep, K. C., Schisterman, E. F., Mumford, S. L., Perkins, N. J., Ye, A., Pollack, A. Z., . . .
Stanford, J. B. (2013). Validation of different instruments for caffeine
measurement among premenopausal women in the BioCycle study. American
Journal of Epidemiology, 177(7), 690-699.
205 | P a g e
Schmid, S. M., Hallschmid, M., Jauch-Chara, K., Born, J., & Schultes, B. (2008). A single
night of sleep deprivation increases ghrelin levels and feelings of hunger in
normal‐weight healthy men. Journal of Sleep Research, 17(3), 331-334.
Schreiber, J. B., Nora, A., Stage, F. K., Barlow, E. A., & King, J. (2006). Reporting structural
equation modeling and confirmatory factor analysis results: A review. The
Journal of Educational Research, 99(6), 323-338.
Scully, M., Dixon, H., White, V., & Beckmann, K. (2007). Dietary, physical activity and
sedentary behaviour among Australian secondary students in 2005. Health
Promotion International, 22(3), 236-245.
Sejnowski, T. J., & Destexhe, A. (2000). Why do we sleep? Brain Research, 886(1), 208-
223.
Sekine, M., Yamagami, T., Handa, K., Saito, T., Nanri, S., Kawaminami, K., . . .
Kagamimori, S. (2002). A dose–response relationship between short sleeping
hours and childhood obesity: results of the Toyama Birth Cohort Study. Child:
Care, Health and Development, 28(2), 163-170.
Selwitz, R. H., Ismail, A. I., & Pitts, N. B. (2007). Dental caries. The Lancet, 369(9555), 51-
59.
Shapiro, C. M., Griesel, R. D., Bartel, P. R., & Jooste, P. L. (1975). Sleep patterns afted
graded exercise. Journal of Applied Physiology, 39(2), 187-190.
Shi, Z., McEvoy, M., Luu, J., & Attia, J. (2008). Dietary fat and sleep duration in Chinese
men and women. International Journal of Obesity, 32(12), 1835-1840.
Shilo, L., Sabbah, H., Hadari, R., Kovatz, S., Weinberg, U., Dolev, S., . . . Shenkman, L.
(2002). The effects of coffee consumption on sleep and melatonin secretion.
Sleep Medicine, 3(3), 271-273.
Shimazu, T. (1987). Neuronal regulation of hepatic glucose metabolism in mammals.
Diabetes/Metabolism Reviews, 3(1), 185-206.
Šidák, Z. (1967). Rectangular confidence regions for the means of multivariate normal
distributions. Journal of the American Statistical Association, 62(318), 626-633.
Siegel, J. M. (2005). Clues to the functions of mammalian sleep. Nature, 437(7063),
1264-1271.
Silber, B., & Schmitt, J. (2010). Effects of tryptophan loading on human cognition, mood,
and sleep. Neuroscience and Biobehavioral Reviews, 34(3), 387-407.
Silber, M. H., Ancoli-Israel, S., Bonnet, M. H., Chokroverty, S., Grigg-Damberger, M. M.,
Hirshkowitz, M., . . . Penzel, T. (2007). The visual scoring of sleep in adults.
Journal of Clinical Sleep Medicine, 3(2), 121-131.
Simon, S. L., Field, J., Miller, L. E., DiFrancesco, M., & Beebe, D. W. (2015). Sweet/dessert
foods are more appealing to adolescents after sleep restriction. PloS one, 10(2).
doi:10.1371/journal. pone.0115434.
Sleep Health Foundation. (2011). Sleep Tips for Children. Retrieved from
http://www.sleephealthfoundation.org.au/public-information/fact-sheets-a-
z/224-sleep-tips-for-children.html
Sleep Health Foundation. (2013). Caffeine, Food, Alcohol, Smoking and Sleep. Retrieved
from http://www.sleephealthfoundation.org.au/public-information/fact-sheets-
a-z/262-caffeine-food-alcohol-smoking-and-sleep.html
Slimani, N., Deharveng, G., Charrondière, R. U., van Kappel, A. L., Ocké, M. C., Welch, A.,
. . . Pala, V. (1999). Structure of the standardized computerized 24-h diet recall
206 | P a g e
interview used as reference method in the 22 centers participating in the EPIC
project. Computer Methods and Programs in Biomedicine, 58(3), 251-266.
Smedje, H., Broman, J., & Hetta, J. (2001). Associations between disturbed sleep and
behavioural difficulties in 635 children aged six to eight years: a study based on
parents' perceptions. European Child and Adolescent Psychiatry, 10(1), 1-9.
Smith, A. (2002). Effects of caffeine on human behavior. Food and Chemical Toxicology,
40(9), 1243-1255.
Smith, A. F., Jobe, J. B., & Mingay, D. J. (1991). Retrieval from memory of dietary
information. Applied Cognitive Psychology, 5(3), 269-296.
Smith, C. (2001). Sleep states and memory processes in humans: procedural versus
declarative memory systems. Sleep Medicine Reviews, 5(6), 491-506.
Spaeth, A. M., Dinges, D. F., & Goel, N. (2013). Effects of experimental sleep restriction
on weight gain, caloric intake, and meal timing in healthy adults. SLEEP, 36(7),
981-990.
Spiegel, K., Leproult, R., Colecchia, E., L'Hermite-Balériaux, M., Nie, Z., Copinschi, G., &
Van Cauter, E. (2000). Adaptation of the 24-h growth hormone profile to a state
of sleep debt. American Journal of Physiology-Regulatory, Integrative and
Comparative Physiology, 279(3), 874-883.
Spiegel, K., Tasali, E., Penev, P., & Cauter, E. (2004). Brief communication: sleep
curtailment in healthy young men is associated with decreased leptin levels,
elevated ghrelin levels, and increased hunger and appetite. Annals of Internal
Medicine, 141(11), 846.
Stavric, B., & Gilbert, S. G. (1990). Caffeine Metabolism - A problem in extrapolating
results from animal studies to humans Acta Pharmaceutica Jugoslavica, 40(3),
475-489.
Stavric, B., Klassen, R., Watkinson, B., Karpinski, K., Stapley, R., & Fried, P. (1988).
Variability in caffeine consumption from coffee and tea: possible significance for
epidemiological studies. Food and Chemical Toxicology, 26(2), 111-118.
Steinfeldt, L., Anand, J., & Murayi, T. (2013). Food Reporting Patterns in the USDA
Automated Multiple-Pass Method. Procedia Food Science, 2, 145-156.
Stickgold, R. (2001). Toward a cognitive neuroscience of sleep. Sleep Medicine Reviews,
5(6), 417-421.
Stickgold, R., Hobson, J. A., Fosse, R., & Fosse, M. (2001). Sleep, learning, and dreams:
off-line memory reprocessing. Science, 294(5544), 1052-1057.
Strecker, R. E., Morairty, S., Thakkar, M. M., Porkka-Heiskanen, T., Basheer, R., Dauphin,
L. J., . . . McCarley, R. W. (2000). Adenosinergic modulation of basal forebrain and
preoptic/anterior hypothalamic neuronal activity in the control of behavioral
state. Behavioural Brain Research, 115(2), 183-204.
Stutts, J. C., Wilkins, J. W., Osberg, J. S., & Vaughn, B. V. (2003). Driver risk factors for
sleep-related crashes. Accident Analysis & Prevention, 35(3), 321-331.
Subar, A. F., Kipnis, V., Troiano, R. P., Midthune, D., Schoeller, D. A., Bingham, S., . . .
Ballard-Barbash, R. (2003). Using intake biomarkers to evaluate the extent of
dietary misreporting in a large sample of adults: the OPEN study. American
Journal of Epidemiology, 158(1), 1-13.
Sullivan, B. L., Brown, J., Williams, P. G., & Meyer, B. J. (2008). Dietary validation of a
new Australian food-frequency questionnaire that estimates long-chain n-3
polyunsaturated fatty acids. British Journal of Nutrition, 99(03), 660-666.
207 | P a g e
Sutherland, G. R., & McNaughton, B. (2000). Memory trace reactivation in hippocampal
and neocortical neuronal ensembles. Current Opinion in Neurobiology, 10(2),
180-186.
Taheri, S., Lin, L., Austin, D., Young, T., & Mignot, E. (2004). Short sleep duration is
associated with reduced leptin, elevated ghrelin, and increased body mass index.
PLoS Med, 1(3), 210-217. doi: 10.1371/journal.pmed.0010062
Taren, D. L., Tobar, M., Hill, A., Howell, W., Shisslak, C., Bell, I., & Ritenbaugh, C. (1999).
The association of energy intake bias with psychological scores of women.
Tarokh, L., Saletin, J. M., & Carskadon, M. A. (2016). Sleep in adolescence: Physiology,
cognition and mental health. Neuroscience & Biobehavioral Reviews, 70, 182-
188.
Thompson, F. E., & Byers, T. (1994). Dietary assessment resource manual. Journal of
Nutrition, 124(Suppl.11), 2245-2317.
Thomson, T., & Subar, A. (2001). Nutrition in the Prevention and Treatment of Disease.
In C. J. Boushey, A. M. Coulston, C. L. Rock, & E. Monsen (Eds.), Dietary
Assessment Methodology. London: Academic Press.
Tirone, T. A., & Brunicardi, F. C. (2001). Overview of glucose regulation. World Journal of
Surgery, 25(4), 461-467.
Todd, K., Hudes, M., & Calloway, D. H. (1983). Food intake measurement: problems and
approaches. The American Journal of Clinical Nutrition, 37(1), 139-146.
Tyas, S. L., & Pederson, L. L. (1998). Psychosocial factors related to adolescent smoking:
a critical review of the literature. Tobacco Control, 7(4), 409-420.
Uddin, L. Q., Supekar, K., & Menon, V. (2010). Typical and atypical development of
functional human brain networks: insights from resting-state FMRI. Frontiers in
Systems Neuroscience, 4, 21.
Van Cauter, E., & Knutson, K. L. (2008). Sleep and the epidemic of obesity in children and
adults. European Journal of Endocrinology, 159(suppl 1), S59-S66.
van den Brandt, P. A., Goldbohm, R. A., Veer, P. V. T., Volovics, A., Hermus, R. J., &
Sturmans, F. (1990). A large-scale prospective cohort study on diet and cancer in
The Netherlands. Journal of Clinical Epidemiology, 43(3), 285-295.
Van Dongen, H. P., Maislin, G., Mullington, J. M., & Dinges, D. F. (2003). The cumulative
cost of additional wakefulness: dose-response effects on neurobehavioral
functions and sleep physiology from chronic sleep restriction and total sleep
deprivation. SLEEP, 26(2), 117-129.
van Staveren, W. A., Ocké, M. C., & de vries, J. H. M. (2012). Estimation of Dietary Intake
Present Knowledge in Nutrition (pp. 1012-1026): Wiley-Blackwell.
Vartanian, L. R., Schwartz, M. B., & Brownell, K. D. (2007). Effects of soft drink
consumption on nutrition and health: a systematic review and meta-analysis.
American Journal of Public Health, 97(4), 667-675.
Viera, A. J., & Garrett, J. M. (2005). Understanding interobserver agreement: the kappa
statistic. Family Medicine, 37(5), 360-363.
Walker, M. P. (2005). A refined model of sleep and the time course of memory
formation. Behavioral and Brain Sciences, 28(01), 51-64.
Walker, M. P. (2009). The role of sleep in cognition and emotion. Annals of the New York
Academy of Sciences, 1156(1), 168-197.
208 | P a g e
Walker, M. P., & Stickgold, R. (2006). Sleep, memory, and plasticity. Annual Review of
Psychology, 57, 139-166.
Walsh, L. J. (2000). Preventive dentistry for the general dental practitioner. Australian
Dental Journal, 45(2), 76-82.
Wang, A., Wang, S., Zhu, C., Huang, H., Wu, L., Wan, X., . . . He, L. (2016). Coffee and
cancer risk: A meta-analysis of prospective observational studies. Scientific
Reports, 6.
Warner, M. (2005). A jolt of caffeine, by the can. New York Times, C1.
Warzak, W. J., Evans, S., Floress, M. T., Gross, A. C., & Stoolman, S. (2011). Caffeine
consumption in young children. Journal of Pediatrics, 158, 508-509.
doi:http://dx.doi.org/10.1016/j.jpeds.2010.11.022
Watanabe, M., Wang, D.-H., Ijichi, A., Shirai, C., Zou, Y., Kubo, M., . . . Ogino, K. (2014).
The influence of lifestyle on the incidence of dental caries among 3-year-old
Japanese children. International Journal of Environmental Research and Public
Health, 11(12), 12611-12622.
Sciences and Nutrition. (In Press). doi: 10.1080/09637486.2016.1268102
Watson, J. F., Collins, C. E., Sibbritt, D. W., Dibley, M. J., & Garg, M. L. (2009).
Reproducibility and comparative validity of a food frequency questionnaire for
Australian children and adolescents. International Journal of Behavioral Nutrition
and Physical Activity, 6(1), 62.
Watson, N. F., Badr, M. S., Belenky, G., Bliwise, D. L., Buxton, O. M., Buysse, D., . . . Kushida, C. (2015).
Recommended amount of sleep for a healthy adult: a joint consensus statement of the American
Academy of Sleep Medicine and Sleep Research Society. J Clin Sleep Med, 11(6), 591-592.
Weikel, J., Held, K., Wichniak, A., Ising, M., Schmid, D., Mathias, S., . . . Steiger, A. (2003).
Ghrelin promotes slow-wave sleep in man. Pharmacopsychiatry, 36(5), 308.
Wein, E. E., Sabry, J., & Evers, F. (1990). Recalled estimates of food portion size. Journal
of the Canadian Dietetic Association, 51(3), 400-403.
Weiss, A., Xu, F., Storfer-Isser, A., Thomas, A., Ievers-Landis, C. E., & Redline, S. (2010).
The association of sleep duration with adolescents’ fat and carbohydrate
consumption. Sleep, 33(9), 1201-1209.
Weiss, R., Dziura, J., Burgert, T. S., Tamborlane, W. V., Taksali, S. E., Yeckel, C. W., . . .
Morrison, J. (2004). Obesity and the metabolic syndrome in children and
adolescents. New England Journal of Medicine, 350(23), 2362-2374.
Weissman, D. H., Roberts, K., Visscher, K., & Woldorff, M. (2006). The neural bases of
momentary lapses in attention. Nature Neuroscience, 9(7), 971-978.
Westerlund, L., Ray, C., & Roos, E. (2009). Associations between sleeping habits and food
consumption patterns among 10–11-year-old children in Finland. British Journal
of Nutrition, 102(10), 1531-1537.
Whitney, E. N., Rolfes, S. R., Crowe, T., Cameron-Smith, D., & Walsh, A. (2011).
Understanding Nutrition: Cengage Learning.
Whittier, A., Sanchez, S., Castaneda, B., Sanchez, E., Gelaye, B., Yanez, D., & Williams, M.
A. (2014). Eveningness chronotype, daytime sleepiness, caffeine consumption,
and use of other stimulants among Peruvian university students. Journal of
Caffeine Research, 4(1), 21-27. doi:http://dx.doi.org/10.1089/jcr.2013.0029
209 | P a g e
Wiggs, L., & Stores, G. (1996). Severe sleep disturbance and daytime challenging
behaviour in children with severe learning disabilities. Journal of Intellectual
Disability Research, 40(6), 518-528.
Wilhelm, B., Stuiber, G., Ludtke, H., & Wilhelm, H. (2014). The effect of caffeine on
spontaneous pupillary oscillations. Ophthalmic and Physiological Optics : the
Journal of the British College of Ophthalmic Opticians (Optometrists), 34(1), 73-
81.
Willett, W. (2012). Nutritional Epidemiology: Oxford University Press.
Willett, W. C. (1994). Future directions in the development of food-frequency
questionnaires. The American Journal of Clinical Nutrition, 59(1), 171S-174S.
Wilson, M. A., & McNaughton, B. L. (1994). Reactivation of hippocampal ensemble
memories during sleep. Science, 265(5172), 676-679.
Wolf, A. M., & Colditz, G. A. (1998). Current estimates of the economic cost of obesity in
the United States. Obesity Research, 6(2), 97-106.
Wolfson, A. R., & Carskadon, M. A. (1998). Sleep schedules and daytime functioning in
adolescents. Child Development, 69(4), 875-887.
Wolraich, M. L., Stumbo, P. J., Milich, R., Chenard, C., & Schultz, F. (1986). Dietary
characteristics of hyperactive and control boys. Journal of the American Dietetic
Association, 86(4), 500-504.
Wood, A. (2016, March 11). Sugar is sweet- and also a killer, Newspaper. The Daily
Telegraph. Retrieved from http://www.dailytelegraph.com.au/news/nsw/sugar-
is-sweet--and-also-a-killer/news-story/1e284dff95f2fa6867dc6c4f3dd02f3d
World Health Organization. (2015). Guideline: sugars intake for adults and children.
Wright, K. P., Hull, J. T., & Czeisler, C. A. (2002). Relationship between alertness,
performance, and body temperature in humans. American Journal of Physiology-
Regulatory, Integrative and Comparative Physiology, 283(6), R1370-R1377.
Wyatt, J. K., Ritz-De Cecco, A., Czeisler, C. A., & Dijk, D.-J. (1999). Circadian temperature
and melatonin rhythms, sleep, and neurobehavioral function in humans living on
a 20-h day. American Journal of Physiology-Regulatory, Integrative and
Comparative Physiology, 277(4), R1152-R1163.
Xie, L., Kang, H., Xu, Q., Chen, M. J., Liao, Y., Thiyagarajan, M., . . . Iliff, J. J. (2013). Sleep
drives metabolite clearance from the adult brain. Science, 342(6156), 373-377.
Yang, A., Palmer, A. A., & de Wit, H. (2010). Genetics of caffeine consumption and
responses to caffeine. Psychopharmacology, 211(3), 245-257.
Yuksel, H., Sogut, A., Yilmaz, O., Demet, M., Ergin, D., & Kirmaz, C. (2007). Evaluation of
sleep quality and anxiety–depression parameters in asthmatic children and their
mothers. Respiratory Medicine, 101(12), 2550-2554.
Zhang, J., Lam, S. P., Li, S. X., Li, A. M., Kong, A. P., & Wing, Y. K. (2014). Restless legs
symptoms in adolescents: epidemiology, heritability, and pubertal effects.
Journal of Psychosomatic Research, 76(2), 158-164.
doi:http://dx.doi.org/10.1016/j.jpsychores.2013.11.017
Zinkhan, M., Berger, K., Hense, S., Nagel, M., Obst, A., Koch, B., . . . Young, P. (2014).
Agreement of different methods for assessing sleep characteristics: a comparison
of two actigraphs, wrist and hip placement, and self-report with
polysomnography. Sleep Medicine, 15(9), 1107-1114.
210 | P a g e
Zohar, D., Tzischinsky, O., Epstein, R., & Lavie, P. (2005). The effects of sleep loss on
medical residents' emotional reactions to work events: a cognitive-energy model.
SLEEP, 28(1), 47-54.
Zulkifli, S. N., & Yu, S. (1992). The food frequency method for dietary assessment.
Journal of the American Dietetic Association, 92(6), 681-685.
Zunhammer, M., Eichhammer, P., & Busch, V. (2014). Sleep Quality during Exam Stress:
The Role of Alcohol, Caffeine and Nicotine. PLoS ONE, 9 (10) (no
pagination)(e109490). doi:http://dx.doi.org/10.1371/journal.pone.0109490
Zylber‐Katz, E., Granit, L., & Levy, M. (1984). Relationship between caffeine
concentrations in plasma and saliva. Clinical Pharmacology and Therapeutics,
36(1), 133-137.
211 | P a g e
Appendices
212 | P a g e
Appendix A
Correspondence with Journal of Clinical Sleep Medicine:
Hello Ms. Watson:
Thank you for your inquiry and congratulations on the upcoming publication of your JCSM
article. Since this is your own work, we do give you permission to use it for your thesis. You can
cite it this way:
Watson, E., Banks, S., Coates, A., and Kohler, M. The Relationship between Caffeine, Sleep and
Behavior in Children, J Clin Sleep Med. 2017 (in press).
In several weeks, the citation should be available ahead of print on PubMed. Right now, the
paper is included on our accepted papers list on the JCSM Website. Here is the direct link to that
page: http://www.aasmnet.org/jcsm/AcceptedPapers.aspx
I hope this helps!
Best regards and Happy New Year!
Rosanne
Rosanne Money
Journal Editorial Office
www.aasmnet.org
Follow Us: Facebook | Twitter | LinkedIn

Correspondence with Nutrients:
Dear Dr. Watson,
Thank you very much for your interest in said material.
All MDPI journals are Open Access and subject to the Creative Commons Attribution
License (CC BY). The CC BY permits unrestricted use, distribution, and reproduction of
the material in any medium, even commercially, *provided the original work is properly
cited*. You do not have to pay anything for permission.
For more information on the CC BY License, please see here:

https://creativecommons.org/licenses/by/4.0/legalcode
Best regards,
Larissa Eisele
Correspondence with International Journal of Food Sciences and Nutrition:
Thank you for your email.
As the author of the original article, you have the right to include the article in a thesis or
dissertation that is not to be published commercially, provided that acknowledgement in
the journal is noted.
This is an Accepted Manuscript of an article published by Taylor & Francis in

Journal of XXX on DATE, available online: http://wwww.tandfonline.com/DOI
Best wishes
Sherry Howard
On behalf of Author Services
Appendix B
Centre for Sleep Research

Main Project: A questionnaire investigating the interactions between diet, sleep and behaviour in
children
Sub Study: Validation of a questionnaire to monitor consumption of caffeine: pilot study
Researchers: Emily Watson, Dr Mark Kohler, Dr Siobhan Banks & Dr Alison Coates
Dear Sir/Madam,
You have been invited to participate in a research study to validate a questionnaire to measure
caffeine consumption and to look at the relationship between habitual caffeine consumption and
sleep quality. Before agreeing to participate in the study, it is important that you read and
understand the following explanation of the questionnaire.
What is the study about?
The main study (A questionnaire investigating the interactions between diet, sleep and behaviour in
children) is investigating the relationship between dietary consumption of a range of foods (including
caffeine intake), sleep and cognitive performance in children. To do this, we have developed a
questionnaire to capture average intake of caffeine over a week. Before using this questionnaire it is
important that we first validate it against a robust dietary assessment method and check that it is
reliable. This particular study, therefore, is piloting the caffeine questionnaire to determine if the
questionnaire is valid (i.e. compares with actual caffeine intake), reliable (i.e. repeatable) and easy to
use. It will be piloted on two population samples, both adults and children.
What does the study involve?
If you would like to participate in the study, you will be invited to complete a short questionnaire
asking about sleep habits and history of medical conditions which may impact the outcomes of
interest in this study. If you are eligible and you chose to participate in this study you will be asked
to:
 Complete a food diary recording all beverages and chocolate you consume for seven
consecutive days. This involves recording the type, time, amount and brand of drink or
chocolate consumed.
 Complete the caffeine questionnaire we have developed. This questionnaire will be

completed twice over the two days post completion of the food diary. This information
gathered from this questionnaire is regarding your caffeine consumption over the previous
week.
 Adults participating are also asked to complete a Pittsburgh Sleep Index Questionnaire
(PSIQ). The PSIQ will be completed only once, the day after the completion of food diary
with the caffeine questionnaire. It will collect information regarding sleep quality and
disturbances.
The study investigators are PhD candidate Miss Emily Watson, Dr. Mark Kohler, Dr. Siobhan Banks &
Dr. Alison Coates. It is expected that the food diary and questionnaires will take approximately 10
minutes per day.
Do I have to participate?
No. As with all research at the University, participation is completely voluntary. If you do decide to
participate, you are free to withdraw at any stage, without prejudice.
Who can participate?
You can participate if you are:
 An adult over the age of 18 OR a prepubescent child aged between ages 8-12.
 Not involved in paid employment that includes night work i.e. paid after 7pm
 Not clinically diagnosed with any sleep disorders
 Not taking any prescribed medications or supplements to aid sleep or alertness
 Consuming caffeinated beverages or foods at least once a day.
 Not suffering from any conditions that limits caffeine intake e.g. pregnancy
Will my information be kept private?
After completing the screening questionnaire, each participant will be given an identification number
and personal identifiable information will be removed and stored separately to your questionnaire
information. You will not be identified in any way in the analysis of the data or when the results are
published. Also, information on the questionnaire will be grouped for reporting and individual
responses will not be presented in any way. All information supplied will be locked away at a secured
location at the Centre for Sleep Research, University of South Australia for 5 years and the researcher
will not supply this information to the public without explicit permission. Every effort is made so that
the information you supply will remain completely confidential.
Can I be informed of the results of mine or my child’s questionnaire?
The results from this study will be used to further future use of the questionnaire. Therefore no
individual results will be available, however if you would like to see a summary of all the information
gained, please provide an email address.
How do I know that this research is legitimate?
The University of South Australia’s Human Research Ethics Committee has reviewed this study.
Should you wish to discuss the project with someone not directly involved, in particular in relation to
matters concerning policies, information about the conduct of the study, or your rights as a
participant, please contact the UniSA Ethics Officer, Ms Vicki Allen on 8302 3118; fax 8302 3921;
email: Vicki.allen@unisa.edu.au
Yes, I or my child would like to participate. What do I do now?
Your involvement in this study would be greatly appreciated. If you are interested please complete
the attached questionnaire. If you would like more information please contact either Miss Emily
Watson (8302 2453 - emily.watson@mymail.unisa.edu.au), Dr Siobhan Banks, (8302 1712 -
siobhan.banks@unisa.edu.au) Dr. Mark Kohler (8302 4919 - mark.kohler@unisa.edu.au) or Dr.
Alison Coates (8302 2313 – alison.coates@unisa.edu.au).
Appendix C

Main Project: A questionnaire investigating the interactions between diet, sleep and behaviour in
children
Sub Study: A questionnaire investigating the consumption of caffeine: pilot study

Researchers: Emily Watson, Dr Mark Kohler, Dr Siobhan Banks & Dr Alison Coates
 I have read the participant Information Sheet, and the nature and the purpose of the
research project have been explained to me. I understand and agree for myself to take part.
 I understand the purpose of the research project, the benefits and risks associated with
participation and what commitment is required.
 I understand that I can withdraw from the study at any stage and that this will not affect my
status now or in the future.
 I understand that while information gained during the study may be published, I will not be
identified and all personal results will remain confidential.
 I am aware that the information collected during this study may be used for future research
but at no time will participants be identified.
 I understand data will be stored in a locked filing cabinet at the University of South Australia
that can be accessed only by the researchers involved in this study.
 I confirm that I am over 18 years of age
 This project has been approved by the University of South Australia’s Human Research Ethics
Committee
 I understand that I may not directly benefit from taking part in the study.
PARTICIPANT NAME:
PARTICIPANT SIGNATURE:
DATE:
Researcher Certification:
I have explained the study to subject and consider that he/she understand what is involved
RESEARCHER NAME
RESEARCHER SIGNATURE: DATE:

Appendix D
Screening Questionnaire
In this questionnaire general questions are asked about medial conditions and medications
related to sleep and sleep habits. All information will be kept strictly confidential. If you have any
concerns about the questions in this form or have difficulty in answering any questions please do
not hesitate to contact, Miss Emily Watson (8302 2453 - emily.watson@mymail.unisa.edu.au),
Dr Siobhan Banks, (8302 1712 - siobhan.banks@unisa.edu.au) Dr. Mark Kohler (8302 4919 -
mark.kohler@unisa.edu.au) or Dr. Alison Coates (8302 2313 – alison.coates@unisa.edu.au).
1. Today’s Date: ___________________________________
2. Full name: _____________________________________________
3. Gender: Male/Female (please circle one) VALIDATION STUDY ONLY
3. Are you over the age of 18? □ Yes. □ No.
If yes, what is your DOB? _________________ VALIDATION STUDY ONLY
If no, you are not eligible. Thank you for your time
Eligibility Assessment
PLEASE COMPLETE by ticking the appropriate answer:
Do you have any sleep related conditions (e.g. insomnia, night terrors, sleep apnea)
□ Yes. If yes, please complete the table below.
□ No. If no, go to Q4.
Sleep related condition name Diagnosis Date Frequency of occurrence
e.g Insomnia June 2010 Most nights
Are you proficient in reading and writing in English? VALIDATION STUDY ONLY
□ Yes. □ No.
Are you regularly taking any prescribed or over the counter medication to help you sleep (e.g.
sleeping tablet, herbal supplement) or help alertness?
□ Yes. If yes, please complete the table below.
□ No. If no, go to Q5.
Medication name Dose Reason for Use Frequency taken
e.g Nature’s own

500 mg Once/day
complete sleep
In the last month, have you been involved in night work i.e. attended paid work after 7pm?
□ Yes. □ No.
If yes, you are not eligible. Thank you for your time
Do you consume a caffeinated beverage at least once per day? (e.g. coffee, tea, coca cola,
mountain dew, iced coffee or tea.)
□ Yes. □ No.
Do you have any conditions that affect caffeine intake e.g. pregnancy?
□ Yes. □ No.
If yes, you are not eligible. Thank you for your time
Study Participation Requirements
Would you be able to keep a food diary of your beverages consumed (other than water) and
chocolate eaten for seven consecutive days, taking approximately 10 minutes per day?
□ Yes. □ No.
Would you be able to meet with the researchers to complete additional questionnaires on the
two days post food diary completion – expected completion time 10minutes?
□ Yes. □ No.
Thank you for your co-operation
If you have any questions regarding the study please do not hesitate to contact, Miss Emily
Watson (8302 2453 - emily.watson@mymail.unisa.edu.au), Dr Siobhan Banks, (8302 1712 -
siobhan.banks@unisa.edu.au) Dr. Mark Kohler (8302 4919 - mark.kohler@unisa.edu.au) or Dr.
Alison Coates (8302 2313 – alison.coates@unisa.edu.au).
Appendix E
ID: PCQ________
Appendix F
ID: PCQ________
ID: PCQ________
ID: PCQ________
ID: PCQ________
ID: PCQ________
Date: VCQ:
Appendix G
CAFFEINE INTAKE QUESTIONNAIRE
Please think about the foods and drinks you have had over the last week.
Remember there are no right or wrong answers.
Place a cross, X, through the pictures of the drinks not drunk and answer all questions
below the drinks you have consumed over the last week.
EXAMPLES:
Did not drink Red Bull over the last Had two small cans of Red Bull over the
week last week
Red Bull
Red Bull
TM
TM
All Red Bull Flavours

What size did you drink mostly?
Small Can (250mL) Medium Can
(355mL) × Small Can (250mL) Medium Can
(355mL)
Large Can (473mL) Bottle (330mL)
Large Can (473mL) Bottle (330mL)
Glass
Glass
How many times over the last week did you drink
Red Bull? How many times over the last week did you drink
Red Bull?
1 2
1 × 2
3 4
3 4
Other ______________________
(please specify) Other ______________________
(please specify)
Date: VCQ:
Using the previous examples as a guide please answer the following questions, remembering to only
consider what you have had over the last week:
Red Bull V
TM
TM
What type of V drink did you drink the

most?
All V drinks Pocket Rocket

Small Can Medium Can What size did you drink?
(250mL) (355mL) Small can Large can
(250mL) (500mL)
Large Can Bottle (330mL)
(473mL) Bottle (350mL) Glass
Glass Bottle (500mL) Pocket Rocket
How many times over the last week did you

drink Red Bull? How many times over the last week did
1 2 you drink it?
1 2
3 4
3 4
Other ______________________
(please specify) Other ______________________
(please specify)
Place an X through the pictures of the drinks not drunk and answer questions about the drinks you have consumed over the last week.
Date: VCQ:
Red Eye
Rockstar
TM
TM
What type of Red Eye drink did you drink?

Platinum
All Rockstar Flavours
Other ______________________
(please specify
What size did you drink?
What size did you drink? Small Can (355mL) Medium Can/Bottle (500mL)
Bottle (330mL) Glass Glass Large Can (710mL)
How many times over the last week did you drink it?
How many times over the last week did
you drink it? 1 2
1 2
3 4
3 4 Other ______________________
Other ______________________ (please specify)
(please specify)
Monster
Mother
TM
TM
What type of Monster drink did you drink?

All Mother Flavours ANY monster flavour not Zero or sugar free
including sugar free
Small Can
(250mL) Small can (340mL) Large can (500mL)
Large Can Other_________

(500mL) Resealable (550mL) Glass
drink it? How many times over the last week did you drink it?
1 2
3 4 1 2
Other ______________________ 3 4
(please specify)
Other ______________________
(please specify)
Date: VCQ:
Powerade Fuel Energy

Wicked
TM TM
What size did you drink? What size did you drink?
Large Can Glass Can Glass

(500mL) How many times over the last week did
you drink it? 1 2
1 2 3 4
3 4 Other ______________________
(please specify)
Other ______________________
(please specify)
Other energy drinks:

Other energy drinks consumed in the last week
that have not been mentioned (e.g. 28 black,
pink):
1.
2.
Approximate size of the drink:
1.
2.
it?
1.
2.
Date: VCQ:
Mountain Dew Coke

TM TM

Can (375mL) 600mL Bottle
What type of Coca Cola drink did you drink over
Glass the last week?
Diet Coke/ Diet Original Coke/ Coke
drink it?
Vanilla Coke Zero/ Vanilla
1 2 Coke/Cherry Coke
3 4 What size did you most often drink?
200mL Can 375mL Can
Other ______________________
(please specify)
450mL Bottle 600ml Bottle
Glass
How many cans/bottles/glasses over the last week

did you drink?
Pepsi 1 2
TM
3 4
Other ______________________
(please specify)
What type of Pepsi drink did you drink over

the last week? Other caffinated carbonated
Original Diet drinks:
Other energy drinks consumed in the last week
Max Max Kick
that have not been mentioned (e.g. Dr Pepper):
What size did you most often drink? 1.

200mL Can 375mL Can
450mL Bottle 600ml Bottle 2.
Glass
Approx. size of the drink:
How many cans/bottles/glasses over the last
1.
week did you drink?
1 2
2.
3 4 How many times over the last week did you drink
Other ______________________ it?
(please specify)
1.
2.
Date: VCQ:
Hot Coffee
Iced Coffee
TM
What type of hot coffee drink do you drink the

Over the last week what type of Iced Coffee drink most?
did you drink the most? Espresso (For example: Flat
Iced Coffee Brand: white, cappuccino, short black,
____________________________ latte)
(i.e. Farmers Union, Dare, Feel Good)
Instant – Brand ____________
Small Carton Medium Carton (600mL)
(375mL) Other ______________________
(please specify)
Glass Large Carton (750mL)
it? 1 2
1 2
3 4
3 4
Other ______________________
Other ______________________ (please specify)
(please specify)
Iced Tea Hot Tea
Over the last week what type of Iced Tea drink did Over the last week what type of hot tea did you
you drink the most? drink the most?
Lipton Homemade Green Black Herbal/Fruit
Nerada Nestea Type of tea ________________________________
Snapple How many cups over the last week did you drink?
Iced Tea Flavour ____________________________ 1 2
What size did you drink? 3 4

Bottle (500mL) Can (250mL) Other ______________________
(please specify)
Glass
it?
1 2
3 4
Other ______________________
(please specify)
Date: VCQ:
Iced Chocolate
TM
Over the last week what type of Iced Chocolate

drink did you drink the most?
Iced Chocolate Brand:

____________________________
(For Example: Farmers Union, Dare, Feel Good)
Glass Small Carton (375ml)

Large Carton (600ml)
it?
1 2
3 4
Other ______________________
(please specify)
Hot Chocolate
Over the last week what type of Hot Chocolate

drink did you drink the most?
Hot Chocolate Brand:
____________________________
(For example: Milo, Cadbury, Ovaltine)
How many teaspoons?
1-2 3-4
5+ Don’t Know
How many glasses over the last week did you
drink?
1 2
3 4
Other ______________________
(please specify)
Place an X through the pictures of the drinks not drunk and answer questions about the drinks you have consumed over the last
week.
Date: VCQ:
Chocolate
If you have some chocolate in the last week please fill in the table. Using
the examples below as a guide please be as specific as possible:
Was the chocolate

Brand and flavour of Amount of chocolate
MILK/DARK/WHITE
chocolate: eaten:
(please select one)
For example:
1. Cadbury Freddo Frog 1 Solid milk chocolate
2. Cadbury Crunchie
4 rows of 220g family block milk
Chocolate
THANK – YOU
Thank you for taking the time to complete this questionnaire.
Appendix H
Chocolate and Beverage Record

Instructions
To allow us to assess your beverage and chocolate intake we will need you
to keep a record of everything you drink (including water) and all chocolate
you eat for seven consecutive days. It is important that these days be as
normal as possible. Please do not change any eating patterns on the days
that you record. Remember we are only interested in your usual dietary
intake, not the perfect diet!
If you have any question regarding how to complete the food record, please
do not hesitate to contact the Emily Watson on 83022453
General Guidelines:
 Please keep the diary for seven consecutive days.

 Write down all food containing chocolate eaten both at home and away
from home.
 Beverages of all kinds, including water and alcohol, must be recorded.
For juices and cordials please record how these were made (strength).
For packaged foods record the brand name and the weight or volume.
 For homemade foods please provide the recipe with all the ingredients
and number of serves made and the portion or fraction of the recipe you
ate
 Record amounts (in mL, L and grams) as accurately as possible
 If you are out and unable to measure the food, describe what is eaten as
fully as possible and make the best guess you can for the quantity.
EXAMPLE:
Day Number: ONE Day MONDAY Date: 1/10/08
Record ALL food with chocolate and drink consumed during the next seven
days.
Please remember to record:
 TIME OF CONSUMPTION
 TYPE OF FOOD OR DRINK (eg. Chocolate chip cookie or glass of milk)
 QUANTITY/VOLUME OF FOOD/DRINK (eg. 4 chocolate chip cookies or 500ml
glass of milk)
 BRAND NAMES (eg Coles, Nestle, Pura, Nescafe, Coopers)
FOOD TIME QUANTITY/VOLUME DETAILS OF FOOD & DRINK

CONSUMED CONSUMED
EARLY 7am 250ml Water
MORNING 7.30 200ml Nescafe ‘Indriya’ Soy Milk Latte
BREAKFAST 8am 250ml Water
DURING 10am 4 Coles brand chocolate chip cookies

MORNING
MIDDAY 12pm 200ml Nescafe ‘Indriya’ Soy Milk Latte

One row Cadbury crunchie 210g block of chocolate
DURING 3pm 400ml Coca cola

AFTERNOON
EVENING 2pm 250ml Twinings English breakfast tea – brewed for

MEAL 25secs
DURING 10pm 200ml Milk with soy milk

EVENING /
SUPPER
Appendix I
PITTSBURGH SLEEP QUALITY INDEX (PSQI)
The following questions relate to your usual sleep habits during the past month only. Your
answers should indicate the most accurate reply for the majority of days and nights in the past
month. Please answer all questions.
1. During the past month, when have you usually gone to bed at night?
USUAL BED TIME _______________
2. During the past month, how long (in minutes) has it usually take you to fall asleep each
night?
NUMBER OF MINUTES _____________
3. During the past month, when have you usually gotten up in the morning?
USUAL GETTING UP TIME ___________
4. During the past month, how many hours of actual sleep did you get at night? (This may
be different than the number of hours you spend in bed.)
HOURS OF SLEEP PER NIGHT ______________
For each of the remaining questions, circle the one best response. Please answer all
questions.
5. During the past month, how often have you had trouble sleeping because you….
(a) Cannot get to sleep within 30 minutes
Not during the Less than once a Once or twice a Three or more
past month week week times a week
(b) Wake up in the middle of the night or early morning
(c) Have to get up to use the bathroom
(d) Cannot breathe comfortably
(e) Cough or snore loudly
(f) Feel too cold
(g) Feel too hot
(h) Had bad dreams
(i) Have pain
(j) Other reason(s), please describe

______________________________________________________________________________
6. During the past month, how would you rate your sleep quality overall?
Very Good Fairly Good Fairly Bad Very Bad
7. During the past month, how often have you taken medicine (prescribed or ‘over the
counter’ to help you sleep?
Not during the past Less than once a Once or twice a week Three or more times
month week a week
8. During the past month, how often have you had trouble staying awake while driving,
eating meals or engaging in social activity?
Not during the past Less than once a Once or twice a week Three or more times
month week a week
9. During the past month, how much of a problem has it been for you to keep up enough
enthusiasm to get things done?
No problem at all Only a very slight Somewhat of a A very big problem
problem problem
10. Do you have a bed partner or roommate?
No bed partner or roommate __________

Partner/roommate in other room __________
Partner in same room, but not same bed __________
Partner in same bed __________
If you have a roommate or bed partner, ask him/her how often in the past month you have
had
a) Loud snoring
Not during the past Less than once per Once or twice a week Three or more times
month week a week
b) Long pauses between breaths while asleep
month week a week
c) Legs twitching or jerking while you sleep
month week a week
d) Episodes of disorientation or confusion during sleep
month week a week
e) Other restlessness while you sleep
Please
describe_______________________________________________________________________
month week a week
Appendix J

Project Title: A questionnaire investigating the interactions between diet, sleep and behaviour in children
Researchers: Emily Watson, Dr. Mark Kohler, Dr. Siobhan Banks and Dr. Alison Coates
INFORMATION SHEET
Dear Parent/Guardian,
You and your child have been invited to participate in a research questionnaire investigating the
relationship between child diet, sleep and behaviour. Before agreeing to participate in the study, it is
important that you read and understand the following explanation of the questionnaire.
If you have read the information below and would like your child to take part please complete the following
steps:
1. Please sign and date the consent form (coloured form).

2. Complete the parent participant booklet about your child (this will take approximately 20mins).
3. Either complete the Tanner stage questionnaire for your child or give it to your child to complete;
ensuring appropriate gender is completed.
4. Place all information back in the prepaid envelope and send it in the post back to University of
South Australia.
What is the study about?
Sleep is very important to our health and well-being. Sleep is also important for normal daytime learning,
better school performance, physical health and emotional health. Currently, the relationship between diet
and sleep is of increased interest to researchers, due to the rise in obesity. The relationship is of particular
interest in children as they need adequate sleep to grow, develop and function well.
The questionnaire will ask you as a parent about your observations of your child’s diet, sleep and
behaviour. The questionnaire will also ask you about your beliefs about the connection between diet and
sleep. The results of the questionnaires will help us to understand the relationship between diet, sleep and
behaviour. This information will be useful for improving child health and performance.
What are the benefits to me and my child?
In addition to the parental questionnaires we will be conducting an interactive research activity during class
time at your child’s school and we have planned some informative activities that discuss the importance of
sleep and nutrition. The University of South Australia are world leaders in research in both sleep and
nutrition and so you can be confident that the information presented will be based on the latest scientific
evidence. From the information we collect, you can also receive a summary of your child’s sleep and
dietary patterns. All children participating in this research will also be entered into the draw to win one of
five IPod shuffles.
Does my child have to participate?
No. As with all research at the University, participation is completely voluntary. If you do decide to
participate, you and your child are free to withdraw at any stage, without prejudice.
Who can participate?
You can participate if your child is:
 Between 8-12 years of age

 In good health
 Not clinically diagnosed with sleep or behavioural disorders, e.g. sleep apnoea, ADHD
 Not diabetic
Will my information be kept private?
After completing the questionnaire, it will be given an identification number and personal identifiable
information will be removed. Neither you nor your child will be identified in any way in the analysis of the
data or when the results are published in scientific journals. Also, information on the questionnaire will be
grouped for reporting and individual responses will not be presented in any way. All information supplied
will be locked away at a secured location at the Centre for Sleep Research, University of South Australia, for
5 years and the researcher will not supply this information to the public without explicit permission. Every
effort is made so that the information you supply will remain completely confidential.
Can I be informed of the results of my child’s questionnaire?
Yes. A summary of the results can be provided to you at the completion of the study if you indicate at the
end of the questionnaire together with an email address.
How do I know that this research is legitimate?
The University of South Australia’s Human Research Ethics Committee has reviewed this study.
Furthermore, this research has been reviewed by the Department of Education and Child Development and
South Australia Commission of Catholic Schools. Should you wish to discuss the project with someone not
directly involved, in particular relation to matters concerning policies, information about the conduct of the
study, or your rights as a participant, please contact the UniSA Ethics Officer, Ms Vicki Allen on 8302 3118;
fax 8302 3921; email: Vicki.allen@unisa.edu.au.
Your involvement in this study would be greatly appreciated. If you are interested please complete the
included questionnaire and either send it back to school with your child or mail it in the supplied envelope.
If you would like more information please contact either Miss Emily Watson (8302 2453 -
emily.watson@mymail.unisa.edu.au), Dr Siobhan Banks, (8302 1712 - siobhan.banks@unisa.edu.au) Dr.
Mark Kohler (8302 4919 - mark.kohler@unisa.edu.au) or Dr. Alison Coates (8302 2313 -
alison.coates@unisa.edu.au).
Appendix K

Project Title: A questionnaire investigating the interactions between diet, sleep and behaviour in
children
Researchers: Emily Watson, Dr. Mark Kohler, Dr. Siobhan Banks & Dr. Alison Coates
CONSENT FORM
 I have read the participant and parent/guardian Information Sheet, and the nature and the
purpose of the research project have been explained to me. I understand and agree for my child
and myself to take part.
 I understand the purpose of the research project, the benefits and risks associated with
participation and what commitment is required.
 I understand that I/my child can withdraw from the study at any stage and that this will not affect
their status now or in the future.
 I understand that while information gained during the study may be published, I/my child will not
be identified and all personal results will remain confidential.
 I am aware that the information collected during this study may be used for future research but at
no time will participants be identified.
 I understand data will be stored in a locked filing cabinet at the University of South Australia that
can be accessed only by the researchers involved in this study.
 I confirm that I am over 18 years of age and am the parent/guardian of the child named on this
consent form.
 This project has been approved by the University of South Australia’s Human Research Ethics
Committee
 I understand that I may not directly benefit from taking part in the study.
NAME OF CHILD:
NAME OF PARENT/GUARDIAN:
SIGNATURE:
DATE:
Appendix L
Participants
Participants
Questionnaire Study
Investigating the relationship between
diet, sleep and behaviour in children
Needed!
We are looking for volunteers to complete a questionnaire study. A questionnaire
pack will go home with your child this week.
You can participate if you are a parent/legal guardian with a child aged 8-12
years who is in good health, has no clinically diagnosed sleep or behavioural
disorders (e.g. sleep apnoea, ADHD) and has not been diagnosed with diabetes.
Both the child and a parent/legal guardian are required to complete a battery of
questionnaires, each taking approximately 15-20mins. The questionnaires then need
to be returned to Emily Watson at University of South Australia in a supplied pre-
paid envelope.
All questionnaires that are correctly filled will go in the running to win one
of five iPod Shuffle’s and all participants can receive a copy of their results
for their diet and sleep questionnaires.
The University of South Australia’s Human Research Ethics Committee and the Department for Education
and Child Development have reviewed this study.
Emily Watson
emily.watson@mymail.unisa.edu.au
8302 2453
facebook.com/questionnairestudy
Appendix M
Section One: General Information

What is today’s date (day/month/year)? _______/__________/________
What is your child’s date of birth (day/month/year)? _______/__________/________
What is your child’s gender (please circle)? (1) Male (2) Female
Number of children in the immediate family? ___________
Birth order of the child (eldest, 2nd, youngest, only child etc)? ______________________
Child’s ethnicity (please circle)? (1) Caucasian (2) Indigenous (3) Asian (4) Other (please specify):
Australian
Is English the predominate language spoken at

home? Yes No Please Specify__________________
Does your child (please circle)? (1) Sleep alone (2) Share a bedroom (3) Other (please specify):
Your child’s bedroom is (please circle)? (1) (2) (3) (4) (5) (6) (7)
Not at all noisy Very noisy
What is your relationship to the child (please circle)? (a) (b) (c) Other (please specify):
Mother Father
What is the mother/female caregiver’s date of birth (day/month/year)? ______/__________/________
What is the mother’s highest level of (1) (2) (3) (4) (5) (6)
education (please circle)? Primary High Bachelor TAFE Honours Post
School School Degree Graduate
What is the mother’s occupation? ____________________
What is the father/male caregiver’s date of birth (day/month/year)? ______/__________/________

(1) (2) (3) (4) (5) (6)
What is the father’s highest level of Primary High Bachelor TAFE Honours Post
education (please circle)? School School Degree Graduate
What is the father’s occupation? ____________________
Do either parent’s occupations require shift work (please circle)? Yes No
If yes which parent? _________________________________
What is your postcode? ___________________________________
Has your child been clinically diagnosed with ANY other behavioural Yes No
disorders? e.g. ADHD (please circle)
If yes please specify _____________________
Has your child ever been clinically diagnosed with any sleep disorders Yes No
(is yes please circle)?
if yes please specify __________________
Has your child been diagnosed with diabetes (please circle)? Yes No
What is your child’s weight in kg? _______________
What is your child’s height in cm? _________________

Appendix N
SECTION 2: TANNER STAGE QUESTIONNAIRE (MALE)

Please answer the questions below on behalf of your child, or give to child to complete. It will be
kept completely confidential.
The questions below are about changes that may be happening to your body.
These changes are normal and happen to young people at different ages.
Please answer the following questions by placing an  in the appropriate box that best describes
changes to your body that you have noticed. If you do not understand a question or do not know
the answer, just put an X next to “I don’t know”. Do your best to answer carefully.
Definitely
complete
started
started
started
Seems
I don’t
Barely
know
Not
1. Would you say that your growth in height

has:
2. Would you say that your body hair growth

has:
Body hair" means hair any place other than your
head, such as under your arms
3. Have you noticed any skin changes,

especially pimples?
4. Have you noticed a deepening of your

voice?
5. Have you begun to grow hair on your face?

SECTION 2: TANNER STAGE QUESTIONNAIRE (FEMALE)
Please answer the questions below on behalf of your child, or give to your child to complete. It
will be kept completely confidential.
The questions below are about changes that may be happening to your body. These changes are
normal and happen to young people at different ages.
Please answer the following questions by placing an  in the appropriate box that best
describes changes to your body that you have noticed. If you do not understand a question or
do not know the answer, just put an X next to “I don’t know”. Do your best to answer carefully.
Definitely
complete
started
started
started
Seems
I don’t
Barely
know
Not
1. Would you say that your growth in height
has:
2. Would you say that your body hair growth

has:
"Body hair" means hair any place other than your
head, such as under your arms.
3. Have you noticed any skin changes,

especially pimples?
4. Have you noticed that your breasts have

begun to grow?
Please circle the answer that suits you the most for the following question:
5a. Have you begun to menstruate Yes No I don’t Know
(started to have your period)?
If you circled yes to 5a please write an answer to 5b.

5b. If yes, how old were you when
you started to menstruate _________________________
(when you had your first
period)?
The End.

PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

PDF

Uploaded by

Copyright:

Available Formats

Dietary Caffeine and Sugar and their

Relationship with Sleep and Behaviour in

Emily Jane Watson

Submitted in fulfilment of the requirements for the degree of

Doctor of Philosophy in Psychology

1.2 SLEEP .............................................................................................................................................................. 2

1.3 SLEEP NEED ...................................................................................................................................................... 4

1.4 WHY MIGHT SLEEP BECOME DISRUPTED? ............................................................................................................... 6

1.5 SUBJECTIVE SLEEP MEASURES ............................................................................................................................... 7

1.6 CONSEQUENCES OF NOT ENOUGH SLEEP, OR POOR SLEEP........................................................................................... 9

CHAPTER 2 CAFFEINE & SLEEP.............................................................................................................. 15

2.2 CAFFEINE ....................................................................................................................................................... 16

2.2.1 How much caffeine do Australians consume? .................................................................................... 16

2.3.1 Caffeine food frequency questionnaires ............................................................................................. 25

2.4.1 How caffeine impacts sleep ................................................................................................................ 28

2.6 CAFFEINE, SLEEP AND BEHAVIOUR IN CHILDREN...................................................................................................... 52

CHAPTER 3 SUGAR AND SLEEP IN CHILDREN ........................................................................................ 54

3.2 DIETARY SUGAR .............................................................................................................................................. 55

3.4 HOW MUCH SUGAR ARE CHILDREN CONSUMING? .................................................................................................. 59

3.5 SUGAR AND THE BODY ...................................................................................................................................... 63

3.6 SUGAR AND SLEEP............................................................................................................................................ 63

3.7 SUGAR AND BEHAVIOUR.................................................................................................................................... 71

CHAPTER 4 STUDY AIMS ...................................................................................................................... 73

4.2 STUDY TWO .................................................................................................................................................... 75

CHAPTER 5 GENERAL METHODOLOGY ................................................................................................. 77

5.2 ETHICAL APPROVAL.......................................................................................................................................... 78

5.3 STUDY ONE .................................................................................................................................................... 78

5.3.1 Informed Consent ................................................................................................................................ 78

5.4.1 Informed Consent ................................................................................................................................ 84

6.2 ABSTRACT: ..................................................................................................................................................... 94

6.3 PILOT STUDY ................................................................................................................................................... 95

6.4 INTRODUCTION ............................................................................................................................................... 96

6.5 MATERIALS AND METHODS ............................................................................................................................... 97

6.5.1 Participants ......................................................................................................................................... 98

6.7 DISCUSSION.................................................................................................................................................. 107

6.8 CONCLUSION: ............................................................................................................................................... 112

7.2 ABSTRACT: ................................................................................................................................................... 114

7.3 INTRODUCTION ............................................................................................................................................. 115

7.4 MATERIALS AND METHODS ............................................................................................................................. 117

7.6 DISCUSSION.................................................................................................................................................. 125

7.7 CONCLUSIONS............................................................................................................................................... 127

8.2 ABSTRACT .................................................................................................................................................... 130

8.3 BRIEF SUMMARY: .......................................................................................................................................... 131

8.4 INTRODUCTION ............................................................................................................................................. 132

8.5 METHOD ..................................................................................................................................................... 134

8.5.1 Participants ....................................................................................................................................... 134

8.7 DISCUSSION.................................................................................................................................................. 147

9.2 ABSTRACT .................................................................................................................................................... 154

9.3 INTRODUCTION ............................................................................................................................................. 156

9.4 METHOD ..................................................................................................................................................... 158

9.4.1 Participants ....................................................................................................................................... 158

9.6 DISCUSSION.................................................................................................................................................. 168

CHAPTER 10 GENERAL DISCUSSION ................................................................................................... 172

10.2 CAFFEINE FOOD FREQUENCY QUESTIONNAIRE .................................................................................................... 175

10.3 IMPACT OF CAFFEINE .................................................................................................................................... 176

10.4 RELATIONSHIP BETWEEN CAFFEINE AND SUGAR.................................................................................................. 178

10.5 SUGAR INTAKE IN CHILDREN ........................................................................................................................... 179

10.6 IMPACT OF SUGAR ....................................................................................................................................... 179

10.7 FUTURE RESEARCH ....................................................................................................................................... 180

10.8 CONCLUDING REMARKS ................................................................................................................................ 183

ACAES-FFQ Australian Child and Adolescent Eating Survey – Food Frequency

FDA The Food and Drug Administration

PCS puberty category scale