Aging & Mental Health, March 2006; 10(2): 7986

ORIGINAL ARTICLE

The relationship between caregiving self-efficacy and depressive

symptoms in dementia family caregivers

C. M. GILLIAM & A. M. STEFFEN

Department of Psychology, University of Missouri, St. Louis, Missouri, USA

(Received 12 November 2004; accepted 4 April 2005)

Abstract
The present study was intended to replicate the findings of Steffen et al. (2002) of a negative relationship between
caregiving self-efficacy and depressive symptoms among family dementia caregivers. Female family caregivers (N 74)
of community-dwelling individuals diagnosed with dementia completed a telephone interview and self-report assessment
packet that included measures of caregiving self-efficacy and depressive symptoms. There was a direct negative relationship
between caregiving self-efficacy and depressive symptoms after controlling for objective stressors. There was no support,
however, for the hypothesis that caregiving self-efficacy would operate as a moderator, such that the relationship between
objective stressors (cognitive impairment and behavior problems) and caregivers depressive symptoms would be strongest
for caregivers reporting lower levels of self-efficacy. The results of this study suggest that caregiving self-efficacy has a strong,
direct relationship with depressed symptoms for dementia family caregivers. Longitudinal research is needed to determine
if it is an appropriate focus of future intervention research.

Introduction
Caregivers of family members diagnosed with
dementia are engaged in a challenging and important
role that often consumes their lives. The demands
and emotional strains associated with dementia
caregiving leave caregivers vulnerable to psychological and health consequences, most frequently,
depression. The relationship between depressive
symptoms and caregiving has been found in virtually
all studies of dementia caregivers (Schulz et al.,
1995). Not all, however, experience similar negative
effects of caregiving. Some individuals experience
significant distress; others are able to manage
the process of caregiving without experiencing
psychosocial impairment (Aneshensel et al., 1995).
In order to develop effective interventions, it is
important to identify and understand the factors
that protect some caregivers from negative consequences frequently associated with caregiving.
Although objective stressors have been proposed as
leading to depressive symptoms among caregivers,
the results are ambiguous. Some studies find
a direct relationship between depressive symptoms
and the objective stressors of caregiving, such as
cognitive impairment, behavioral problems, and
activities of daily living (ADL) deficiencies of the
patient (Alspaugh et al., 1999; Donaldson et al.,
1998; Nagaratnam et al., 1998; Teri, 1997).

Others have not, suggesting that other variables

besides objective stressors are responsible for the
heterogeneity in depression among family dementia
caregivers (Boss et al., 1990; Gaugler et al., 2000;
Li et al., 1999; Schulz et al., 1995). These conflicting results imply that primary objective stressors,
the actual demands of caregiving, may not be
directly related to depressive symptoms. Instead,
other factors such as a caregivers perception of the
caregiving situation or his/her ability to manage
caregiving-related demands may be more strongly
related.
One such variable proposed as playing an important role in the relationship between caregiving
and depressive symptoms is self-efficacy (Bandura,
1977; Fortinsky et al., 2002; Gignac & Gottlieb,
1996; Zeiss et al., 1999). Self-efficacy (also called
perceived self-efficacy) is an individuals assessment
of his or her ability to successfully master a specific
task (Bandura, 1997). These efficacy beliefs are
an individuals estimates of her or his own ability
to mobilize the motivation, cognitive resources, and
courses of action needed to meet given situational
demands (Bandura & Wood, 1989, p. 408). Selfefficacy is by definition concerned with the individuals assessment of ability to perform a specific
task, and differs from a more general assessment

Correspondence: Christian M. Gilliam, Psychology Department, 325 Stadler Hall, One University Blvd., St. Louis,
Missouri 63121, USA. Tel: 1 (314) 516 5391. Fax: 1 (314) 516 5392. E-mail: cmgg3c@umsl.edu
ISSN 1360-7863 print/ISSN 1364-6915 online 2006 Taylor & Francis
DOI: 10.1080/13607860500310658

80

C. M. Gilliam & A. M. Steffen

of abilities, such as self-mastery or a global evaluation of the self encompassed in the concept of selfesteem. One can have high self-efficacy in a certain
domain but not another. Unlike self-esteem and
self-mastery, which are conceived as relatively stable
constructs, self-efficacy often changes within the
same individual over time and in response to specific
experiences (Bandura, 1997).
Those with a high sense of caregiving efficacy may
be protected from the negative consequences
of this role by focusing on what they are capable
of accomplishing, rather than on their failures.
Rather than viewing the demands of caregiving
as tasks to be avoided, those with a high sense
of caregiving self-efficacy may view them as challenges to be mastered. In short, a resilient sense
of self-efficacy enables people to endure hardships
and persevere against great odds (Bandura, 1997,
p. 22) that are associated with caregiving.
Individuals with a low sense of caregiving selfefficacy, on the other hand, are more vulnerable to
the stressors of caregiving and the potential negative
consequences, especially depression. In the face
of challenges, which undoubtedly occur frequently
in dementia caregiving, these individuals are more
likely to focus on past failures and doubt their ability
to effectively respond to the demands of caregiving.
Of the studies examining the relationship between
self-efficacy and depression in non-caregiving
samples, some suggest a direct causal relationship
between self-efficacy and depressive symptoms,
whereas others suggest a buffering, or moderating
effect of self-efficacy on depression. Stanley and
Maddux (1986), for example, examined the
direct causal relationship between self-efficacy for
interpersonal skills and depressed mood. Using
an experimental manipulation of participants
self-efficacy regarding interpersonal skills, the
investigators found that those with an induced low
self-efficacy expectancies reported greater depressed
mood than those with induced high self-efficacy
expectancies. Induction of depressed or elated
mood, however, had no corresponding effect on
the participants self-efficacy regarding social interactions, suggesting a one-way causal relationship
between a specific type of self-efficacy and depressed
mood. In a longitudinal study of cardiac rehabilitation patients, early-treatment changes in exercise
self-efficacy predicted late-treatment improvements
in activity level, depressive symptoms, and patientstaff working alliance. These positive outcomes,
however, did not predict changes in exercise selfefficacy. Results from this longitudinal intervention
study further suggest a one-way causal relationship
between certain types of self-efficacy and depressive
symptoms (Evon & Burns, 2004).
Among older adults, self-efficacy in specific
areas has also been found to play a moderating
role between key domains of an older adults life
(e.g., physical health, financial situation, relationship

with adult children, and social support from friends)

and depressive symptoms (Chou & Chi, 2001).
Moderators help clarify why people react differently
to similar situations. In the case of caregivers,
for example, a moderator can identify a cognitive
characteristic that may influence the relationship
between stressors and depression. Two dementia
caregivers whose impaired relatives have the same
level of cognitive impairment and behavioral problems, for example, may not experience the same
amount of depressive symptoms. The caregiver
with high self-efficacy and high stressors may only
experience a low level of depressive symptoms while
the caregiver with low self-efficacy and high stressors
may experience a higher level of depressive
symptoms.
Self-efficacy has been applied to explain the
diverse reactions of individuals to stressful life
experiences, and has more recently been applied
to the stressor of caregiving for a family member with
dementia. To date, the few studies in this area have
found a negative relationship between self-efficacy
and depression among dementia caregivers (Gignac
& Gottlieb, 1996; Steffen et al., 2002; Zeiss et al.,
1999). Although these investigators have examined
different types of self-efficacies, such as coping
self-efficacy, caregiving self-efficacy, and problemsolving self-efficacy, they have all found that those
with higher self-efficacy were less likely to report
depressive symptoms.
The aim of this study was to clarify the role
of caregiving self-efficacy in the relationship between
objective primary stressors and depressive symptoms. Specifically, we examined whether self-efficacy
for responding to problematic patient behaviors
moderates the relationship between the objective
stressors of caregiving (cognitive impairment and
behavioral problems) and caregiver depressive
symptoms.

Method
Participants
The data for this paper were collected as part of
an intervention study for female dementia caregivers
with depressive symptoms. A variety of strategies
were used to recruit participants from nine central
US states (Illinois, Indiana, Iowa, Kansas, Michigan,
Minnesota, Missouri, Nebraska, and Wisconsin),
including calling agency staff at all chapters of the
Alzheimers Association and Area Agency on Aging
(AAA), and providing brochures and articles about
the program. Caregivers who called in for more
information about the program received a telephone
screening interview to determine interest and eligibility. Criteria for inclusion in the study included
being female aged 3080, and a primary caregiver
who lives with a family member with a physician
confirmed diagnosis of dementia. In order to be

81

Caregiving self-efficacy and depressive symptoms

included in the study, the caregiver needed to report
depressive symptoms as indicated by a score of three
or higher (out of a possible score of 10) on the
Boston Short Form of the Center for Epidemiologic
Studies of Depression Scale (CES-D) (Kohut et al.,
1993), and at least moderate distress following
at least two weekly-occurring behavioral problems
of the patient, as measured by the Revised Memory
and Behavior Problems Checklist (Teri et al., 1992).
Caregivers were excluded from the study if they
exhibited disabling hearing impairment as measured
by the Hearing Handicap Inventory for the Elderly
(Ventry & Winstein, 1982), disabling visual impairment as measured by the Low Vision Quality of Life
Scale (Wolffsohn & Cochrane, 2000), or significant
cognitive impairment, as measured by the telephone version of the Short Portable Mental Status
Questionnaire (Roccaforte et al., 1994). Other
exclusionary criteria included significant suicidal
ideation, current alcohol or drug abuse, or not
having access to a telephone or VCR.
As shown in Table I, the majority of caregivers
were Caucasian (79.7%), with an average age of
59 years. Most caregivers were married (71%), with
approximately half of caregivers (52.2%) caring for
their husband, and the other half caring for a parent
(38% mother; 6% father). Less than 5% of caregivers were caring for other family members (grandparent, aunt or mother-in-law). Less than half
(36.5%) of the caregivers worked either full or
part-time outside of the home, and approximately
half reported an income of at least $40,000 (54.1%).
Only a small number of caregivers reported having
significant financial difficulty (9.5%). The caregivers
were well educated, with the majority reporting

Table I. Demographic characteristics of participants (N 74).

Caregiver characteristics
Age
Education
Some high school
High school diploma
Post high school training
Ethnic background
Caucasian
African-American
Other
Marital status
Married
Not married/widowed
Employment status
Full-time
Part-time
Not employed
Income
Less than $20,000
Less than $40,000
Above $40,000
Length of caregiving

Means/percentages

completion of some type of post-high school

education.
The average participant reported being a caregiver
for approximately three years. Although an attempt
was made to assess the amount of time the caregivers
spent providing direct assistance to their relative,
caregivers often reported spending 24 hours per day
caregiving; as such, this variable was not included
as it did not appear to accurately assess the actual
amount of time caregivers spent providing assistance or supervision to their relatives on a daily
basis. Approximately 50% of the caregivers reported
receiving one type of formal services to assist caring
for their family member, such as a homemaker, a
home health aid, a visiting nurse, delivery of cooked
meals, provision of transportation, or having their
family member attend a day care. Approximately
24% of the caregivers reported that they received
none of the formal services described above, and
approximately 26% of the caregivers reported
receiving two or more of the services described
above. The most frequently reported formal support
was senior day care attendance by the family
member (38%).
As shown in Table II, the average dementia
patient was a Caucasian (78%) male (58%) in his
mid seventies. The majority of the patients were
married, with the more than half having at least
a high school diploma (66.5%). The average patient
needed assistance in two areas among the activities
of daily living, such as bathing or toileting, and the
majority of the patients were described by their
caregivers as being in good physical health.
Measures
Caregiving self-efficacy. Self-efficacy was measured
using the Revised Scale for Caregiving Self-Efficacy
(Steffen et al., 2002). The scale consists of 15 items;

Range

58.9 years
(10.5)

74 (33.185.2
years)

5.4%
20.3%
88.2%

4
15
55

79.7%
18.9%
1.4%

59
14
1

67.6%
32.4%

50
24

24.3%
12.2%
63.5%

18
9
47

16.7%
29.2%
54.1%
2.94 years

12
21
39

Table II.

Patient characteristics (N 74).

Caregiver characteristics
Age

0.069.96
years

Gender
Male
Female
Relation to caregiver
Mother
Father
Husband
Other relative
Type of dementing illness
Alzheimers disease
Undecided dementia
Vascular dementia
Mixed dementia
Dementia secondary to
Parkinsons disease
Other type of dementia
ADL impairment

Means/percentages

Range

75.4 years (9.1)

74

(54.892.5
years)

58.2%
41.8%

39
28

37.3%
6%
52.2%
4.5%

25
4
35
3

60.8%
21.6%
8.1%
5.4%
1.4%

45
16
6
4
1

2.7%
2.01 (1.75)

2
74

06

82

C. M. Gilliam & A. M. Steffen

caregivers were asked to rate their level of confidence

(from 0100%) that they could perform each activity
if they gave their best effort. For example, caregivers
were asked to rate their confidence as 0% if they
believe they could not perform an activity at all, 20%
if they believed it was unlikely but not impossible
to perform the activity, 5060% if they were
moderately certain they could perform the activity,
and 100% if they were certain they could perform
the activity with their best effort. The measure
begins with a detailed explanation and opportunity
to practice reporting confidence levels with noncaregiving related tasks. Each item on the scale asks
the caregiver to assess their confidence level for
being able to perform a behaviorally specific task.
The scale is divided into three subscales assessing
three domains of caregiving self-efficacy. (1) Selfcare and obtaining respite (5 items): this scale
measures caregivers self-efficacy regarding the
ability to ask for assistance (e.g., How confident
are you that you can ask a friend/family member to
stay with your family member for a day when you
need to see the doctor?). (2) Responding to disruptive patient behaviors (8 items): this scale assesses
caregivers self-efficacy regarding the ability to
respond to the patients disruptive behaviors effectively. (e.g., When your family member asks you
four times in the first hour after lunch when lunch is,
how confident are you that you can answer him/her
without raising your voice?). (3) Controlling upsetting thoughts activated by caregiving activities: this
scale measures caregivers self-efficacy regarding
controlling negative or upsetting thoughts concerned
with caregiving (5 items). (e.g., How confident are
you that you can control worrying about future
problems that may come up with your family
member?). Due to our specific interest in caregivers
responses to difficult behaviors of the patient, only
the scale for responding to disruptive patient behaviors (5 items) was examined in relation to depressive
symptoms for the present study. This scale has
demonstrated good reliability and validity and high
internal consistency in past studies (Cronbachs
alpha 0.82) with two-week test retest reliability
moderate and in the acceptable range (r 0.70;
Steffen et al., 2002). In the present study, the
scale was normally distributed with high internal
consistency (Cronbachs alpha 0.88).
Depressive symptoms. Depressive symptoms were
measured by the Beck Depression InventorySecond Edition (BDI-II) Long Form (Beck, Steer,
& Brown, 1996). The BDI-II consists of 21 items
designed to measure the degree of depressive symptoms in adults and adolescents age 13 and older.
Participants were asked to pick the statement that
best described the way that they had been feeling in
the last two weeks, including the day of the interview
(Beck, Steer, & Brown, 1996). The BDI-II has good
reliability and validity. It has strong internal

consistency, with a coefficient alpha of 0.92, and

good one week test-retest reliability (r 0.93,
p < 0.001). In the present study, the items were
normally distributed and demonstrated high internal
consistency (Cronbachs alpha 0.89).
Behavior problems. The Revised Memory and
Behavior Problems Checklist (RMBPC; Roth
et al., 2003; Teri et al., 1992) was used to measure
caregivers report of the dementia patients observable behavior problems, such as memory problems,
depressive symptoms, and disruptive behaviors.
The checklist consists of 25 items, which represent
possible problems that the patient may be exhibiting.
The caregiver was asked whether a particular
behavior or memory problem had occurred in the
past week (Roth et al., 2002), with a score created
for the total number of behavior problems endorsed
by the caregiver. The checklist has good reliability
and convergent and discriminant validity. For example, endorsement of memory related problems
correlated negatively with the Mini-Mental Status
Exam and positively with the diagnosis of dementia
of the family member (r
0.48/0.45, p < 0.001)
(Teri et al., 1992). In the present study, the total
score was normally distributed and had adequate
internal consistency (Cronbachs alpha 0.74).
Telephone measure of cognitive impairment. This scale
measures the degree of cognitive impairment of the
family member, such as memory loss, communication deficits, and recognition failures, through
reports by the caregiver. Caregivers were asked to
rate how difficult it is for the relative to perform
each of the seven items, on a scale from 0 (not at
all difficult) to 5 (cant do at all) (Aneshensel et al.,
1995). The scale has been found to adequately
evaluate the level of cognitive impairment for
brief screening purposes. The scale correlates
highly (r 0.65) with the Mini-Mental State Exam
(Folstein et al., 1975). The scale demonstrated
adequate internal consistency in the present study
(Cronbachs alpha 0.82).
Descriptive information. A brief demographic
questionnaire assessed the caregiver and the family
members ethnic background, educational level,
income and financial status, marital status, age,
and general health. The Index of Activities of
Daily Living (Katz et al., 1963) was also utilized to
obtain descriptive information regarding the family
members ability perform activities of daily living
with or without assistance, such as bathing, toileting,
and eating.
Procedure
Once a caregivers eligibility and interest in participating in the study were determined, a baseline
telephone interview was scheduled; consent forms

Caregiving self-efficacy and depressive symptoms

and response cards to assist with the telephone
interview process were mailed to the caregiver.
At the beginning of the baseline telephone call,
informed consent was obtained. Following the
informed consent procedures, the measures used
in the present study were administered over the
telephone in the following order: The Index of
Activities of Daily Living (Katz et al., 1963), The
Revised Memory and Behavior Problems Checklist
(Teri et al., 1992), The Telephone Measure of
Cognitive Impairment (Aneshensel et al., 1995),
The Beck Depression Inventory-Second Edition
(Beck, Steer, & Brown, 1996), and The SelfEfficacy Scale for Responding to Disruptive Patient
Behaviors (Steffen et al., 2002). These measures
were completed over the phone rather than mailed
to the caregiver to increase the likelihood of the
caregiver responding and completing the measures
accurately. Demographic data were collected in
a questionnaire format that participants mailed
back to the research office.

confidence in responding to various behavior

problems on the RMBPC.
The average dementia patient in the study showed
approximately 1112 different types of behavioral
problems within a week of the interview, and had
quite a bit of difficulty performing various cognitive
tasks. Pearson correlations were conducted to
examine any problems with multicollinearity; as
illustrated in Table III, there were no concerns
with multicollinearity in the data.
To test the studys hypotheses, two hierarchical
regression analyses with scores on the BDI-II as the
dependent variable were conducted. Scores on the
Telephone Measure of Cognitive Impairment and
the Revised Memory and Behavior Problems
Checklist were entered into block 1. In block 2,
self-efficacy for responding to disruptive behaviors
was entered. The product of self-efficacy and
behavior problems (Regression #1) or the product
of self-efficacy and cognitive impairment (Regression
#2) was entered into block three.
Contrary to our hypothesis, self-efficacy for
responding to problematic patient behaviors did
not act as a moderator between patient cognitive
impairment or behavioral problems and caregiver
depressive symptoms. As illustrated in Table IV,
neither interaction significantly improved the ability
of the model to predict scores on the BDI-II.
The interaction between self-efficacy and behavior
problems or the interaction between self-efficacy
and cognitive impairment accounted for less than
1% of the variance in scores on the BDI-II. Further,
the relationship between objective stressors and
depressive symptoms was not significant, accounting
for less than 2% of the variance in scores on the
BDI-II. Neither behavior problems (beta 0.54,
p 0.65) nor cognitive impairment (beta 0.12,
p 0.34) significantly predicted depressive symptoms of the caregivers. In comparison, self-efficacy

Results
Table III shows variable distributions and intercorrelation among study variables. On average, caregivers reported mild levels of depressive symptoms
on the BDI-II (M 15.8). The average confidence
level for responding to various disruptive behaviors
among caregivers was 67%. The level of depressive
symptoms reported among this studys participants
is comparable to those of other intervention studies
assessing depressive symptoms among caregivers
of dementia patients. Two intervention studies
assessing depressive symptoms with the BDI-II
among dementia family caregivers reported an
average BDI-II score of 13 (Coon et al., 2003) and
19.3 (Gallagher-Thompson & Steffen, 1994).
Participants also displayed a moderate level of

Table III.

Pearson correlation of cognitive impairment and behavior problems (N 74).

Variable
1.
2.
3.
4.
5.

Mean (SD)

Beck Depression Inventory

Cognitive Impairment
Behavior Problems
Self-Efficacy
ADLs

15.76
2.58
11.55
67.23
2.01

(3.33)
(1.07)
(3.33)
(19.82)
(1.75)

Range

033
0.434.71
420
1598.75
06

0.11
0.05

0.40*
0.05

0.02
0.11
0.44*

0.13

0.04

0.00

*Correlations significant at the 0.01 level.

Table IV.

Hierarchical multiple regressions predicting scores on BDI-II (N 72).

Block
(1)
(2)
(3)
(3)

83

Behavior problems and cognitive impairment

Caregiving self-efficacy
Product of behavior problems and self-efficacy
Product of cognitive impairment and self-efficacy

R2

R2 Change

Significance of change

0.016
0.196
0.198
0.199

0.016
0.180
0.002
0.003

0.571
0.000
0.710
0.597

84

C. M. Gilliam & A. M. Steffen

Table V. Hierarchical multiple regressions predicting scores on BDI-II with lethargy removed
(N 72).
Block
(1)
(2)
(3)
(3)

Behavior problems and cognitive impairment

Caregiving self-efficacy
Product of behavior problems and self-efficacy
Product of cognitive impairment and self-efficacy

accounted for approximately 18% of the variance

after controlling for objective stressors, and significantly predicted scores on the BDI-II. As
hypothesized, self-efficacy had a significant negative relationship with depressive symptoms
(beta
0.43, p < 0.001). The Pearson r for a
direct correlation between self-efficacy and depression (r
0.40) and the partial correlation after
controlling for behavioral problems are nearly
identical (r
0.41), providing further evidence
that the relationship between self-efficacy and
depression was not affected by patient behavioral
problems.
As discussed in the Measurement section of this
paper, the items used to measure caregiving selfefficacy in the present study are behaviorally specific
descriptions of common disruptive behaviors by
dementia patients, and are different in content
to the BDI-II items. Despite the apparent difference
in item content between the BDI-II and the selfefficacy items, the significant correlation between
the BDI-II and caregiving self-efficacy suggests that
there may be a potential confound between caregivers self-efficacy for responding to disruptive
behaviors and depressive symptoms. In an attempt
to tease apart a potential confound between selfefficacy and depressive symptoms in these caregivers, hierarchical regression analyses to test the
main hypotheses were also conducted with a revised
dependent variable that excluded items assessing
somatic or lethargic symptoms on the BDI-II (item
15: loss of energy; item 16: changes in sleeping
patterns; item 20: fatigue/tiredness). As shown in
Table V, removal of these three BDI-II items did
not result in any significant changes to the analyses.
No moderating relationship was found between the
recalculated BDI-II scores and self-efficacy after
controlling for objective stressors.

Discussion
The present study hypothesized that caregiving selfefficacy functions as a moderator between objective
stressors and caregivers depressive symptoms. The
relationship between self-efficacy and depressive
symptoms appears to be a direct relationship,
and not a moderating relationship as hypothesized.
Whereas objective stressors did not demonstrate
a significant impact on depressive symptoms,

R2

R2 Change

Significance of change

0.015
0.181
0.183
0.183

0.015
0.166
0.003
0.003

0.597
0.000
0.644
0.644

self-efficacy maintained a significant negative

relationship with depressive symptoms, even after
controlling for objective stressors.
The significant relationship between caregiving
self-efficacy and depressive symptoms found in this
study is consistent with both the literature on
self-efficacy and depression in general, as well as
with the literature on self-efficacy and depression
specific to dementia caregivers. Using a different,
non-intervention sample of dementia caregivers,
Palmer and Steffen (2004) also found a direct
negative correlation (r
0.38) between caregiving
self-efficacy and depressive symptoms. Furthermore,
in the sample, initially lower levels of caregiving
self-efficacy predicted length of time to nursing
home placement over seven years, while controlling
for initial levels of depressive symptoms.
In addition to providing further evidence supporting the relationship between caregiving self-efficacy
and depressive symptoms, the results of this study
provide further understanding of the relationship
between self-efficacy and caregiver depression.
First, by controlling for the objective stressors of
caregiving, the results demonstrate that self-efficacy
has a significant relationship with depressive symptoms even after accounting for differences in the
objective stressors of caregiving. In other words,
regardless of the level of cognitive impairment or
behavioral problems that the caregivers face,
self-efficacy maintains a significant relationship
with depressive symptoms. The results also suggest
that the caregivers self-efficacy is not simply determined by the objective challenges of caregiving.
The findings are consistent with Banduras (1977)
self-efficacy theory.
There are at least three published studies examining caregiving self-efficacy in relation to depressive
symptoms (Fortinsky, Kercher, & Burant, 2002;
Steffen et al., 2002; Zeiss et al., 1999). Considering
that self-efficacy is concerned with the caregivers
perceived ability to execute a particular domain
of tasks, and not a general sense of global mastery,
and that self-efficacy can differ in level, generality,
and strength, it is not surprising that there are a
variety of ways of defining and measuring caregiving
self-efficacy (Bandura, 1977). For example, Zeiss
and her colleagues (1999) examined caregivers
self-efficacy for self-care and problem solving.
Although they found a significant relationship
between depression and self-efficacy, Zeiss and

Caregiving self-efficacy and depressive symptoms

her colleagues reported that their measure of
self-efficacy was not a strong predictor of caregiver
distress. In addition, their measurement of problemsolving self-efficacy was generalized, and did not
focus on specific situations or caregiving tasks.
The self-efficacy scale used in this study is a revision
of the scale used by Zeiss et al. (1999) to improve
upon the ceiling effects and include a broader range
of items that assess the caregivers self-efficacy
in responding to various behavioral problems
(Steffen et al., 2002). The relationship found
between caregiving self-efficacy and depression
in this study was stronger than that found in the
study by Steffen et al. (2002) (r
0.31 and
r
0.34 in the study by Steffen et al., as compared
to r
0.43 in this study). Given that the scale used
to measure caregiving self-efficacy in the two studies
was identical, the stronger relationship between
self-efficacy for responding to disruptive patient
behaviors found in this study can be attributed to
differences in caregiver characteristics between the
two studies. In order to be included into the study,
the caregivers in this study were reporting distress
regarding patient behavioral problems and some
depressive symptoms, as well as an interest in
participating in a caregiver intervention program.
The association between caregiving self-efficacy and
depressive symptoms, therefore, may be stronger
when applied to caregivers who are at least mildly
depressed or distressed, and help-seeking.
A recent study by Fortinsky and his colleagues
(2002) examined the relationship between symptom
management self-efficacy and caregiver depression,
and found very similar results to those of this study.
Our research, however, unlike the study by Zeiss
et al., or by Fortinsky et al., which used a more
global measure of caregiving self-efficacy, specifically
examined self-efficacy in regards to the ability to
respond to common disruptive behaviors, such as
repeated questioning, or following the caregiver.
The measurement of self-efficacy used in this
study assessed caregiving self-efficacy in response
to specific and diverse situations frequently encountered by dementia caregivers. The intermediate level
of generality assessed in this measure yields a greater
predictive power while still measuring the domain
of the caregivers self-efficacy in responding to
disruptive behaviors (Bandura, 1997).
Although the results of this study contribute
further to the understanding of self-efficacy and
depressive symptoms among dementia family caregivers, there are some limitations. First, due to the
correlational design of this study, no conclusions can
be drawn regarding a causal relationship between
caregiving self-efficacy and depressive symptoms.
Present results suggest that caregiving self-efficacy
and depressive symptoms have a significant association; however, such an association may simply
reflect a potential confound between self-efficacy
and depressive symptoms. The current study cannot

85

answer the question of why caregiving self-efficacy

and depressive symptoms are related. Future
research, particularly longitudinal and intervention
research will help shed light on the relationship
between caregiving self-efficacy and depressive
symptoms.
Further, in addition to the general problem of
relying solely on self-report measures, the measurement of behavioral problems may have included
more memory or cognitive problems associated with
dementia rather than the behavioral problems found
to be most distressing to caregivers, such as violent
behavior, or constant following of the caregiver.
In addition, most of the caregivers in the study did
not report the more distressing behavioral problems
included in the measure. This may account for the
non-significant relationship found between behavioral problems and caregivers depressive symptoms.
An obvious limitation to the study is the participant characteristics. As in most caregiver studies,
the study participants were self-selected into the
study. The caregivers were all female, mostly
Caucasian, well educated, and mostly from a high
socioeconomic status. The results of this study,
therefore, many not generalize to caregivers from
lower socioeconomic status, with lower education,
caregivers uninterested or not capable of participating in an intervention study, for caregivers of ethnic
minority, or to male caregivers. In fact, there is
some research to suggest that the consequences of
long-term caregiving differ for African-American and
Caucasian caregivers, as well as for male and female
caregivers (Roth, Haley, Owen, & Clay et al, 2001;
Gallicchio, Siddiqi, Langenberg, & Baumgarten,
2002). Caregivers financial status also undoubtedly
influences the nature of caregiving; those with
financial resources can more easily acquire professional assistance, such as a home health aid, or
respite. Adequate financial resources, therefore, may
ease some of the stressors of caregiving.
Of particular interest within self-efficacy research
among dementia caregivers is the relative importance
of different types of caregiving self-efficacy.
Caregiving is a challenging role that demands that
caregivers be able to respond to a variety of different
situations and tasks. Caregivers must be able to
coordinate medical care, be responsible for the
finances, as well as assist the impaired relative
in daily activities of living and keeping the home
safe, among many others. The different types of
self-efficacy measured within caregiving research
demonstrate the diverse range of responsibilities
for caregivers (Steffen et al., 2002; Fortinsky,
Kercher, & Burant, 2002; Zeiss et al., 1999;
Gignac & Gottlieb, 1996). In fact, the measurement
of caregiving self-efficacy used for this study, the
caregivers self-efficacy in responding to disruptive
behaviors, is only one of the three types of caregiving self-efficacy measured in the Scale for
Caregiving Self-Efficacy (Steffen et al., 2002).

86

C. M. Gilliam & A. M. Steffen

Although the caregivers self-efficacy for responding

to disruptive behaviors was chosen for this study, the
other types of self-efficacy measured by the Scale
for Caregiving Self-Efficacy, self-efficacy for self-care
and obtaining respite, and for controlling upsetting
thoughts, may also prove to be informative in
understanding caregivers depressive symptoms.
Comparisons of different types of caregiving selfefficacy may shed further light on the relationship
between self-efficacy and caregiving.

References
Alspaugh, M. L., Stephens, M. P., Townsend, A. L., & Zarit,
S. H., et al. (1999). Longitudinal patterns of risk for depression
in dementia caregivers: Objective and subjective primary stress
as predictors. Psychology and Aging, 14, 3443.
Aneshensel, C. S., Pearlin, L. I., Mullan, J. T., Zarit, S. H., &
Whitlatch, C. J. (1995). Profiles in caregiving: The unexpected
career. San Diego: Academic Press, Inc.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory
of behavioural change. Psychological Review, 84, 191215.
Bandura, A. (1997). Self-efficacy: The exercise of control. New York:
W. H. Freeman and Company.
Bandura, A., & Wood, R. E. (1989). Effect of perceived
controllability and performance standards on self-regulation
of complex decision making. Journal of Personality and Social
Psychology, 56, 805814.
Beck, A. T., Steer, R. A., & Brown, G. K. (1996). The Beck
Depression Inventory Second Edition Manual. San Antonio:
Psychological Corporation.
Boss, P., Caron, W., Horbal, J., & Mortimer, J. (1990). Predictors
of depression in caregivers of dementia patients: Boundary
ambiguity and mastery. Family Process, 29, 245254.
Chou, K. L., & Chi, I. (2001). Social comparison in Chinese
older adults. Aging & Mental Health, 5, 242252.
Coon, D. W., Thompson, L., Steffen, A., Sorocco, K., &
Gallagher-Thompson, D. (2003). Anger and depression
management: Psycho-educational skills training interventions
for women caregivers of a relative with dementia. The
Gerontologist, 43, 678689.
Donaldson, C., Tarrier, N., & Burns, A. (1998). Determinants
of carer stress in Alzheimers disease. International Journal
of Geriatric Psychiatry, 13, 248256.
Evon, D., & Burns, J. (2004). Process and outcome in cardiac
rehabilitation: An examination of cross-lagged effects.
Journal of Consulting and Clinical Psychology, 72, 605616.
Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975).
Mini Mental State. A practical method for grading the
cognitive state of patients for the clinician. Journal of
Psychiatric Research, 12, 189198.
Fortinsky, R. H., Kercher, K., & Burant, C. J. (2002).
Measurement and correlates of family caregiver selfefficacy for managing dementia. Aging & Mental Health, 6,
153160.
Gallicchio, L., Siddiqi, N., Langenberg, P., & Baumgarten, M.
(2002). Gender differences in burden and depression
among informal caregivers of demented elders in the community. International Journal of Geriatric Psychiatry, 17(2),
154163.
Gallagher-Thompson, D., & Steffen, A. M. (1994). Comparative
effects of cognitive-behavioural and brief psychodynamic
psychotherapies for depressed family caregivers. Journal of
Consulting and Clinical Psychology, 62, 543549.

Gaugler, J. E., Pearlin, L. I., Davey, A., & Zarit, S. H. (2000).

Modelling caregiver adaptation over time: The longitudinal
impact of behaviour problems. Psychology and Aging, 15,
437450.
Gignac, M. A. M., & Gottlieb, B. H. (1996). Caregivers
appraisals of efficacy in coping with dementia. Psychology and
Aging, 11, 214225.
Katz, S., Ford, A. B., Moskowitz, R. W., Jackson, B. W., & Jaffe,
M. (1963). Studies of illness in the aged: The index of ADL,
a standardized measure of biological and psychosocial function.
Journal of the American Medical Association, 185, 914919.
Kohut, F. J., Berkman, L. F., Evans, D. A., & Cornoni-Huntley, J.
(1993). Two shorter forms of the CES-D depression symptoms
index. Aging & Mental Health, 5, 179193.
Li, L. W., Seltzer, M. M., & Greenberg, J. S. (1999). Change in
depressive symptoms among daughter caregivers: An 18-month
longitudinal study. Psychology and Aging, 14, 206219.
Nagaratnam, N., Lewis-Jones, M., Scott, D., & Palazzi, L.
(1998). Behavioural and psychiatric manifestations in dementia
patients in a community: Caregiver burden and outcome.
Alzheimer Disease & Associated Disorders, 12, 330334.
Palmer, M., & Steffen, A. (2004). Caregiving self-efficacy
as a predictor of nursing home placement in dementia
patient-caregiver dyads. Paper presented at the Association
for the Advancement of Behaviour Therapy, New Orleans, LA.
Roccaforte, W. H., Burke, W. J., Bayer, B. L., & Wengel, S.
(1994). Reliability and validity of the Short Portable Mental
Status Questionnaire administered by telephone. Journal of
Geriatric Psychiatry and Neurology, 7, 3338.
Roth, D. L., Burgio, L. D., Gitlin, L. N., Gallagher-Thompson, D.,
Coon, D. W., Belle, S. H., et al. (2003). Psychometric
analysis of the Revised Memory and Behaviour Problems
Checklist: Factor structure of occurrence and reaction ratings.
Psychology and Aging, 18, 906915.
Roth, D. L., Haley, W. E., Owen, J. E., Clay, O. J., & Goode,
K. T. (2001). Latent growth models of the longitudinal effects
of dementia caregiving: A comparison of African, American
and White family caregivers. Psychology and Aging, 16(3),
427436.
Schulz, R., OBrien, A. T., Bookwala, J., & Fleissner, K. (1995).
Psychiatric and physical morbidity effects of dementia
caregiving: Prevalence, correlates, and causes. The
Gerontologist, 35, 771791.
Stanley, M. A., & Maddux, J. E. (1986). Self-efficacy expectancy
and depressed mood: An investigation of causal relationships.
Journal of Social Behaviour and Personality, 1, 575586.
Steffen, A. M., McKibbin, C., Zeiss, A. M., GallagherThompson, D., & Bandura, A. (2002). The revised scale
for caregiving self-efficacy: Reliability and validity studies.
Journals of Gerontology: Psychological Sciences, 57, 7486.
Teri, L. (1997). Behavior and caregiver burden: Behavioral
problems in patients with Alzheimers disease and its association with caregiver distress. Alzheimer Disease & Associated
Disorders, 11, S35S38.
Teri, L., Truaz, P., Logsdon, R., Zarit, S., Uomoto, J., &
Vitaliano, P. P. (1992). Assessment of behavioral problems
in dementia: The Revised Memory and Behavior Problems
Checklist. Psychology and Aging, 7, 622631.
Ventry, I. M., & Weinstein, B. (1982). The hearing handicap
inventory for the elderly. A new tool. Ear and Hearing, 3,
128134.
Wolffsohn, J. S., & Cochrane, A. L. (2000). Design of the
Low Vision-Quality-of-Life Questionnaire (LVQOL) and
measuring outcome of low-vision rehabilitation. American
Journal of Ophthalmology, 130, 793802.
Zeiss, A. M., Gallagher-Thompson, D., Lovett, S., Rose, J., &
McKibbin, C. (1999). Self-efficacy as a mediator of caregiver
coping: Development and testing of an assessment model.
Journal of Clinical Geropsychology, 5, 221230.