You are on page 1of 27

SUMMARY

GLOBAL BURDEN OF DISEASE AND INJUR Y SERIES

THE GLOBAL
BURDEN OF DISEASE
A comprehensive assessment of mortality
and disability from diseases, injuries, and
risk factors in 1990 and projected to 2020

EDITED BY

CHRISTOPHER]. L. MURRAY
HARVARD UNIVERSITY
BOSTON, MA, USA

tt.'·
~ lijIo,~W
~ :I ~
~ 'I~~
1ff
WORLD HEALTH WORLD
ORGANIZATION BANK

HARVARD SCHOOL OF
PUBLIC HEALTH

PUBLISHED BY THE HARVARD SCHOOL OF PUBLIC HEALTH ON BEHALF OF


THE WORLD HEALTH ORGANIZATION AND THE WORLD BANK
DISTRIBUTED BY HARVARD UNIVERSITY PRESS
CONTENT S

INTR O D U CT I O N I

1. THE GBD ' s A P P RO AC H TO MEASU RIN G HEA LTH STATUS 6


2. How T HE WOR LD D I ES T O DAY 14

ISBN : 0-9655466-0-8 3. DI SABILIT Y: T HE I N VIS I BLE BUR D E N 20


Library of Congress Cataloging-in -Publication (CIP) Data applied for. 4. TH E G LO BAL B URD E N O F DI S EA S E I N 1990

S. RI S K F ACTO R S F OR DEATH AN D DISAB I LI T Y 27

6. LOO K I NG AH EAD: TH E HE A LTH OF THE WOR LD IN 2020 3I


7. FROM THE S E RI ES 40
T able of Co ntents for Volum e I 40
List o f Ann ex T ables 41
Global H ealth Statistics: Samp le tabl e fro m Volume 11 4 2-
The G lo ba l Bu rden of D isease and Inju ry Series 43

This booklet is a summary of the The Global Burden of Disease: A


Comprehensive Assessment of Mortality and Disability from Diseases,
Injuries, and Risk Factors in I990 and Projected to 2020, edited by
Christopher ].L. Murray and Alan D. Lopez, published by the Harvard
School of Public Health on behalf of the World Health Organization and the
World Bank and distributed by Harvard University Press. It also contains
examples of the detailed information on the epidemiology of 240 conditions
around the world published in Global Health Statistics: A Compendium of
Th e Glob al Burden of Disease Study and the The Glob al Burd en of Disease a nd Incidence, Prevalence and Mortality Estimates for Over 200 Conditions, by
Global Health Statistics would no t have been possible with out the financial support Christopher J.L. Murray and Alan D. Lopez, published by the Harvard
of the Edna McConnell C lark Foundation , the Rocke feller Foundation , the World School of Public Health on behalf of the World Health Organization and the
Bank a nd the WHO Ad H oc Comm ittee o n H ealth Research Relating to Future World Bank and distributed by Harvard University Press.
Interv ent ion Option s.
This summary has been prepared by Phyllida Brown . The preparation and
printing of the summary were made possible through a grant from the Eli Lilly
Foundation and Harvard Medical International.
© Copyright 1996 World Health Organization. All rights reserved.
"Publication of the Global Burden of Disease and Injury Series marks the
transition to a new era. . .1 firmly predict that by the turn of the cent ury the
official reporting of health outcomes in dozens of countries and globally will
embody the approach and standards described in this series."
INTRODUCTION
Dean T. [amison, Professor of Public Health, Uniuer-
sity of California Los Angeles and Chairman, WHO
Ad Hoc Committee of Health Research Relating to
Future Interventi on O ptio ns

The next two decades will see dramatic changes in the health needs of the
world 's populations. In the developing regions where four-fifths of the
planet's people live, noncommunicable diseases such as depression and
heart disease are fast replacing the traditional enemies, such as infectious
diseases and malnutrition, as the leading causes of disability and premature
death. By the year 2020, noncommunicable diseases are expected to
account for seven out of every ten deaths in the developing regions,
compared with less than half today. Injuries, both unintentional and
intentional, are also growing in importance, and by 2020 could rival
infectious diseases worldwide as a source of ill health.
These changes are expected because of the rapid aging of the developing
world's populations, As a population's birth rate falls, the number of
adults relative to children increases, and the population's commonest
health problems become those of adults rather than those of children. In
China, some other parts of Asia and Latin America, this so -called
"epidemiological transition" is already much further advanced than many
public health specialists appreciate. In all regions the rapidity of change,
and the very large absolute numbers involved, will pose serious challenges
to health-care systems and force difficult decisions about the allocation
of scarce resources. Yet, until now, many governments have lacked even
ktGIOI\IAL GItOUr l NCS
the most basic data they needed to inform debate and to assess priorities
o ~::i~~~~~.t=:i::~~~lleb~ .;ru lC)
Ewl'lom,n -dt tNl't'hi .... fIllf.ahikf,If_\11ff.
0 ~ ::t:~~j:~~~~~rn fOrA!)
Auuff. r'-')' ...ANCC1'IH IArAI) for public health .
~=~~~~~~~':i:l:!tErf~ IFSEf 0 ~~~~rs..~~i~~~~~SS)
CIlfAIn iOl:ult\ln dOhrot'" :AI'5 r.1 A(t'iClW ",h-uhu;~ IASSI Now, for the first time, this gap has been filled with a landmark
o Aftcinv-
I..dl;l tlNDI
Jl'loNtINDI
Wrin Amtriu ,,"It ,ht Unhhullll.AQ
Amtinc.1l..uin.a)'dQnkfAI.O
publication. Researchers at the Harvard School of Public Health and the
InckIINt>, AtnrfiqllC'l.,incccuuIM(,UCI

C g:~ :~~~:
Ch"",ICUNI
• ~~~cJt;;:::~C~ ~'b~~:
Croinanr nloytfl_I('fICotllC.\tO)
World Health Organization, with more than 100 collaborators from
around the world, have produced a comprehensive, internally consistent
and comparable set of estimates of current patterns of mortality and
disability from d isease and injury for all regions of the world, with
projections to the year 2020. The Global Burden of Disease and Injury
Series, the ten-volume result of their study, is a unique resource that
2 The Global Burden of Disease SUMMARY 3

provides policy -makers with their well advanced, suggesting that public health policy, with its traditional
first comprehensive picture of the Deaths from emphasis on infectious disease, has not kept pace with events. In addition,
world's current and future health it makes a number of startling individual observations. Just four examples
needs .
noncommunicable are highlighted here:
The researchers involved in this diseases are expected to • The burdens of mental illnesses, such as depression, alcohol dependence
a m biti o us five -year effo rt and schizophrenia, have been seriously underestimated by traditional
developed a new approach to
climb from 28. I million
approaches that take account only of deaths and not disability. While
measuring health status. Their a year in I990 to 49.7 psychiatric conditions are responsible for little more than one per cent
method quantifies not merely the of deaths, they account for almost 11 per cent of disease burden
number of de aths but also the
million by 2020-an
worldwide.
impact of premature death and increase in absolute
disability on a population , a nd • Adults under the age of 70 in Sub-Saharan Africa today face a higher
combines these into a single unit
numbers of 77 per cent. probability of death from a noncommunicable disease than adults of
of measur ement of the overall the same age in the Established Market Economies.
" b u rd en of disease" on the • Men living in the Formerly Socialist Economies of Europe have a
population. The series also presents the first estimates of the proportion disturbingly poor, and deteriorating, health status, including a 28 per
of mortality and disability that can be attributed to certain risk factors cent risk of death between the ages of 15 and 60 .
for disea se, includi ng tobacco, alcohol, poor water and san itat ion , and
• By 2020, tobacco is expected to kill more people than any single disease,
unsafe sex .
surpassing even the HIV epidemic.
A TIMELY ASSESSMENT OF GLOBAL HEALTH NEEDS The Global Burden of Disease
Far from confirming what was already known, the stu dy offers significant Adults under the age of study (GBD) has involved an
surprises. Overall, it shows that th e epidemiologica l transition is already estimated forty person-years of
70 in Sub-Saharan Africa effort. An extraordinarily large
today face a higher volume of data-on 483 separate
Figure I Projected trends in death by broad cause Group, sequelae of 107 diseases and
developing regions probability of death from in ju ries, and 14 million death
Developing regions a noncommunicable cenificates--has been subjected
40 to rigorous analysis using both
disease than adults of the newly developed and well-
same age in the established methods. Volume I,
The Global Burden of Dis ease,
-,g VI
C
30
Established Market summarizes the key concepts ,
methods and results and Volume
Economies.
11, Global Health Statistics ,
I 20
presents the mass of underlying
...
s:
VI

epidemiological and demographic data. Details of the remaining eight


Cl
'"41 volumes, which deal with specific conditions and country-based analyses,
10
are shown on page 43 of this booklet.
The burden of disease and injury ha s been calculated for eight
O L.L ---L -L- ----J demographic regions (see Map ): the Established Market Economies
1990 2000 2010 2020 (largely the OECD countries) (EME); the Formerly Socialist Economies
of Europe (FSE); India ((ND); China (CHN I; Other Asia and Islands (OAI);
- Noncommunicable diseases Sub-Saharan Africa (SSA); Latin America and the Caribbean (LAC) and
Communicable. perinatal. maternal and nutritional conditions the Middle Eastern Crescent (MEC). Burden is analysed by five age groups,
- Injuries by sex and by cause.
4 The Global Burden of Disease SUMMARY 5

This booklet presents the key findings of Volume I. The first section Figure 3 DALYs attributable to diar rhoea . HIV and tobacco.
summarizes the concepts and methods involved in developing a single 1990-2020 (baseline sce nario)
measure of health status. Sections 2 through 5 present the results for the
1990 assessments and section 6 summarizes the 2020 projections. The final Voll p 317

section provides details and samples of the content of Volumes I and 11. 140

Figure 2 Change in the rank order of disease burden for 15 leading


120
causes, world. 1990-2020
Disease burden measured in Disability-Adjusted Ufe Years (DALYs)
Vol t p 375 -III
c:
~
100

1990 2020 80
(Baseline scenario)
].
III
Disease or injury Disease or injury ~
60
Lower respiratory Ischaemic heart disease (3
infections
Diarrhoeal diseases 2 Unipolar major 40
depression
Conditions arising during 3 Road traffic accidents
the perinatal period 20
Unipolar major 4 Cerebrovascular disease
depression
Ischaemic heart disease 5 Chronic obstructive o 1990 1995 2000 2005 2010 2015 2020
pulmonary disease
Cerebrovascular disease 6 Lower respiratory
infections - -- - Diarrhoea - - - HIV - - - - Tobacco
Tuberculosis Tuberculosis
Measles War
Road traffic accidents Diarrhoeal diseases Figure 4 By 2020. tobacco is expected to cause more premature
Congenital anomalies HIV death and disability than any single disease
Malaria Conditions arising
during the perinatal Vol I P 316 (See Map for explanation of regional composition)
period 25% .-- - - - - - - - - - - - - - - - - --,
Chronic obstructive Violence ~
pulmonary disease
~

Falls
Iron-deficiency anaemia
Congenital anomalies
Self-inflicted injuries
~c
il~
.. ~
20%
eu 0
Protein-energy Trachea, bronchus and on",

malnutrition lung cancers :ii ar 15%


.~ ut
-.::l '"
0

..'" ..
.!:!
v
"0 v

16 19
oS 10%
~~
c:-
eu.J:)
17 24
.. 13
v
rf,e'"
5%
19 25 e..
28 37
EME FSE CHN LAC OAI MEC IND SSAWORLD
33 39 Region
6 The Global Burden of Dis ease SUMMARY

1. THE GBD's APPROACH TO MEASURING HEALTH different countries, many of them variants of the so -called Quality-
STATUS Adjusted Life Year or QALY. For the GBD, an internationally
standardized form of the QAL Y has been developed, called the Disability-
Adjusted Life Year (DALY). The DAL Y expresses years of life lost to
I. I WHY THE GBD IS DIFFERENT premature death and years lived with a disability of specified severity and
In general, statistics on the health status of populations suffer from several duration . One DALY is thus one lost year of healthy life. Here, a
limitations that reduce their practical value to policy -makers: "premature" death is defined as one that occurs before the age to which
the dying person could have expected to survive if they were a member of
• First, they are partial and fragmented. In many countries even the most
a standardized model population with a life expectancy at birth equal to
basic data-the number of deaths from particular causes each year-
that of the world's longest-surviving population, Japan.
are not available. Even where mortality data a re available, they fail to
To calculate total DALYs for a
capture the impact of non-fatal outcomes of d isease and injury, such
given condition in a population,
as dementia or blindness, on population health. One D AL Y is one lost years of life lost (YLLs) and years
• Second, estimates of the numbers killed or affected by particular year of healthy life. lived with disability of known
conditions or diseases may be exaggerated beyond their severity and duration (YLDs) for
demographically plausible limits by well-intentioned epidemiologists that condition must each be
who also find themselves acting as advocates for the affected estimated, and then the total summed. For example, to calculate DAL Ys
populations in competition for scarce resources. If the currently incurred through road traffic accidents in India in 1990, add the total years
available epidemiological estimates for all conditions were right, some of life lost in fatal road accidents and the total years of life lived with
people in a given age group or region would have to die twice over to disabilities by survivors of such accidents.
account for all the deaths that are claimed.
1.3 A SOCIETY'S VALUES ARE EXPLICITLY BUILT INTO ITS MEASURES OF
• Third, traditional health statistics do not allow policy-makers to HEALTH STATUS
compare the relative cost-effectiveness of different interventions, such
as, for example, the treatment of ischaemic heart disease versus long- It might appear that quantifying disease burden is a neutral exercise,
term care for schizophrenia. At a time when people's expectations of entirely free of value choices. However, this is far from the case. Disease
health services are growing and funds are tightly constrained, such burden is, in effect, the gap between a population's actual health status
information is essential to aid the rational allocation of resources. and some "ideal" , or reference status. In order to measure burden, a society
has to decide what the ideal or reference status should be. This involves
The GBD set out to address these problems with three explicit aims: making five value choices:
1. to incorporate non-fatal conditions into assessments of health status; • How long " should " people live? If health researchers are to estimate
how many years of life are lost through death at any given age, they
2. to disentangle epidemiology from advocacy in order to produce
must decide on the number of years for which a person at that age
objective, independent and demographically plausible assessments of
should expect to survive in the ideal, or reference, population. That
the burdens of particular conditions and diseases; and
could be, for example, 60, 80 or 90 years from birth.
3. to measure disease and injury burden in a currency that can also be used
• Are years of healthy life worth more in young adulthood than in early
to assess the cost-effectiveness of interventions, in terms of the cost per
or late life?
unit of disease burden averted.
• Is a year of healthy life now worth more to society than a year of healthy
life in 30 years' time?
r.2 A SINGLE MEASURE OF DISEASE BURDEN

In order to capture the impact of both premature death and disability in • Are all people equal? For example, should one socioeconomic group's
a single measure, a common currency is required. Since the late 1940s, years of healthy life count for more than another's?
researchers have generally agreed that time is an appropriate currency: time • How do you compare years of life lost due to premature death and years
(in years) lost through premature death, and time (in years) lived with a of life lived with disabilities of differing severities?
disability. A range of such time-based measures has been developed in
8 The Global Burden of D isease
SUMMARY 9

Health researchers developing a measure of disease burden must Are Years of Healthy Life Worth More in Young Adulthood than in
recognize their responsibility to reflect societies' preferred answers to these Early or Late Life?
five questions, but also to guard against and "filter out" unjustifiable
If individuals are forced to choose between saving a year of life for a 2 year-
pr~fer~nc~s suc~ as .racism, sexism or economic discrimination that may
old and saving it for a 22 year-old, most prefer to save the 22 year-old. A
be institutionalized 10 certain societies. It is unlikely that any measure can
range of studies confirms this broad social preference to "weight" the value
reflect a perfect vision of the ideal society; but its choices should be
of a year lived by a young adult more heavily than one lived by a very
acceptable to as many people of as many different cultures as possible.
young child or an older adult. Adults are widely perceived to play a critical
The Egalitarian Principles on which the DALY Is Based role in the family, community and society. The GBO researchers therefore
incorporated age-weighting into the OAL Y. They assume that the relative
The GBO has sought to develop a measure based on explicit and
value of a year of life rises rapidly from zero at birth to a peak in the early
transparent value choices that may be readily debated and modified.
twenties, after which it steadily declines.
Overall, the OALY has a strongly egalitarian flavour. It is built on the
principle that only two characteristics of individuals that are not directly
related to their health-their age and their sex-should be taken into
Figure 5 The relative value of a year of life lived at different ages, as
consideration when calculating the burden of a given health outcome in
incorporated into DALYs
that individual. Other characteristics, such as socioeconomic status, race
or level of education, are not considered, so, for example, years of healthy Voll p60
life lived by the director of a bank are regarded as no more valuable than
those lived by a poor rural peasant. In the remainder of this section the 2.0
social choices that affect the '
OALY are each discussed briefly.
DAL Ys are intended to .c:
.... 1.5
How Long Should People Live? 0()

In accordance with the GBO's


be a transparent tool to '(jj
~
1.0
egalitarian principles, the study enhance dialogue on the ~
'0
assumes a standard life table for M

all populations, with life


major health challenges "'(jj
a:: 0.5
expectancies at birth fixed at 82.5 facing humanity.
years for women and 80 years for
men. A standard life expectancy 40 60 80 100
20
allows deaths in all communities at the same age to contribute equally to
the burden of disease. Alternatives, such as using different life expectancies Age (years)
for different populations that more closely match their actual life
expectancies, interfere with the egalitarian principle. For example, if a 35
year-old woman dies in childbirth in an African country where she might Is a Year of Healthy Life Now Worth More to Society than a Year of
have expected to live another 30 years, her years of life lost would be Healthy Life in 30 Years' Time?
deemed unfairly to be fewer than those for a 35 year-old woman who dies
If a person is offered $100 today or $100 in a year's time, that person is
in childbirth in Japan, when she might otherwise have expected to live
likely to prefer $100 today. Future dollars are thus discounted-valued
another 48 years.
lower-against current dollars. Whether a year of healthy life, like a dollar,
Life expectancy is not equal for men and women. Accordingly, the GBO
is also deemed to be preferable now rather than later, is a matter of intense
has given men a lower reference life expectancy than women. However,
debate among economists, medical ethicists and public health planners,
since much of the difference between men and women is determined by
because discounting future health affects both measurements of disease
men's higher exposure to various risks such as alcohol, tobacco and
burden and estimates of the cost-effectiveness of an intervention.
occupational injury, rather than purely biological differences this choice
There are arguments for and against discounting and The Global
is arguably a form of discrimination against men and could be modified
Burden of Disease discusses them in depth. The GBO researchers decided,
in future revisions of the OALY.
however, to discount future life years by 3 per cent per year. This means
10 The Global Burden of Disease SUMMARY II

that a year of healthy life bought Table I Gaugingthe severity of disability: disability classes and
for 10 years hence is worth weights set by the GBD protocol for 22 indicator conditions
If individuals are forced
around 24 per cent less than one
Voll P 40
bought for now, as discounting is to choose between saving
represented as an exponential a year or life for a 2 Disability Severity
decay function. Because the class weights Indicator conditions
impact of discounting is year-old and saving it for I 0.00-0.02 V'til'go on face. weight-for-hejght less than 2 standard deviations
significant, the GBD publishes a 22 year-old, most 2 0.02-0.12 Watery diarrhoea. severe sore throat. severe anaemia
a lterna ti ve resu Its based on 3 0.12-0.24 Radius fracture in a stiff cast, infertility. erectile dysfunction.
DAL Ys without discounting. prefer to save the 22 rheumatoid arthritis. angina
Discounting future health year-old. 4 0.24-0.36 Below-the-knee amputation. deafness
reduces the relative impact of a 5 0.36-0.50 Rectovaginal fistula. mild mental retardation. Down syndrome
child death compared with an 6 0.50-0.70 Unipolar major depression, blindness. paraplegia
adult death. For example, with age-weighting also incorporated, a year- 7 0.70-1.00 Active psychosis. dementia. severe migraine. quadriplegia
old girl's death causes a loss of 34 years of life while a 25 year-old woman's
Note: These weights were escabl shed using the person trade-off method with an international group of health
death results in a loss of 33 years of life. Discounting also reduces the value workers who met at WHO In Geneva in August 1995. Each condition is actually a detailed case. For example.
of interventions that payoff largely in the future-such as vaccinating angina in this exercise is defined as reproducible chest pain. when walking 50 meters or more. that the
against hepatitis B, which may prevent thousands of cases of liver cancer, individualwould rate as a 5 on a subJectve pa'n scale from 0 to 10.

but some decades later.


How Do You Compare Time Lost Due to Premature Death with Time the GBD is that the more explicitly these preferences are set out, the more
Lived with Disability? meaningfully they may be debated.
While death is not difficult to define, disability is. All non-fatal health The GBD therefore developed a protocol based on the person trade-off
outcomes of disease are different from each other in their causes, nature, method. In a formal exercise involving health workers from all regions of
and their impact on the individual, and the impact on the individual is in the world, the severity of a set of 22 indicator disabling conditions-such
turn mediated by the way the surrounding community responds. Yet, in as blindness, depression, and conditions that cause pain-was weighted
order to quantify time lived with a non-fatal health outcome and assess between 0 (perfect health) and 1 (equivalent to death). These weights were
disabilities in a way that will help to inform health policy , disability must then grouped into seven classes where class I has a weight between 0.00
be defined, measured and valued in a clear framework that inevitably and 0.02 and Class VII a weight between 0.7 and 1(see Table 1).
involves simplifying reality. Despite their diverse cultural backgrounds, the participants reached
There is surprisingly wide agreement between cultures on what consensus on these weights. Their choices also closely match the pooled
constitutes a severe or a mild disability. For example, a year lived with results of nine additional exercises with other participants using the same
blindness appears to most people to be a more severe disability than a year protocol (Figure 6).
lived with watery diarrhoea, while quadriplegia is regarded as more severe In essence, the weight is set by the number of people with a given
than blindness.These judgements must be made formal and explicit if they condition whose claim on a fixed healthcare budget is equal, in the
are to be incorporated into measurements of disease burden. judgement of a partipant, to that of 1000 healthy people. For example, if
Two methods are commonly used to formalize social preferences for the participant judges that 1000 entirely healthy people would have an
different states of health. Both involve asking people to make judgements equal claim on the resources as 8000 people with some severe disability,
about the trade-off between quantity and quality of life. This can be the weight assigned to that particular disability is equal to 1 minus 1000
expressed as a trade-off in time (how many years lived with a given divided by 8000, or 0.875. If 1000 entirely healthy people were judged
disability would you trade for a fixed period of perfect health?) or a trade- to have an equal claim on the resources as 2000 people with a particular,
off between persons (would you prefer to save one life-year for 1000 less severe, disability, the weight assigned would be equal to 1 minus 1000
perfectly healthy individuals as opposed to saving one life-year for 2000 divided by 2000, or 0.5.
individuals in a worse health srater). While such trade-offs may affront For the GBD protocol, each participant is asked two versions of the
our perceptions about what is morally acceptable, they are practised person trade-off question, one about extending life for people in a given
implicitly throughout the world's health care systems. The philosophy of health state versus extending life for healthy people, the second about
u. The Global Burden of Disease
SUMMARY 13

Figure 6 In close agreement: weights set for 22 indicator disabling


Overall, the rankings of diseases and the distribution of burden by
conditions based on the combined results of nine exercises
broad cause Group are largely unaffected by age-weighting and only
using the GBD protocol. compared with the results of the
slightly affected by changing the method for weighting disability. Changes
Geneva meeting
to the discount rate, by contrast, may have a more significant effect on the
Voll piO overall results. A higher discount rate results in an increased burden in
older age groups, while a lower discount rate results in an increased burden
1.0 in younger age groups. Changes in the age distribution of burden, in turn,
affect the distribution by cause, because communicable and perinatal
~
s: conditions are commonest in children while noncommunicable diseases
.~ 0.8
~
are commonest in older adults. The most significant effect of changing the
discount and age weights is to reduce the importance of several psychiatric
~ 0.6 -
conditions (Figure 7).
:0
~ Ultimately, however, the accuracy of the underlying basic
:s 0,4
epidemiological data from which disease burden is calculated will influence
'tl
QI
e the final results much more than the discount rate, the age weight, or the
:0
E 0.2 disability weighting method. If, for example, estimates of the incidence of
o blindness are off by a factor of two, then it follows that this will be reflected
U
Of:Do"-- ....- - ----'- - - .L-- - -'-- - .....J in the results. The GBD researchers conclude that researchers' efforts
o 0.2 0,4 0.6 0.8 1.0 should be invested in improving the basic data rather than in spending
Geneva meeting disability weights excessive energy on analysing the effects of small adjustments to the
measure itself.

Figure 7 Relationship between the rank order of causes of global


giving health back to people in a given health state versus extending life burden using DALYs calculated without age-weighting and
for healthy people. Two questions are asked because people's answers to discounting and standard DALYs calculated with age-
each one are invariably inconsistent with the other, and the process of weighting and a three per cent discount rate, 1990
making them consistent forces the participant ro think through the (highest rank is the largest cause)
implications of their decision in greater depth. Vol I P 282
The implications of choosing between the claims of different groups in
100
a society are profound, so the process of setting weights cannot be Unipolar major depression
undertaken lightly. While some other studies have used relatively rapid 'i:'
QI
methods, such as telephone surveys, the GBD protocol is a deliberative DO"E 80
process in which a comparatively small group of participants (between 8 .5 0
".,:,t.
and 12 ) are confronted with the implications of their decision, encouraged -foe
to discuss their choices with their peers, and allowed to revise their initial l'~"60
, ID
ID ...
choices. This is time-consuming: at most 20-25 conditions can be discussed ~f
in a full working day. Once the 22 indicator conditions have been 0"
e § 40
weighted, the participants assigned the remaining conditions across the '" 0
seven classes. ~:;(
«=5
1.4 How MUCH DO DIFFERENT VALUE CHOICES AFFECT THE RESULTS?
Cl ...0
QI
20
N

To ga uge the impact of changing these social choices on the final measures
of disease burden, the GBD researchers re-calculated their assessments with 20 40 60 80 100
alternative age-weighting and discount rates, and with alternative methods
Standard DALYs (rank order)
for weighting the severity of disabilities.
14 The Global Burden of Dlsease SUMMARY

2. How THE WORLD DIES TODAY Figure 8 Deaths by broad cause Group. 1990
Developing regions Developed regions
Most developing countries still have only limited information about how
their populations die. One of the chief objectives of the GBD has been to (10.7%)
develop comprehensive internally consistent estimates of how many people
died of each major cause in 1990 worldwide. No such data set was
available before the GBD began work. The methods are briefly described,
followed by a selection of key results.

2.I How THE GBD ESTIMATED THE DEATHS DUE TO EACH CAUSE

Deaths were classified using a tree structure. The first level of


disaggregation comprises three broad cause Groups:
• Group I: Communicable, maternal, perinatal and nutritional Group I o Group 11 DGroup U
conditions;
• Group 11: Noncommunicable diseases; and • Cause-or-death models. Models are used to check the validity of existing
• Group Ill: Injuries. data by putting demographic limits on epidemiological estimates. Such
models estimate the distribution of deaths by cause in a population from
Each group was then subdivided into sub-categories: for example, historical studies of the relationship between deaths from twelve broad
cardiovascular diseases and malignant neoplasms (cancers) are two sub- groups of causes, such as cardiovascular diseases and infectious diseases,
categories of Group Il . Beyond this level, there are two further and the total number of deaths. The models developed for this study
disaggregation levels so that 107 individual causes for which codes have drew on a dataser of 103 observations from 67 countries between 1950
been included in the Ninth Revision of the International Classification of and 1991.
Diseases (ICD-9), such as tuberculosis, stomach cancer, or road traffic
accidents, can be listed separately. 2.2 RESULTS: DEATHS IN 1990
A demographic data set giving information on population size and the Worldwide, one death in every three is from a Group I cause
distribution of deaths for each region was developed especially for the (communicable, maternal, and perinatal conditions and nutritional
GBD. Next, to reach estimates on the number of deaths by cause, the deficiencies). Virtually all of these deaths are in the developing regions.
researchers drew on four broad sources of data: One death in ten is from Group III causes (injuries) and just over half of
• Vital registration systems. These are complete only for the Established all deaths are from Group 11 causes (noncommunicable diseases). Figure
Market Economies and the Formerly Socialist Economies of Europe, 8 shows the proportionate burden from each cause for developing and
but some vital registration information is available for all regions except developed regions.
China, India and Sub -Saharan Africa (excepting South Africa). More surprising, perhaps, is the finding that, for several major
developing regions, more people already die of Group 11 causes than Group
• Sample death registration systems. In China, a set of 145 Disease I causes. In Latin America and the Caribbean, there are almost twice as
Surveillance Points, representative of both rural and urban areas, and many deaths from noncommunicable diseases as from Group I causes. In
covering about 10 million people, provides information on deaths by China, there are four -and-a-half times as many deaths from
cause. In India, Maharashtra State provides full medical certification noncommunicable diseases as from Group I causes. The balance has also
for at least 80 per cent of urban deaths, while a rural surveillance system tipped towards Group 11 causes in the Middle Eastern Crescent and the
including more than 1300 primary health care centres nationwide was region comprising As ia beyond India and China, and the Pacific
used to assess rural death patterns. islands(OAI). Only in India and Sub-Saharan Africa do Group I causes still
• Epidemiological assessments. Epidemiologists have made estimates of dominate, accounting for 51 per cent and 65 per cent of deaths
deaths for specific causes, such as malaria, in certain regions. These respectively.
estimates combine information from surveys on the incidence or The results also dispel any remaining notions that noncommunicable
prevalence of the disease with data on case-fatality rates for both treated diseases are related to affluence. When the estimates are expressed in terms
and untreated cases. of the probability of dying at a given age in a given region, a striking picture
16 T he Global Burden of Disease SUMMARY 1

Correcting for rniscodlng and "garbage" codes emerges. For adults under the age of 70, the probability of dying from a
noncommunicable disease is greater in both Sub-Saharan Africa and India
Under the rules of the International Classificationof Diseases,deaths must be than in the Established Market Economies. The results show that
assigned to one cause only, and this must be the underlying cause: for premature mortality rates from noncomrnunicable diseases are higher in
example, if someone has a heart attack and later dies of pneumonia in popularions with high mortality and low income than in the industrialized
hospital, the cause of death must be recorded as ischaemic heart disease. countries (Figure 9).
Miscoding of deaths-that is, assigning the death to the "wrong" cause in
The health of men in the Formerly Socialist Economies of Europe is
ICD ·9, occurs regularly in all countries with registration systems. For
example, the choice of codes for cardiovascular diseases is notoriously surprisingly poor. Men face a 28 per cent risk of death between the ages
variable between industrialized countries, with a significant proportion of of 15 and 60, the highest risk in any region except Sub-Saharan Africa.
ischaemic heart disease deaths being attributed to ill-defined codes such as This excess is explained by a higher rate of noncommunicable diseases and
heart failure. When national statistics are corrected for probable miscoding, also by a higher risk of death from injury than for men in the Established
substantially different pictures emerge. Miscoding may help to account, for Market Economies.
example, for some of the unexplained variation in rates of death from
ischaemic heart disease between the developed nations. Before correction, Leading Causes of Death in 1990
the national data suggest there are more than six times as many deaths from Just over 50 million people died worldwide in 1990, with ischaemic heart
ischaemic heart disease in adults in Finland (the country with the highest disease (lHD) causing more deaths than any other disease or injury. Only
rate) as in Japan (the country with the lowest rate). After correction, the 2.7 million of the 6.3 million who died of IHD were in the developed
difference is reduced to 2.3 to 1. To correct for such miscoding, the GBD
developed specific algorithms to redistribute the miscoded deaths.
A varying proportion of deaths in all countries are ascribed to so-called Figure 9 Noncommunicable diseases are not diseases of affluence:
"garbage" codes, such as "Symptoms, signs, and ill-defined conditions," risks of death from each broad cause Group. by region,
which account for less than 3 per cent of certificates in the Established 1990
Market Economies but as many as one-fifth in some developing countries.
These deaths are redistributed across other causes. Vol I p 171
Males Females
40 Probab ility of death between ages 0 and IS 40 ProbabIlity of deat h between ages 0 and IS
Box Figure I Mortality rates from ischaemic heart disease for selected countries.
corrected and uncorrected gJO gJO
~ ~

I
;;: 20 ;;: 20
Vol I p 132 .<l .<l
.s2 .s
a.
10
et 100 [
__-""-
=.;;;-= .=-..--...:I.
"O[ IHO death rates before correction

.. ,-
IHO death rates after correction

~O
EHE fse CHN LAC OAf HEC INO SSA

Probab ility of death between ages IS and 60 ~O


EHE FSE CHN LAC OAI HEC
=...........

Probabilityof death between ages IS and 60

~ 300 [ I
....
.
gJO
~
;;: 20
~ 200 I"' ., J
:I -
c. I, I 10

...
1!
llllOO ~
1
EHE FSE CHN LAC OAI HEC INO SSA
Q
! I I .1 I,
~ ProbabIlity of death between ages 60 and 70 40 Probability of death between ages 60 and 70

f
~
20

~ 10
o ..:... ........_ ...... ..
..;;;;....:.....;~~

£HE FSE CHN LAC OAf HEC INO SSA EHE FSE CHN LAC OAI MEC INO SSA

Probability of death from : _ Group I _ Group 11 _ Gr oup III


I8 The Global Burden of Disease SUMMARY

world. Cerebrovascular disease (stroke) killed 4.4 million people, of whom Injuries and the Burden of Suicide
only 1.4 million were in the developed world. Lower respiratory infections Worldwide in 1990, about 5 million people died of injuries of all types,
(pneumonia) killed 4.3 million people, all but 0.4 million of them in the two-thirds of them men. Most of these deaths are heavily concentrated
developing world. Diarrhoeal diseases caused 2.9 miIlion deaths, virtually among young adults. In this age group, road traffic accidents, suicide, war,
all in the developing world. The ten leading causes together accounted for fire and violence all figured within the ten leading causes of death. The
just over half of all deaths (Table 2). significance of injuries has been largely overlooked by the health sector
in many countries.
The Traditional Enemies in Developing Countries Among adults aged 15-44 worldwide, road traffic accidents were the
Despite the epidemiological transition, deaths from communicable leading cause of death for men and the fifth most important for women.
diseases, maternal and perinatal conditions and nutritional deficiencies For women aged between 15 and 44, suicide was second only to
continue to take a heavy, and largely avoidable, toll even though there have tuberculosis as a cause of death. In China alone, more than 180 000
been spectacular successes in their control over the past 30 years. Fully 17.3 women killed themselves in 1990. In India, women face an appallingly high
million deaths in 1990 were due to this group of causes, and more than risk of dying in fires: in 1990 alone, more than 87000 Indian women died
16.5 million of these were in developing regions, mainly India and Sub - this way. In Sub-Saharan Africa, by contrast, the most important cause
Saharan Africa. Of all Group I deaths, 4 out of 10 were due to either of injury deaths for both women and men is war.
pneumonia or diarrhoeal disease, which together accounted for more than
7 million deaths. Perinatal conditions were responsible for more than 2.4 2..3 RESULTS: THE TOLL OF PREMATURE DEATH

million deaths, and tuberculosis another 2.0 million deaths. The vast In contrast to crude numbers of deaths, a time-based measure such as years
majority of these deaths could have been prevented with existing of life lost (YLLs, see Section 1 above) allows public health researchers to
interventions. identify those causes that account for premature deaths, a more
Because Group I conditions affect children disproportionately, the age informative measure for the design of health policies. For example, injuries
structure of deaths also varies sharply between regions. A baby girl born affect mainly young people so, while injuries accounted for only 10 per
in Sub-Saharan Africa faces a 22 per cent risk of death before age 15. In cent of deaths in 1990, they accounted for 15 per cent of YLLs.
China the risk is less than 5 per cent and in the Established Market Accordingly noncommunicable diseases, which affect mainly older people,
Economies the risk is just 1.1 per cent. accounted for only 31 per cent of YLLs, compared with 56 per cent of
deaths (Figure 10).

Figure 10 The leading causes of death and of premature death.


Table 2 The ten leading causes of death. 1990 world, 1990
Voll P 178-179 Percent of total deaths or YLLs
Developed Regions Developing Regions Causes 0 5 10
Deaths Cumu- Deaths Cumu- Ischaemic hear, di$ea5e Iiiii. iiiiliiiiiiiiiiiiiiiiliiiiii••iiilliiiiiiiiii.iiiiiiliiiiiiiiiiiiii' l
('OOOs) latlve" ('OOOs) latlve ~ CerebrovaKubr disease jiii illlIiI _
All Causes All Causes 39 554

i~~~~~E~~~~±~=t--'
Ischaemic heart disease Lower respiratory infections 3915 9.9
2 Cerebrovascular disease 2 Ischaemic heart disease 3 565 1 18.9 Diarrhoealinfections
Lower respiratory diseases
3 Trachea. bronchus and 3 Cerebrovascular disease 2 954 26.01
lung cancer Conditions ansing during the perinatal period

"I Lower respiratory 385 46.1 "I Diarrhoeal diseases 2940 33.8 Chroni( obsuuc.tive: pulmonary disease
infections
5 Chronic obstructive 324 49.1 5 Conditions arising during 2361 38.7 Tuberculosis ~~~
pulmonary disease the F'erinatal period Measles
6 Colon and rectum cancers 277 I 51.6 6 Tuberculosis 1922 43.01 Deaths _
Road tnffic accidents " "- ""\;._
7 Stomach cancer 241 53.8 7 Chronic obstructive 1887 46.1 YLLs _
pulmonary disease Trachea.bronchus and lung cancers

8 Road traffic accidents 222 55 .8 8 Measles 1058 48.7 Malaria ~:::::::~:...J


9 Self·lnflicted injur Je,-s_ _......;.19-.;3;;.., 57.6
10 Diabetes mellitus 176 59.2
9 Malaria
10 Road traffic accidents
856
777
50 .9
52.8 Congenitalanomalies E===_.:..---=-_i.-....:...~ __-.i_ L.. .:. ......::....J
20 The Global Burden of Disease SUMMARY 2.I

3. DISABILITY: THE INVISIBLE BURDEN explicitly considered were made on the basis ofinformation on the ratio
of total premature mortality (years of life lost) to disability (YLDs) for each
broad cause Group.
3.1 How THE GBD ASSESSED DISABILITY
A disease or injury may have multiple disabling effects, or sequelae. For 3.2. RESULTS: THE UNSEEN BURDEN OF PSYCHIATRIC DISEASE
example, diabetes may result in diabetic foot, retinopathy, neuropathy, or The GBD's findings demonstrate clearly that disability plays a central role
amputation. To estimate the total burden of disability, the study measured in determining the overall health status of a population. Yet that role has
the amount of time lived with each of the various disabling sequelae of until now been almost invisible to public health. The leading causes of
diseases and injuries, in both treated and untreated states, and weighted disability are shown to be substantially different from the leading causes
for their severity, in each population. In all, 483 disabling sequelae of of death, thus casting serious doubt on the practice of judging a
diseases and injuries were analysed for this study, for all regions and age population's health from its mortality statistics alone.
groups, and for both sexes.
To calculate the number of years lived with a disabling condition, it is Mental Illnesses
necessary to know its incidence, the average age of its onset, the average Most significantly, the study shows that the burden of psychiatric
duration of the disability (whether just months or lifelong) and the severity conditions has been heavily underestimated. Of the ten leading causes of
weight for the condition. The GBD researchers requested epidemiological disability worldwide in 1990, measured in years lived with a disability, five
specialists to estimate each of these variables for each condition based on were psychiatric conditions: unipolar depression, alcohol use, bipolar
an in-depth review of published and unpublished studies. They also affective disorder (manic depression) schizophrenia and obsessive-
estimated the prevalence of each sequela, its case-fatality rates, remission compulsive disorder. Unipolar depression alone was responsible for more
rates and death rates. This information allowed them to correct the than one in every ten years of life lived with a disability worldwide.
preliminary estimates for internal consistency-that is, to ensure that Altogether, psychiatric and neurological conditions accounted for 28 per
...
Cl)
estimated prevalence was consistent with estimated incidence and vice cent of all YLDs, compared with 1.4 per cent of all deaths and 1.1 per cent
e versa. Consistency was validated using a model developed specially for the of years of life lost . The predominance of these conditions is by no means
~
;>
study, known as DISMoD. Frequently, inconsistencies were detected, and restricted to the rich countries, although their burden is highest in the
epidemiological specialists were asked to revise their initial estimates. Established Market Economies. They were the most important contributor
'"
cu
Overall, the team took the disability estimates through three complete to YLDs in all regions except Sub-Saharan Africa, where they accounted
.~
t'Il
e rounds of revision in a process lasting nearly five years. for a relatively modest 16 per cent of the total.
e The severity weights of different conditions were set using the protocol Alcohol use is the leading cause of male disability-and the tenth largest
=
'"
Cl
described in Section 1. A set of 22 indicator conditions ranging from mild in women-in the developed regions. More surprisingly, perhaps, it is also
o to severe disabilities were assigned to seven classes of severity weight the fourth largest cause in men in developing regions. The remaining
.~
between 0 (perfect health) and 1 (equivalent to death). Following this important causes of YLDs were anaemia, falls, road traffic accidents,
'" exercise, each remaining condition was assigned a distribution across these chronic obstructive pulmonary disease and osteoarrhritis (Table 3).
'"
~ seven classes.

,
Table 3 The leading causes of disability, world, 1990
The number of years lived with a given disability for each individual -------
were calculated from the incidence of the disability, with the "stream" of Total Percent of
(millions) total
disability arising from it measured from the age of onset, for the estimated t All Causes 472.7
duration of the disability, multiplied by the condition's severity weight.
To calculate the YLDs due to a condition in any given population, the I Unipolar major depression 50.8 10.7
2 Iron-deficiency anaemia 22.0 4.7
number of YLDs lost per incident case must be multiplied by the number
3 Falls 22.0 4.6
of incident cases. For example, a case of asthma carries a disability weight .. Alcohol use 15.8 3.3
of 0.1 if untreated and 0.06 if treated. If the incidence of asthma in males 5 Chronic obstructive pulmonary disease 14.7 3.1
aged 15-44 years is 1 million, the untreated proportion is 35 per cent, and 6 B·polar disorder 14.1 3.0
the average duration is seven years, then this sequela alone is estimated 7 Congenital anomalies 13.5 2.9
to cause 664 000 YLDs in a given year for that demographic group. 8 Osteoarthritis 13.3 2.8
Unlike the estimates of years of life lost, not all sequelae of all conditions 9 Schizophrenia 12.1 2.6
10 Obsessive-compulsive disorders 10.2 2.2
could be explicitly assessed for YLDs. Estimates for conditions not
2.2. The Global Burden of Disease SUMMARY 23

3.3 RESULTS: LONGER LlFESPAN MEANS LONGER "HEALTH SPAN" TOO 4. THE GLOBAL BURDEN OF DISEASE IN 1990
The GBD provides support for the theory that people in the high-income,
low-mortality populations of the Established Market Economies not only A key aim of the GBD was to quantify the combined burden of fatal and
live longer, but remain healthier for longer too. In recent years , researchers non-fatal health outcomes in a single measure, the Disability-Adjusted Life
have been divided between those who say that ill health is " compressed " Year (DALY). Thi s sect ion presents the key results of the assessments of
into the last few years of life in these populations, and those who argue overall burden for each region.
that longer life merely exposes people to a longer period of poor health. The methods for developing the DALY are described in section 1. To
The new results suggest that older people in the developed world are calculate DALYs due to each disease or injury in a given year and
healthier than their counterparts in developing countries (Figure 11). population, the researchers added together: (a) the years of life lost through
They found that babies born in Sub -Saharan Africa can expect to spend all deaths in that year, and (b) the years of life expected to be lived with
about 15 per cent of their lifespan disabled, compared with just 8 per cent a disability for all cases beginning in that year summed, and weighted for
for babies born in the Established Market Economies. A 60 year-old in the severity of the condition.
Sub-Saharan Africa can expect to spend about half his or her remaining
years with a disability, whereas a 60 year-old in the Established Market 4.1 RESULTS : STARK REGIONAL IMBALANCES IN THE BURDEN OF

Economies is likely to spend just one-fifth of those years disabled. These DISEASE
results suggest that the proportion of the lifespan lived with a disability The peoples of Sub-Saharan Africa and India together bore more than four-
falls as life expectancy rises. tenths of the total global burden of disease in 1990, although they make
up only 26 per cent of the world's population. By contrast, the Establ ished
Market Economies and the Formerly Socialist Economies of Europe, with
Figure 11 Proportion of expected remaining Iifespan at age 60 that will about a fifth of the world's population between them, together bore less
be lived disabled. by region. than 12 per cent of the total disease burden. China emerged as substantially
the most "healthy" of the developing regions , with 15 per cent of the global
Voll p 232
disease burden and a fifth of the world's population. Put differently, about
579 years of healthy life were lost for every 1000 people in Sub-Saharan
60 Africa, compared with just 124 for every 1000 people in the Established
"'0
Cll ....., Market Economies. This assessment demonstrates clearly the glaring
::0 50 0-
~ inequalities of world health at the end of the 20th century.
~ The rates of premature death varied sharply between regions, with rates
c 40 I-
a
VI
.... .... 7 times higher in Sub -Saharan Africa than in the Established Market
~ 30 I- Economies. By contrast, the rates of disability were less varied, with
...o Sub-Saharan Africa having twice the rate of YLDs as the rich countries
...
c
20 1- ....
(Figure 12).
Cll
V
1.. 10 I-
rf. Figure 12 DALYs per 1000 population by region broken down into
o ~ w.... LL I YLL and YLD rates. 1990
EME FSE CHN LAC OAI MEC IND SSA
Vol I P 255 600
Region
500
• Males • Females
g
o 400

~ 300
a.
s"' 200 YLD rate
IX:
100

EME FSE CHN LAC OAI


Region
The Global Burden of Disease SUMMARY

Traditional Enemies Remain a Significant Force Figure 14 The relationship between the rank order of causes of global
The grip of the traditional enemies-eommunicable, maternal, perinatal ill-health when measured using deaths alone or total disease
conditions and nutritional deficiencies-persists as a problem for the whole
Vol I P 287 burden. 1990
Lower respiratory
world. Even though these Group I conditions accounted for only 7 per cent Tuberculosis in(ee:tions
of the burden in the Established Market Economies and less than 9 per 100
o
Unipolar major depreSSion
Malaria\ ~/_15thaemit
cent in the Former Socialist Economies, they nevertheless made up more Anaemia Falls \ .'I'lI t'P heart disease
[J [J 1:1'"'"[J 'Stroke
than four-tenths the total global burden of disease in 1990, and almost Bipolar disorder Akohol use IJ [J ~01 [J COPD
...... 80 Osteoar:hritis [J 0 HIV\. 1:1 l:r'"'
half of the burden (49 per cent) in developing regions. In Sub -Saharan l-
Q) o 0
n [JO 0
Afr ica, two out of three years of healthy life lost were due to Group I "E []
IJ [] 0 [J [J Trachea, bronchus
o 60 n o o [J \ and lung cancers
conditions. Even in China, where the epidemiological transition is far ..::,(, [J
IJ 0 0
c [J Diabetes
advanced, a quarter of years of healthy life lost were due to this Group. o
~ o [] [J [J 1:1
Worldwide, five out of the ten leading causes of disease burden are Group VI 40 I:l [J
00 [] [J

I conditions: lower respiratory infections (pneumonia); diarrhoeal disease; ~ n 000


o IJ []
IJ
perinatal conditions; tuberculosis and measles. In developing countries, ~ I:l
o [] 0 Co [] Pancreas cancer
20 o 0 0 0
malaria is added to this already daunting list (Figure 13). [] 0 o I:l 0 Prostate cancer
t'P 0 Bladder cancer
Injuries Are a Large, and Neglected, Health Problem in All Regions
20 40 60 80 100
The burden of injury in 1990 was highest in the Formerly Socialist
Deaths (rank order)
Economies of Europe, where almost 19 per cent of all burden was
attributed to this group of causes. China had the second highest injury --------- - - -
burden, Latin America and the Caribbean the third, and Sub-Saharan population's health status. While a few leading conditions-such as lower
Africa the fourth. Even in the Established Market Economies, however, respiratory infections, diarrhoeal diseases and perinatal conditions-are
the burden of injuries-dominated by road traffic accidents-was almost at the top of both lists, there are 14 conditions in the top half of the list
12 per cent of the total. for disease burden that are in the bottom half of the list for deaths.
In almost all regions, unintentional injuries were a much bigger source Depression is the most marked of these, falling within the top ten for
of ill-health in 1990 than intentional injuries such as interpersonal violence disease burden, but the bottom ten for deaths (Figure 14).
and war. The only exception was the Middle Eastern Crescent, where The leading causes of disease burden worldwide in 1990 were broadly
unintentional and intentional injuries took an approximately equal toll similar to those for the developing regions. (Table 4).
because of a particularly high burden of war in the region at the time.
Sex Differences in Disease Burden
Leading Causes of Disease Burden Although in infancy and early childhood, girls and boys suffer from
When causes of death are compared, in rank order, with causes of disease broadly similar health problems, striking sex differences emerge in adults.
burden, substantial differences emerge, again reinforcing the need to take First, and most obviously, women suffer disproportionately from their
non-fatal conditions into account as well as deaths when assessing a reproductive role. Although the burden of reproductive ill-health is almost
entirely confined to the developing regions, it is so great that even
Figure 13 The burden of disease. by broad cause Group. 1990 worldwide, maternal conditions make up three out of the ten leading
Vol I P 257 100 causes of disease burden in women aged between 15 and 44. In developing
regions, five out the ten leading causes of DALYs are related to
:5 80 reproductive ill-health, including the consequences of unsafe abortion and
(§ • Group I
chlamydia. Almost all of this loss of healthy life is avoidable.
B 60
• Group 11 However, poor reproductive health is far from being women's sole
B
--...
o
C
-40 • Group III
concern (Figure 15). In both developing and developed regions, depression
is women's leading cause of disease burden. In developing regions, suicide
~L. 20 is the fourth. Thus, while programmes to reduce the unacceptably high
If burden of poor reproductive health must remain a high priority for years
o EME FSE CHN tAC OAI MEC IND SSA to come, women's psychological health also deserves much more attention.
Region
The Global Burden of Disease SUMMARY 2

Table 4 Ten leading causes of disease burden (DALYs). developing 5. RISK FACTORS FOR DEATH AND DISABILITY
world. 1990
Total Per cent of Exposure to particular hazards, such as tobacco, alcohol, unsafe sex or
(millions) total poor sanitation, can significantly increase individuals' risks of developing
All Causes 1218.2 disease. These hazards, or risk factors, are significant contributors to the
I Lower respiratory infections 110.5 9.1 total global disease burden and health policy makers need accurate
2 Diarrhoeal diseases 99.2 8.1
information on their impact if they are to dev ise effective prevention
3 Conditions arising during the perinatal period 89.2 7.3
4 Unipolar major depression 41.0 3.4
strategies. Until now, however, there have been few attempts to measure
5 Tuberculosis 37.9 3.1 the burdens of these risk factors, or to express them in a currency that can
6 Measles 36.5 3.0 be compared directly with the burdens of individual diseases.
7 Malaria 31.7 2.6 The GBD researchers have sought to overcome this problem. They have
8 Ischaemicheart disease 30.7 25 assessed, for the first time, the mortality and loss of healthy life that can
9 Congenital anomalies 29.4 24 be attributed to each of ten major risk factors in each region. These risk
10 Cerebrovascular disease 29.1 2.4
factors are: malnutrition; poor
water supply, sanitation and
For men aged 15-44, road traffic accidents are the biggest cause of ill- personal/domestic hygiene; unsafe
health and premature death worldwide, and the second biggest in sex; tobacco use; alcohol use;
Unsafe sex was the third
developing regions, surpassed only by depression. Alcohol use, violence, occupation (that is, exposure to biggest risk factor for
tuberculosis, war, bipolar affective disorder, suicide, schizophrenia and hazards through work);
iron-deficiency anaem ia make up the remainder of the list in developing hypertension; physical inactivity;
disease burden.
countries. The high toll of road traffic accidents in developing regions has illicit drug use; and air pollution. In young adult women
received relatively little attention from public health specialists in the past.
5 .1 How THE BURDEN OF RISK in Sub-Saharan Africa,
Figure 15 The ten leading causes of disease burden for women , FACTORS WAS ASSESSED
aged 15-44. 1990 it accounts for an
The burden of disease or injury in
Percent of all causes in developed or developing regions a population today that can be
estimated 30 per cent
Causes 0 5 10 15 20 attributed to past exposure to a of the total burden.
Un ipolar major depression liii~~~_~ __
iiiii~iiii~iiir=t--T~: given risk factor is, essentially, an
Schizophre ni. C:=:"'::::=';= :;=-l estimate of the burden that could
Road traffic. accidents L..- ---
have been averted in the population if that particular risk factor had been
eliminated. More precisely, this is defined as the difference between the
Bipolar disorder ~_IIIIIII("
currently observed burden and the burden that would be observed if past
Obsessiveo<ompulsive disorder
levels of exposure had been equal to a specified, reference distribution of
Alcohol us.
exposure. In general, to calculate this, it is necessary to know: (a) the
On:eoannrj(is relative risk at different levels of exposure for each cause of death and
Chl.mydi. disability linked to the factor; (blrhe distribution of different levels of
Self~nflit(ed injuries exposure in the population; and (c) the burden of disease or injury due to
Rheunutoid arth ritis each of the causes linked to the factor . Depending on the nature of the risk
Developed _ factor, the reference distribution against which relative risk is compared
Developing _ could be zero exposure for the whole population, a population distribution
Ir"Of'l-deficienq anaemia ~. __
of exposure from low to high levels based on observed populations, or an
Obstructed I.bour ~. ._ .
arbitrary distribution. For this study , the researchers used, wherever
Maternal se psis ~_-= possible, zero exposure as the reference, except for risk factors such as
hypertension, where clearly no one can be said to exposed to " zero " levels.
SUMMARY
The Global Burden of Disease

Figure 16 Male deaths attributable to, and averted by. alcohol use.
Table 5 Global burden of disease and injury attributable to selected
risk factors, 1990 Voll p319
Established Market Economies (EME)
As%of
total 100
Risk factor DALYs 'Ul
"'0
Malnutrition 5881 15.9 lij
III
SO
:l
Poor water supply 2668 6.8 o
sanitation and personal
and domestic hygiene :E.
Unsafe sex 1095 2.2 27602 3.0 21 100 4.5 48702 3.5
Tobacco 3038 6.0 26217 2.9 9965 2.1 36182 2.6
Alcoho 774 1.5 19287 2.1 28400 6.0 47687 3.5
Occupation 1129 2.2 22493 2.5 15394 3.3 37887 2.7 15-29 30-44 45-59 60-69 70+
Age group (years)
Hypertension 2918 5.8 17665 1.9 1411 0.3 19076 1.4
Physical inactivity 1991 3.9 11353 1.3 2300 0.5 13653 1.0
0.3 5834 8467 0.6
Sub-Saharan Africa (SSA)
Illicitdrugs 100 0.2 2634 1.2 100
Air pollution 568 1.1 5625 0.6 1630 0.3 7254 0.5 'Ul
"'0
C 75
Source: Authors' estimates are based on data and informatio n contained in individualchapters of Volume IX
"'
III

o so
of the Global Burden of Disl!OSe and Injury Series. :l

:E.
III
25
s:
5.2 RESULTS: THE CONTRIBUTIONS OF RISK FACTORS TO GLOBAL
BURDEN
ti o l.L-""",--i,,,;~.,}....-F-~i-E:=::::::H~~
c
Of the ten risk factors studied, the most significant were malnutrition, poor -25 15-29 30-44 45-59 60-69 70+
water, sanitation and hygiene, unsafe sex, alcohol, tobacco and Age group (years)
occupation. Together, these six hazards accounted for more than one-third Injury - Disease - Protective effects
of total disease burden worldwide in 1990 (see Table 5). Of the six, (on ischaemic heart disease)
malnutrition and poor sanitation were the dominant hazards, responsible
for almost a quarter of the global burden between them. Unsafe sex and
alcohol each contributed approximately 3.5 per cent of the total disease The impact of alcohol varies between regions not only because of
burden, closely followed by tobacco and occupation hazards with just different levels of use in each population, but also because of differences
under 3 per cent each . These are comparable to the burdens caused by in the age structure of those populations (Figure 16). Alcohol has
tuberculosis and measles. consistently been shown to provide some protection against death from
Not surprisingly, the researchers found sharp inequalities between ischaemic heart disease, but to increase the risk of several other diseases,
regions and between men and women in the burdens of most risk factors . such as alcoholic psychoses, pancreatitis, some cancers and cirrhosis of the
For example, the ill-health consequences of unsafe sex-which include liver, as well as many injuries. Because of its protective effect against
both infections and the complications of unwanted pregnancy-are borne ischaemic heart disease, in populations where this condition is common
disproportionately by women in all regions. In young adult women in Sub- and injuries and violence are rare, alcohol may prevent about as many
Saharan Africa, unsafe sex accounts for almost one -third of the total deaths as it causes. In the Established Market Economies, for instance, this
disease burden. is probably the case. Nevertheless, alcohol causes a severe disease burden
Tobacco and alcohol currentl y cause their heaviest burdens in men in in these rich countries, because it causes so many injuries and premature
the developed regions. In these regions, the two together accounted for deaths and thus results in large numbers of years lived with a disability
more than one-fifth of the total burden in 1990. However, the health and years of life lost.
burdens of smoking and drinking are far from being the exclusive preserve In Sub-Saharan Africa, the picture is very different. There, ischaemic
of the industrialized world . The recent rapid increase in tobacco use in heart disease is relatively uncommon, so the protective effect of alcohol
Asia and other developing regions is expected to kill many more people is far outweighed by its harmful effects in increasing the rates of death and
in the coming decades than have so far died in the developed regions (see disability from injuries. The contribution of alcohol to injuries is also
Section 6).
3° The Global Burden o{Disease SUMMARY 31

extremely high in Latin America and the Caribbean, where alcohol use 6.LOOKING AHEAD: THE HEALTH OF THE WORLD IN
accounts for almost 10 per cent of total disease and injury burden, a figure 2020
surpassed only in the developed regions. Ultimately, alcohol is estimated
to have caused about three-quarters of a million more deaths in 1990 than To plan health services effectively, policy-makers need to know how
it averted, with more than four-fifths of the excess deaths in the developing current health needs could develop in the future. For this study, the GBD
regions. developed projections of mortality and disability for each five year period
from 1990 to 2020, by cause, for all regions and both sexes. The findings
5.3 SOME DISEASES ARE THEMSELVES POWERFUL RISK FACTORS FOR
have striking implications for public policy.
OTHER DISEASES
While assessments of the burdens of risk factors such as tobacco and 6 .1 How THE PROJECTIONS WERE DEVELOPED

alcohol can help to guide priorities, it is also useful to think about the The GBD researchers used a set of relatively simple models to develop
impact of certain diseases as risk factors for other diseases. For example, projections of future health trends. Rather than attempt to model the
diabetes mellitus strongly increases an individual's risk of developing effects of the many separate direct, or proximal, determinants of disease
ischaemic heart disease and stroke, while infection with hepatitis B virus from the limited data that are available, it was decided to consider a limited
increases the risk of developing liver cancer and cirrhosis of the liver. number of socio-economic variables: (1) income per capita; (2) the average
Traditional methods of assessing deaths by cause fail to capture these number of years of schooling in adults, termed "human capital"; and (3)
relationships. The GBD researchers have attempted to overcome this for time , a proxy measure for the secular improvement in health this century
a short list of well-studied conditons. They treated each condition as a risk that results in part from accumulating knowledge and technological
factor and estimated how much of the total disease burden would development. These socio-economic variables show clear historical
be averted in each region's population if the condition were eliminated relationships with mortality rates: for example, income growth is closely
(T able 6). The most dramatic differences between directly coded and total related to the improvement in life expectancy that many countries have
burden are for diabetes and hepatitis B and hepatitis C.

Figure 17 The rich get richer: baseline projections of income per


Table 6. Total burden for major diseases. world, 1990 capita by region. 1990-2020
Deaths Voll pHO
Both sexes Males Females DIr-~"t!
(thousands) (thousands) (thousands) total 35
Chagas disease <49.2 28.0 21.2 0.39
559.6 259.1 0.13 ~
Hepatitis B and hepatitis C 818.7 30
(ij
Diabetes mellitus 2758.9 1279." 1"79.6 0.21 c: - EME
Cataracts 1 10404 5.....0 560.4 0.01 0
'i:i
0.02 c<l
Glaucoma 330.5 1.. 1... 189.1 c: 25 - - FSE
Onchocerciasis
Trachoma
Unipolar major depression
18.6
103.1
786.2
11 .8
29.4
"56."
6.8
73.7
329.9
000
0.00
0.00 -"
"-
...C
Q)

cc:
-
In
20
IND

C!'..s CHN
Se>eually transm itted diseases "13.3 183.0 230.3 0.56 0-'"

DALYs
Both sexes
(millions)
Males
(millions)
Females
(millions)
Direct!
total
._-
-:=;,
-0
~-5
Q.
..s
u
IS OAI
SSA
"- 10 LAC
Chagas disease 1.6 0.9 0.7 GAO
~.... ;Ci
Q)

._._ .
Q.
Hepat itis B and hepatitis C 13.3 9.2 ".1 0.16
Diabetes mellitus 26.3 12.9 13.'1 GA2
a-
S ~1iO· - MEC
Cl ~~-
Cataracts 17.9 8.6 9.2 0.42 o "=::::
Glaucoma
Onchocerciasis
5.8
1.2
2.3
0.7
3.5
0.5
0.....
0.76 0 --
;:1" - -•.•.• • .,.. .. O"c:ft
=WB ~. ~~ ~~ . _ . _ . _ . _ . _

Trachoma 2.0 0.6 lA 0.51 1990 1995 2000 2005 2010 2015 2020
Unipolar major depression 69.8 28.3 41.5 0.73 Years
Se>eually transmitted diseases 25.3 9.1 16.2 0.7<4
SUMMARY
The Global Burden of Disease 33

Figure 18 Projected life expectancy at birth in 2020, by region:


achieved this century, Because of their relationships with death rates, these
baseline, optimistic and pessimistic scenarios, compared
socioeconomic variables may be regarded as indirect, or distal ,
with 1990 estimates
determinants of health. In addition, a fourth variable, tobacco use, was
included, because of its overwhelming impact on health status, using Voll P 354
information from more than four decades of research on the time lag
between persistent tobacco use-measured in terms of "smoking
* Estimated life expectanc ies in 1990

intensity"- and its effects on health. 90


Males t Projected life expectancies in 2020

Death rates for all major causes based on historical data for 47 countries
since 1950-91 were related to these four variables to generate the
85 -
projections. A separate model was used for HIV and modifications for the .-. 80
~
-T
interaction between HIV and tuberculosis. Three projection scenarios were ~ 75
developed using different projections of the independent variables. ob
~ 70
*
-
6 .2 RESULTS: PATTERNS OF DEATH-AND LIFE-IN THE 21ST CENTURY

!
e 67
c

v
60
... T T
t
.. - I
r
Life Expectancy Grows Almost Everywhere, but Men Fare Worse t
Life expectancy at birth is expected to grow for women in all regions. By
:5 55
50
2020, infant girls born in the Established Market Economies may expect
to survive to almost 88 years (Figure 18). For men, life expectancy will 45 *
grow much more slowly, mainly because of the impact of the tobacco 40
I , I
• , I I
epidemic. Nevertheless by 2020, males born in Sub-Saharan Africa, whose EME FSE CHN LAC OAI MEC IND SSA
life expectancy at birth was below 50 in 1990, may expect to reach 58
Region
years. Males born in Latin America and the Caribbean, who in 1990 could
have expected to live to 65, may expect to reach 71 years. Estimated life expectancies in J990
However, for men in the Formerly Socialist Economies of Europe, life Females
expectancy is not expected to grow at all between 1990 and 2020. This is 90 + Projected life expectancies in 2020

partly due to the fact that it has actually fallen between 1990 and 1995, 85
so that any positive change is likely to be merely recovering to the 1990
.-. 80
position.
...
III

t
In young children and adolescents under the age of 15, the risk of death '"
QJ 75
is projected to decline dramatically in all regions, falling by about two- ob
~ 70
thirds in Sub-Saharan Africa and India . In adult women, too , the risk of c
death is expected to fall in all regions. For men, the picture is more
~
v
QJ
67 •
complex. Because of the tobacco epidemic, men in the Formerly Socialist ! 60
Economies of Europe and China may expect a higher risk of dying between ~ 55
*
::i
the ages of 15 and 60 than they do today. In other regions, the risk of death
for men in this age group is expected to fall, but more modestly than in
50 ..
women . Remarkably, by 2020, men of this age group in the Formerly 45
Socialist Economies of Europe could face a higher risk of death even than 40
men in Sub -Saharan Africa. EME FSE CHN LAC OA) MEC IND SSA
The Impact of Infectious Disease May Be Reduced Region

Deaths from communicable, maternal and perinatal conditions and


Note: The stars in the figure denote life expectancies at birth in 1990. Baseline life expectancies projected for
nutritional deficiencies (Group I) are expected to fall from 17.3 million 2020 appear as a horizontal bar crossed by a vertica bar that gives upper and lower limits defined by the
in 1990 to 10.3 million in 2020. As a percentage of the total burden, Group optim istic and pess imistic projection scenarios.

1 conditions are expected to drop by more than half, from 34 per cent to
15 per cent.
34 The Global Burden of Diseas e SUMMARY
35

This projected reduction will not keep pace with the emergence of drug-resistant strains of
overall, despite increased burdens Even under the important microbes such as Mycobacterium tuberculosis. If such a
due to HIV and tuberculosis, runs frightening scenario were to prove correct, and if, in addition, case-fatality
pessimistic scenario,
counter to the now widely - rates were to rise because of such drug-resistant strains, the gains of the
accepted belief that infectious deaths from infectious present century could be halted or even reversed. Undoubtedly, the
diseases are making a comeback continuing high toll of Group I causes today leaves no room for
diseases, maternal and
worldwide. It reflects, in part, the complacency. Nonetheless, the evidence to date suggests that, as long as,
relative contraction of the world's perinatal conditions and and only if, current efforts are maintained, Group I causes are likely to
"young" population: the under-IS continue to decline.
nutritional deficiencies
age group is expected to grow by
only 22 per cent between 1990 are expected to fall HIV AND TUBERCULOSIS

and 2020, whereas the cohort of Projections for HIV mortality were developed using a separate model. The
slightly. GBD projections for HIV demonstrate that the death toll from the AIDS
adults aged between 15 and 60 is
expected to grow by more than 55 epidemic may be even greater than feared in future.
per cent. In addition, the In Sub-Saharan Africa, death rates from HIVlAIDS are expected to peak
projection reflects the observed overall decline in Group I conditions over around 2005 with around 800 000 deaths per year. In India, death rates
the past four decades, due to increased income, education and are expected to peak a little later, around 2010, at about half a million a
technological progress in the development of antimicrobials and vaccines . year . The worldwide peak for HIV deaths is expected to be around 2006,
Even under the pessimistic scenario, in which both income growth and with perhaps 1.7 million deaths that year. Clearly, these estimates are
technological progress are expected to be minimal , deaths from these subject to considerable uncertainty.
conditions are still expected to fall slightly to 16.9 million (Figure 19). Projections of the future impact of tuberculosis are also subject to large
Clearly, it should not be taken for granted that the progress of the past uncertainties, but the results are no more reassuring than those for AIDS.
four decades against infectious diseases will be maintained. It is possible, The figure shows baseline, optimistic and pessimistic projections of the
for example, that antibiotic development and other control technologies deaths the disease is expected to cause between 1990 and 2020 (Figure 20 ).

Figure 19 Pessimistic projection of deaths from Group 1causes, world, Figure 20 Projections of deaths from tuberculosis. world. 1990-2020
1990-2020
3500
Vol I P 360
- - - Optimistic scenario
Baseline scenario
20 - ...... 3000 - - - Pessimistic scenario
VI
"U
C
...... 15 I- '"
VI
:l

~
VI
c
.s
~
2500
], 10 I-
VI
-£III
VI
...
..c.
III Q
Ql
2000
Ql
Cl 5-

1500
0 1990 2000 2010 2020
Year
_ Infectious _ Respiratory _ Perinatal _ Nutritional _ Maternal
0
infections 1990 2000 2010 2020
The Global Burden of Disease SUMMARY 37

The Rise of Noncommunicable Diseases However, any age-specific decrease in the rates of these diseases that may
While overall, Group I conditions are expected to decline, deaths from also emerge in low-income countries is likely to be outweighed by the large
noncommunicable diseases (Group 11) are expected to climb from 28.1 and demographically driven increase in the absolute numbers of adults at
million deaths in 1990 to 49 .7 million in 2020, an increase of 77 per cent risk for these diseases, augmented by the tobacco epidemic.
in absolute numbers. In proportionate terms , Group II deaths are expected As with noncommunicable diseases, deaths from injury are also
to increase their share of the total from 55 per cent in 990 to 73 per cent expected to rise for mainly demographic reasons. Young adults are
in 2020. These global figures, impressive as they are, mask the extreme generally exposed to greater risks of injury.
nature of the change that is projected in some developing regions because
6.2 THE BURDEN OF DISEASE IN 2020
they incorporate the projections for the rich nations, which show little
change. In India, deaths from noncommunicable diseases are projected to When disability is taken into account as well as death, a different view of
almost double, from about 4 million to about 8 million a year, while Group the future emerges - and one that emphasizes adult health problems still
I deaths are expected to fall from almost 5 million to below 3 million a further. By 2020, the disease burden due to communicable, maternal and
year. In the developing world as a whole, deaths from noncommunicable perinatal conditions and nutritional deficiencies (Group I) is expected to
diseases are expected to rise from 47 per cent of the burden to almost 70 fall to a fifth of the total. The burden attributable to Group 11, accordingly,
per cent. is expected to rise sharply, and the burden from injuries is also expected
The steep projected increase in the burden of noncommunicable diseases to rise to that of Group I conditions (Figure 22).
worldwide is largely driven by population aging, augmented by the large
numbers of people in developing regions who are now exposed to tobacco. Mental Health: Unmet and Growing Needs
It is important to stress that aging will result in a rise in the absolute As with the 1990 assessments, psychiatric diseases emerge as a highly
numbers of cases of noncommunicable diseases and in their increased share significant component of global disease burden when disability, as well
of the total disease burden for the population as a whole, but not in any as death, is taken into account. The projections show that psychiatric and
change to the rates of those diseases in any given age group. neurological conditions could increase their share of the total global
As studies in the Established Market Economies show, the age-specific burden by almost half, from 10.5 per cent of the total burden to almost
rates of some important noncommunicable diseases, such as ischaemic 15 per cent in 2020. This is a bigger proportionate increase than that for
heart disease and stroke, have been falling steadily for at least two decades. cardiovascular diseases.
Whether these rates are also falling in other regions is much less clear.

Figure 22 Change in the distribution of DALYS. by broad cause


Figure 21 The rise of noncommunicable diseases: Group 11 deaths by Group. developing regions. 1990-2020.
causes. world . 1990-2020 ----- - - - - - -
Vol I P 359

60 Group 11 deaths 1990 2020

'Vi' (17.6%)
.~ 40
]. (68.7%)
."

~ 20 (10.7%)
Cl

o 1990 2000 2010 2020


Year
_ Cardiovascular _ Cancer _ Respiratory • Digestive • Other _ Group I _ Group 11 Group III
The Global Burden of Disease SUMMARY 39

Tobacco's Legacy CONCLUSION


By 2020, the burden of disease attributable m tobacco is expected m The GBD study has provided a new and much needed picture of current
outweigh that caused by any single disease. From its 1990 level of 2.6 per and projected health needs. In particular, it has shown that
cent of all disease burden worldwide, tobacco is expected to increase its noncommunicable diseases are rapidly becoming the dominant causes of
share m just under 9 per cent of the total burden in 2020, compared with ill-health in all developing regions except Sub-Saharan Africa; it has
just under 6 per cent for ischaemic heart disease, the leading projected revealed the extent to which mental health problems have been
disease. This is a global health emergency that many governments have underestimated worldwide; and it has shown the significance of injuries
yet to confront. as a problem for the health sector in all regions. The findings pose new
and immediate challenges to policy-makers and are certain to provoke
Leading Causes of Disease Burden in 2020
debate. Ultimately, the study's impact will be judged in two ways: first,
Figure 2 on page 8 shows how the global pattern of disease burden is by the degree to which it stimulates other researchers m apply the same
expected m shift over the next 24 years. In 1990, the three leading causes rigorous methods of measuring disease burden in all regions; and second,
of disease burden were, in descending order, pneumonia, diarrhoea I m the extent that it changes priorities for public health in the decades
diseases and perinatal conditions. The three conditions projected to take ahead.
their place by 2020 are ischaemic heart disease, depression and road traffic
accidents. Pneumonia is expected to fall to sixth place, diarrhoeal diseases
m ninth and perinatal conditions m eleventh. Notably, measles, currently
in eighth place, is expected to drop m twenty -fifth. However, not all
infectious diseases are expected m decline, despite the projected overall
collapse of Group I conditions. Tuberculosis is expected to remain at its
current level of seventh place, a substantial source of disease burden for
the foreseeable future. Of equally great concern is the finding that HIV,
currently twenty-eighth in the ranking, could be as high as tenth by 2020.
Road Traffic Accidents and Violence
Because of the growth of the adult fraction of the population, the burdens
of several important types of injury are also likely to increase. For example,
young men are the group most frequently involved in road traffic accidents,
so if the young-adult proportion of the population increases sharply, road
traffic accidents are likely to increase too. Indeed, according to the baseline
projection, road traffic accidents could rise to third place from ninth
worldwide. Violence, currently nineteenth, could rise as high as twelfth
place and suicide could climb from seventeenth m fourteenth place.
Not surprisingly, these changes are not expected to be evenly dispersed
worldwide. The total number of lost years of healthy life in the Established
Market Economies is likely m fall slightly, while it will increase slightly
in the Formerly Socialist Economies of Europe. Strikingly, however, Sub
Saharan Africa's future looks disturbingly poor despite the decline in the
burden of Group I conditions that currently dominate its health needs.
Overall, the region faces an increase in the number of lost years of healthy
life between 1990 and 2020, due mainly to a steep projected rise in the
burden of injuries from road accidents, war and violence.
SUMMARY 41
40 The Global Burden of Disease

ANNEX TAB LES TO The Global Burden of Disease


FROM THE SERIES
1 States or territories included in the Global Burden of Disease
TALBE OF CONTENTS: The Global Burden of Disease
Study, by demographic region 398
LIST OF TABLES iii 2 Cause Groups, diseases and injuries, and sequelae included in
the Global Burden of Disease Study 400
LIST OF FIGURES .............. ............. .........••.......•......................... ...... ... . V
3 Age-specific disability weights for the untreated and treated
GLOBAL BURDEN OF DISEASE AND INJURY SERIES FOREWORD forms of sequelae included in the Global Burden of Disease
Ralph H . Henderson xiii Study 412
4 Proportions of cases receiving treatment for each sequela , by
GLOBAL BURDEN OF DISEASE AND INJURY SERIES FOREWORD
region, 1990 417
Dean T. [amison xv
5 Percentages of deaths in each cause Group for varying levels
FOREWORD Wi//iam H. Foege xxv of mortality; regression results from 67 countries 419
6 Deaths by age, sex and cause (thousands), 1990 433
PREFACE xxvii 7 YLLs by age, sex and cause (thousands), 1990 469
ACKNOWLEDGEMENTS xxxi 8 YLDs by age, sex and cause (thousands), 1990 505
9 DAL Ys by age , sex and cause (thousands) , 1990 541
CHAPTER 1. RETHINKING DALYs 10 DAL Ys[O,Oj by age, sex and cause (thousands), 1990 577
Cbristopher j.L, Murray 1
11 Population by age, sex and region (t ho usan ds ), 1990-2020
CHAPTER 2. NON-FATAL HEALTH OUTCOMES: CONCEPTS, .............................................................. ....... ................ ........ 613
INSTRUMENTS AND INDICATORS 12 Deaths by age , sex and cause (thousands), 2000 , baseline
Ann Goerdt, jeff P. Koplan, [ean-Marie Robine, Michel C. scenario 616
Thuriaux, jerome K. van Ginneken 99 13 DAL Ys by age, sex and cause (thousands) , 2000, baseline
scenario 652
CHAPTER 3. ESTIMATING CAUSES OF DEATH: NEW METHODS AND 14 Deaths by age , sex and cause (thousands) , 2010, baseline
GLOBAL AND REGIONAL APPLICATIONS FOR 1990
scenario 688
Christopher j.L. Murray, Alan D. Lopez 118 15 DAL Ys by age , sex and cause (thousands) , 2010 , baseline
CHAPTER 4. GLOBAL AND REGIONAL DESCRIPTIVE EPIDEMIOLOGY scenario 724
OF DISABILITY: INCIDENCE, PREVALENCE, HEALTH EXPECTANCIES 16 Deaths by age, sex and cause (thousands) , 2020, baseline
AND YEARS LIVED WITH DISABILITY scenario 760
Cbristopber j.L. Murray, Alan D. Lopez 201 17 DAL Ys by age, sex and cause (thousands) , 2020, baseline
scenario 796
CHAPTER 5. THE GLOBAL BURDEN OF DISEASE IN 1990: FINAL 18 Deaths by age, sex and cause (thousands), 2020, pessimistic
RESULTS AND THEIR SENSITIVITY TO ALTERNATIVE
scenario 832
EPIDEMIOLOGICAL PERSPECTIVES, DISCOUNT RATES , AGE- 19 DAL Ys by age, sex and cause (thousands), 2020,
WEIGHTS AND DISABILITY WEIGHTS
pessimistic scenario 868
Cbristopher j.L. Murray, Alan D. Lopez 24 7 20 Deaths by age, sex and cause (thousands), 2020, optimistic
CHAPTER 6. QUANTIFYING THE BURDEN OF DISEASE AND INJURY scenario 904
ATTRIBUTABLE TO TEN MAJOR RISK FACTORS 21 DAL Ys by age , sex and cause (thousands) , 2020, optimistic
Christopher j.L. Murray, Alan D. Lopez 295 scenario 940
CHAPTER 7. ALTERNATIVE VISIONS OF THE FUTURE: PROJECTING
MORTALITY AND DISABILITY, 1990-2020
Christopher j.L. Murray , Alan D. Lopez 325
ANNEX TABLES 397
INDEX 97
The Global Burden ofDisease SUMMARY 43

GLOBAL HEALTH STATISTICS provides these detailed


THE GLOBAL BURDEN OF DISEASE AND INJURY
epidemiological estimates for 240 conditions. The following are
examples of these tables: SERIES

C".dro 1Se
I. THE GLOBAL BURDEN OF VI. N E U R 0 - P S YCHI AT R I C
T.bl. '61
Dlebetn MellUue Ol.betea m.n it".. DISEASE: a comprehensive assess- DISORDERS AND GLOBAL
c.... cesce Ca. menr of morrality and disability HEALTH: The epidemiology and
Diab&l89melll'us - Cases
from diseases, injuries and risk burdens of schizophrenia, dem-
Aoe Inddeneo 1990 P'ovalOnte 1990 AVQ. ag& Average Dealhs 1990 [)eelh, 2000 (PJOje(:Iec1)
Humber Rate
factors in 1990 and projected to entia, substance use, epilepsy and
group Number RaIO NlJmber Rale at Clnse1 o...w&ltOt\ Numbef Rale
1"..,./ (COOs) 'pe, tOOOOO} I"""'} 1_ .00000/ I,.... ' ,,..... J I"""') (per '000(0) ('000» (per 100 000) 2020. affective, neurotic and stress-
M....
related disorders.
0.. 3 11,0 7 2T 25 63 6 o 0.1 o 0.1 11. GLOBAL HEALTH STATISTICS: a
5-'4
15 ·44

113
8.0
61 .6
$0
'900
84
, 032
10.0
30.0
51.5
395
o
3
0.0
1.4
o
2
0.0
1.3
compendium of incidence, VII. THE GLOBAL BURDEN OF
45-59 'De $ 16.9 4240 6411 52.3 20.6 7 11.0 7 9.2 prevalence and mortality INJURIES: mortality and disability
60. 56T 936 .4 ' 0 573 '7.463 10.7 ' .0 47 77.' 53 raa
A1l4lgtS ' 096 280.6 ,6nl 420S 592 .6 .8 57 14.6 63 15.2 estimates for over 200 from suicide , violence, war and
Femal. .
cond itions. unintentional injuries.
0.' 3 10.0 6 25 2.' 696 o 0 .1 o 0,1
5-,' 6 '1.0 53 '04 10.0 63.9 o 0. 1 o 0,1
Ill. HEALTH DIMENSIONS OF SEX VIIl. MALNUTRITION AND THE
, 0.7
15,(.4
'09 60.8 , 696
4332
1059 29.' 4$ .2
24 .9
2
5
0 ,9
7.5 5 5.7 AND REPRODUCTION: the global BURDEN OF DISEASE: the global
45·59 404 595.9 6389 52.3
60- 092 .16.0 14790 17490 72 .4 '02 81 P6.1 96 98.3
AlII.1 1213 297 .7 21079 5175 61 .S IS .6 88 21.6 102 23.7 burden of sexually transmitted epidemiology of protein-energy
,( 744 60.<4 145 18.2 164 19.6
Totol 2308 289.3 37 850 '7,8
diseases, HIV, maternal condi- malnutrition, anaerruas and
In,n..• 'ndla · fnde DiabetesmelliluS· Cases tions, perinatal disorders and vitamin deficiencies.
Incidence 1990 PfeY8'o~ 1990 Avg. age AvctagO OealhS 1990 [)Nlh, 2000 (Protecredl co n-genital anomalies.
Number AlIIl. Numbef Ra'e al onsel durallon Nurrbof Ra,. ibrIbor Aale IX. QUANTIFYING GLOBAL HEALTH
('000>/ (p« 100 ooo) {'OOO:JJ (po F '00 O()()) (ye. ,,) (yeDls) I"""') _ ' 00 (00) ('000» (per . "" 0(0)
IV. THEGLOBAL EPIDEMIOLOGY OF RISKS: rhe burden of disease
......
0"" 4 7.0 3 5.' 2.5 0.97 2 3.5 2 3.8
INFECTIOUS DISEASES attributable to selected risk
5 ·'4 7 7.0 7 7.0 10.0 5.21 .. 4.3 3 2.8 factors .
15-44 96 47.7 , 364 . . ..4 29.7 :32.98 5 2.3 .. 1.8 V. GLOBAL PERSPECTIVES ON NON-
45-59 345 7252 3068 6 .....9.9 52 ,2 16.89 .0 20.9 10 16.4
60. 283 fM9.2 4795 16 ,og.5 69.1 7.35 28 94.5 29 17.6 COMMUNICABLE DISEASES: the X. HEALTH PRIORITIES AND
167 ,1 9238 2 102.• 55.1 15.11 49 11.2 48 9 .4
AI_ 734
epidemiology and burdens of BURDEN OF DISEASE ANALYSIS:
F.......
5.6 2.5 0.97 2 3.6 .2 4.0
cancers, cardiovascular diseases, methods and applications from
0-' 4 1.0 3
5·14 7 1.0 7 1.0 10.0 5.26 4 <11 .4 3 2.8 dia betes mellitus, respiratory global, national and sub-national
'5-4' 81 -47.7 \ 236 674 .7 29.7 34.08 4 22 3 \ ."
45·$9 329 714.1 2956 6425.8 52_2 \7.93 11 23 ,3 10 17.4 disorders and other major studies.
60. 16058 .5 69 ,5 7.49 34 1162 39 10""
272 941 .3 4645
conditions.
""ago. 699
, 433
170."
168.7
8647 2151.2
2'28-9
55..
SS.3
15.66
1$.38
$5
104
13.3
'2.2
58
106
12.0
10.6
Totol '8ce5

$SA • ASS· ASS Oiabcles rnafhtus• CaNS


WORKSHOP SOFTWARE PROGAMS

_.
Age lnadence 1990 Prevalence 1990 Avg. age Avetttge OealhS 1990 Dealh$~ (Pf~

9""'. Number Rate Nurnbel Rail. at 0tlSe1 durallon N~' ~l. Number A.il~
1f'6" 00 0(0) The Burden of Disease Unit holds Software Programs for Windows.
~ars) ('000>/ (P« 100 000) (COOs) 1,," ' 00 000) tr-s, I,....' (CIOOsJ (pet,OO OOO) ('000»
workshops periodically throughout the Developed by the Burden of Disea se
, o year. Workshops provide detailed Unir to measure ep idemiological
0-4
5·'4 1
2.0
2.0 ,
1 1.1
1.3
2."
10.0
0.61
2.95 1
0.8
2.0
1
1
0.8
lA
instruction and practical experience in
'5-44 n 20.' 300 289. 29-4 29.501 1 0.' , 0.7 indicators of diseases and injuries.
2 2 8.0
45-59
60.
rt
54
3n.7
514'"
646
87.
3179.1
8320.9
52.1
69.0
15.65
7.10 5
10."
49 .2' 6 39.3
measuring burden of disease and These programs can be purch ased
AI_ 1SS 6'-' 4 , 0. . 722.0 54.' , ..... , 10 4.0 " 3.' undertaking analyses of the cost- by calling 617 495 8498 or by email,
effectiveness of health interventions. egakidou@hsph.harvard.edu.
Fem.1ft
, I.,
0-'
5-11 , 2.0
20 ,
1
'.3
2.4
10.0
0,6'
2.99
o
, 1.0
2.0
r
I
'-0
r." They emphasize both conceptual and
15-4' 22 20.9 301 218 .7 29 .$ 30.66 o 0.' o 0.3
applied skills and will include the use of
45 ·59 82 312 .9 100 3162 ,9 52_2 16.40 3 11.8 2 .0."
60. 73
'80
572 .6 • 099 6635.• 69.0& 7.25
• 65 .7
5.2
9
14
52.'
".0
specialized compurer software. Please
""-'
701., 335
698
65 .6
2107
3929
816 .8
769. 55.8
56.0
' 4.27
'4 .33
'3
23 '6 24 3.5 contact the Burden of Disease Unit for
For epidemiological8OUtt8S see McKeigue and King 1996. For lhe melhod9 u~ to G$tima1e and projec1
more information.
incidence. pnwaJanco,anc1 dealhs see Murrav and lopez 1996l:I. Soo 8XP'an&1OfY notes lor da4nhOtli$ ;lnd cav&DIS.

For more information: Tel: 6 17·495-8498; Fax: 617-4 96-3227; Email: eg., kido u@hsph. harv. rd.ed u;
Weh: hnp :llw\V\V.h sph .h.rv~rd .cduf()Tg~ni z.,ri ()n 5/bdu
ORDER FORM

o PLEASE SEND o CHECK OR MONEY ORDER


ENCLOSED
_copies ofGlobal Burden of Disease payable to:
($39.95 ea.) Sept. 6x9, 900pp. Harvard University Press
ISBN 0-674-35448-6 (MURGLO)
o PLEASE CHARGE TO MY
CREDIT CARD
_copies of Global Health Statistics
o Visa
($39.95 ea.) Sept. 6x9, 1010 pp.
ISBN 0-674-35449-4 (MURGLH)
o American Express
o Mastercard
Name _ Card Number " _

Address _ Exp . DaW_


'- _

signature

Total Payment plus


Send to: $4.50 postage
Customer Service
Harvard University
79 Garden Street
UK/EuropelAfricalIndialMiddle East
Cambridge, MA 02138
contact: Harvard University Press,
TeI: 1-800-448-2242 do Wiley & Sons Ltd. Southern Cross
Fax: 1-800 -962-4983 Trading Estate, 1 OIdlands Way,
Int'l Tel: 617-495-2480 Bognor Regis, W. Sussex PP22 9SA,
Int'l Fax: 617 495-8924 United Kingdom

TeI: 00-44 -1243-779777


Developing Country Price
Fax: 00-44-1243-820250
For residents of developing countries,
Telex: 86111 WILEY
copies of The Global Burden of
email: customer@wiley.co.uk
Disease and Global Health Statistics
are available at a price of US $23.00
Web order form: http://www.hup.harvard .cdu
each + postage (depending on
location ) by contacting:
Ann: E. Gakidou 36-200
Burden of Disease Unit
9 Bow Street
Cambridge, MA 02138 USA (f) WHO LIBRARY
Tel: 1-671-495-8498
Fax: 1-617-496-3227
Email: egakidou@hsph.harvard.edu II~I~1~~~~lmlmm~mll~lm
·00050215·
__
IIIWlmIOO~~IW~i~I~W
"000502t5* _

You might also like