Full text of DOI:10.1207/S15327558IJBM0703_05

INTERNATIONAL JOURNAL OF BEHAVIORAL MEDICINE, 7(3), 256–270
Copyright © 2000, Lawrence Erlbaum Associates, Inc.
Development of a Depression Scale
for Veterans and War Widows
Julie E. Byles, Nick Higginbotham,
Brendan Goodger, Meredith Tavener, Agatha Conrad,
Peter Schofield, and Danielle Anthony
We developed and evaluated a brief measure of depression for use within a population
of older Australian war veterans and war widows. Derived from the Geriatric Depres-
sion Scale (GDS), the 12-item GDS–Veterans is designed to include items that most
closely represent the thoughts and feelings of older veterans in relation to their war ex-
periences. The scale was administered to 1,620 veterans and widows concurrent with
the 36-item Medical Outcomes Study Short Form (MOS SF–36) quality of life mea-
sure. Of those surveyed, 13.5% indicated that they often or always worry about things
that happened during the war, indicating that this item tapped an important dimension
for many of the veteran population. Scores on the GDS-Veterans were strongly corre-
lated with the Mental Health subscale (–.72) and the Mental Health Component Sum-
mary Score (–.68) of the MOS SF–36 quality of life measure.
Key words: depression, aged, war veterans, scale development
Although clinical depression defined according to the Diagnostic and Statistical
Manual of Mental Disorders (3rd ed.; American Psychiatric Association, 1980) is
less common among the elderly than younger adults, the prevalence of depressive
Julie E. Byles, Nick Higginbotham, Brendan Goodger, and Meredith Tavener, Centre for Clinical
Epidemiology and Biostatistics, Faculty of Medicine and Health Science, The University of Newcastle,
New South Wales, Australia; Agatha Conrad, Department of Psychology, Faculty of Science, The Uni-
versity of Newcastle, New South Wales, Australia; Peter Schofield, James Fletcher Hospital, Newcas-
tle, Australia; Danielle Anthony, Centre for Clinical Epidemiology and Biostatistics, Faculty of Medi-
cine and Health Science, The University of Newcastle, New South Wales, Australia.
This research was funded by the Australian Commonwealth Department of Veterans Affairs.
Correspondence concerning this article should be sent to Meredith Tavener, Centre for Clinical Epi-
demiology and Biostatistics, Faculty of Medicine and Health Science, The University of Newcastle,
New South Wales, Australia 2308.

DEPRESSION AMONG WAR VETERANS AND WIDOWS
257
symptoms increases with age (Weissman et al., 1985). It is estimated that around
% of older people in the Australian community suffer major depression
Henderson et al., 1993) and that 20% to 30% of community-dwelling elderly expe-
1
(
rience depressive symptoms (Blazer, Burchett, Service, & George, 1991). How-
ever, in many cases depressive symptoms in the elderly are unrecognized, and
therefore depression is underdiagnosed (Livingston, Manela, & Katona, 1997),
particularly among elderly men (Garrard et al., 1998).
Undiagnosed depression and depressive symptoms can have negative conse-
quences for individuals, their families, and the community (Gurland, Copeland,
Kuriansky, et al., 1983; Henderson et al., 1993; Kay et al., 1985). The most serious
consequences of these depressive feelings are reflected in the high rates of suicide
among people over 65 years of age (Burdekin, 1993). Elderly people who experi-
ence depressive symptoms also are among the highest users of health care services
(Druss, Rohrbaugh, & Rosenheck, 1999; Livingston et al., 1997; Simon, von
Korff, & Barlow, 1995). Woods (1996) confirmed this assertion, stating that the
increased service use associated with depression may overload professionals, thus
reducing the effectiveness of the service itself.
Given that depressive symptoms are common, potentially serious, and
underdiagnosed in the elderly, a brief assessment of depressive symptomatology
would be useful as part of comprehensive geriatric assessment (Carpenter, 1996).
In Australia, the Commonwealth Department of Veterans’ Affairs is developing a
home-based geriatric assessment for community-dwelling older people, which in-
cludes the assessment of depressive symptoms as a necessary component of the in-
tervention (Byles, Harris, Nair, & Butler, 1996).
For war veterans and their widows, depressive symptoms may take on another
dimension. When veterans returned from war they were expected to readjust to
work, marriage, and families, and this adjustment and the subsequent appearance
of psychological malfunctioning initiated detailed postwar research (Elder &
Clipp, 1989). The results of this detailed research concluded that war veterans ex-
perienced chronic impairment 20 years after their return from World War II; their
symptomatology was continual and increased with time.
Memories and emotions associated with combat events that individuals sup-
press and downplay in order to fulfill a “normal life” can reemerge in later life and
lead to depressive symptoms (Elder & Clipp, 1989). The triggers include losses
such as the death of a spouse, siblings, and friends (particularly friends through
war service). These losses leave the surviving elderly person with a diminished so-
cial support network. Other relevant losses may include the loss of health and
physical ability and retirement from employment (Elder & Clipp, 1989). The cop-
ing mechanisms that the aging person has relied on may weaken, increasing the
impact of these losses and exposing the person to a greater risk of depression.
The effect of war in later life has not been examined sufficiently. Investigation
is necessary so that undetected depression among war veterans can be managed ef-

258
BYLES ET AL.
fectively, especially because war veterans may hold a greater risk of experiencing
depressive symptomatology.
Assessment and classification of depression in older people can be difficult
(Snaith, 1987). Older people manifest fewer of the somatic symptoms of depres-
sion, and depressive symptoms may be masked by comorbidities (Kendall, 1976).
The Geriatric Depression Scale (GDS) originated in 1983 (Yesavage et al., 1983)
as a scale to measure depressive symptoms in an elderly population. The questions
focus on sad mood, lack of energy, positive mood, agitation, and social with-
drawal. Higher scores reflect more depression. The GDS originally consisted of 30
items but was later shortened to provide a more concise 15-item scale suitable for
use by health professionals.
Versions of the GDS have been used extensively in clinical and research set-
tings and have been shown to have good reliability and validity. The 30-item ver-
sion has been shown be internally consistent (Cronbach’s a = .94), interrater
reliability has been recorded at .85, and test–retest reliability coefficients range
from .86 to .98 for periods of 1 hr to 10 to 12 days (McDowell & Newell, 1996).
GDS scores correlate with the Hamilton Rating Scale—Depression (.62–.82), the
Beck Depression Inventory (BDI; .85), Montgomery Asberg Depression Rating
Scale (MADRAS; .82), and the Centre for Epidemiological Studies—Depression
(.82 with stroke patients). The 15-item version correlates well with the 30-item
version (coefficients range from .66 to .89). When the Structured Clinical Inter-
view was used as the gold standard, the optimal sensitivity of the 15-item version
for major depression was 92% and the specificity was 81%. The corresponding
values for the 30-item version were 100% and 84% (Lyness et al., 1997). Other re-
search has established that the GDS is appropriate for use in clinical and commu-
nity settings (McDowell & Newell, 1996) and can be used to detect depression in
patients with dementia (Lichtenberg, Marcopoulos, Steiner, & Tabscott, 1992).
Ease of comprehension is an important consideration when choosing a depres-
sion screening scale for the elderly. The GDS has a format such that questions do
not rely on the participant’s memory. “Are you,” “do you,” and “how often” form
the question structure, placing emphasis on the recognition of “now,” and remov-
ing demand from memory function. The Hamilton Depression Rating Scale, for
example, requires both memory function and descriptive expression of recent ex-
periences and feelings (Lichtenberg et al., 1992). When available depression
scales are compared, the GDS and the BDI are comparable, showing the highest
levels of reliability and validity (McDowell & Newell, 1996). Length and time
needed for administration may be the only tangible difference between the two
scales: The GDS is quicker in administration time and therefore was a more appro-
priate scale to use in this study.
To allow recognition of war-related thoughts and concerns, we created a modi-
fied 12-item GDS. This modification of the original 30-item GDS followed the
conduct of three focus groups involving men (two groups) and women (one

DEPRESSION AMONG WAR VETERANS AND WIDOWS
259
group). These groups provided an “insiders’ view” of those factors considered to
be important for the quality of life for older veterans and war widows. As a result,
items from the GDS deemed least suitable for use with Australian veterans and war
widows were excluded (n = 19). Disturbance by thoughts about war-related events
was evident from the group discussions, and a single specific item was included to
cover these ideations. Therefore, a final scale was developed consisting of
1
2-items that were most closely related to the findings of the veteran focus group
discussions.
After pilot testing with 58 veterans and widows, the response options were ex-
panded to a 5-point categorical scale to allow greater discrimination of responses.
To discourage acquiescence bias, approximately half of the items were positively
worded and the rest were negatively worded. In this article we present data on the
validation of this modified 12-item scale (the GDS–Veterans) and the prevalence
of symptoms within a sample of Australian veterans and war widows age 70 years
and older.
METHODS
Validation of the checklist was undertaken as part of the Preventive Care Trial
funded by the Australian Commonwealth Department of Veterans’ Affairs. The
Preventive Care Trial aims to evaluate the benefits of regular home visits to identify
veterans’ health care needs and to meet these needs through primary health care and
other relevant health and community services. The home visits are undertaken by a
trained health professional, using a structured checklist developed specifically for
thetrial. Theareascoveredbythechecklistincludehearing, vision, dental, vaccina-
tions, medications, nutrition, smoking, foot care, continence, mobility, cognitive
function, sleep, community services, social support, depression, and home safety.
For each area, needs are recorded, discussed with the veteran, and also reported to
the person’s local medical officer. Obviously, the process includes an expectation
of effective treatment or intervention. However, frequently only minimal assis-
tance or intervention is required, and the needs may be more for social and instru-
mental support or aids rather than medical intervention.
Sample
Potential participants were selected randomly from the Department of Veterans’
Affairs database of entitled veterans and war widows. To be eligible for inclusion in
the sample, veterans had to be at least 70 years old and had to live in selected postal
areas in 10 geographical regions in two Australian states (6 regions in New South
Wales and 4 regions in Queensland). Half the areas in each state were from urban

260
BYLES ET AL.
zones; the other areas were zoned rural or remote. Within each area, selection was
stratified to allow equal representation of men and women and overrepresentation
of older people (85 years and over).
The Department of Veterans’ Affairs mailed randomly selected veterans and
widows an invitation to participate in the 3-year trial of home visits. A thank
you/reminder card was mailed to all veterans and widows 1 and 3 weeks after the
original invitation. Interested veterans and widows completed and returned a con-
sent form, providing contact details for the baseline telephone survey.
Telephone Survey
Consenting veterans and widows were surveyed using the computer-assisted tele-
phone interview (CATI) system. The telephone survey took an average of 30 min
per participant. The questions included the 36-item Medical Outcomes Study Short
Form (MOS SF–36) quality of life measure, the 11-item Duke’s Social Support In-
dex (DSSI), the GDS–Veterans, items about hospital and nursing home or hostel
admission, and other health and community service utilization in the past 12
months.
The MOS SF–36 is a widely used and well-validated measure of health-related
quality of life (Ware & Sherbourne, 1992). The validity of the MOS SF–36 for use
with elderly populations also has been established in Australia (McCallum, 1995;
McHorney, Ware, Lu, & Sherbourne, 1994; Pitt, Schurink, Nair, Byles, & Heller,
1
1
996) as has its validity for the depressed elderly (Beusterien, Steinwald, & Ware,
996). The instrument provides an eight-scale health profile and two component
summary scores representing physical and mental health. The scales measure
physical functioning; bodily pain; role limitations due to physical health problems;
general health perceptions; vitality, energy, or fatigue; general mental health, cov-
ering psychological distress or well-being; role limitations due to emotional prob-
lems; and social functioning. The DSSI (Koenig et al., 1993) consists of two
subscales. The first measures social interaction including the number of people
who can be relied on, the number of social contacts, and participation in group ac-
tivities. The second subscale involves subjective evaluation of the quality of the
social interaction available. The full scale has been validated for use with older
people in Australia (Goodger, Byles, Higginbotham, & Mishra, 1999).
Analysis
Baseline data collected by CATI were analyzed as follows. For the modified GDS,
negatively worded items were recoded so that a high score related to a more nega-
tiveaffect. TotalGDSscoresthenwerecalculatedforeachparticipantbyaddingto-

DEPRESSION AMONG WAR VETERANS AND WIDOWS
261
gether his or her responses across the 12 items. Item analyses were performed in-
cluding the percentage endorsement of each item and inter-item and item-total
correlations.
We checked the internal consistency of the scale by using Cronbach’s alpha and
we undertook exploratory and confirmatory factor analyses by using principle
components analysis with varimax rotation (Stata Statistical Software, Stata Cor-
poration, College Station, TX). The factor loadings were used to weight item
scores for further analysis.
Univariate and multivariate analyses (analysis of variance) were conducted to
assess the correlation between the weighted GDS scores and the MOS SF–36 and
to identify other factors associated with high scores (age, sex, urban or rural zone,
service use, DSSI, education, occupation, marital status). MOS SF–36 scores were
calculated as recommended by its developers (Ware, Kosinksi, & Keller, 1994),
and Australian population norms were used to calculate component scores (Aus-
tralian Bureau of Statistics, 1995).
All results were weighted to correct for oversampling of people aged 85 years
and older.
RESULTS
Of the total sample of 5,354 veterans, 26 were ineligible (dead, in nursing home, too
young, or out of area), and 10 were not at the address listed. Of the remainder, 1,967
(37%) agreed to participate in the study, and 1,620 were successfully interviewed.
Others either refused at the time of interview (n = 142), could not be contacted for
the interview (n = 73), or were not contacted because the required sample size for
the trial had been exceeded (n = 132).
Of those interviewed, 54% were men (n = 876) and 46% were women (n = 744).
Stated ages ranged from 64 to 97 years. In comparison with the 1996 census (Aus-
tralian Bureau of Statistics, 1996), participants were representative in terms of ed-
ucation level and occupational status. Married women were greatly
underrepresented (4%), because “war widow” is the main entitlement category for
women under Department of Veterans’ Affairs criteria. Of the men, 74% were
married, which is much higher than census data for Australian men of the same age
but comparable to the veteran population (data provided by Australian Common-
wealth Department of Veterans’ Affairs, 1999).
On each of the GDS–Veterans items, the responses were well distributed across
the range of options. However, positively worded items were less frequently en-
dorsedatthedepressedendofthescalethannegativelywordeditems(seeTable1).
Inter-item correlations were acceptable (.2–.3, n = 22), moderate (.3–.4, n =
2
1), or high (>0.4, n = 16). Item-total correlations were all greater than 0.5 ex-
cept for “Worry about things that happened during the war” (correlation = .4).

TABLE 1
Response to Items in the Modified Geriatric Depression Scale
Endorsement (%)
Malesa
Femalesb
Positively Worded Items and Response Options
Are you basically satisfied with your life?
Always
Often
Sometimes
Seldom
49
27
20
2
47
33
16
3
Never
2
1
Are you in good spirits?
Always
Often
Sometimes
Seldom
37
42
18
2
43
44
12
1
Never
1
0
Do you think it is wonderful to be alive now?
Always
Often
Sometimes
Seldom
74
14
8
64
18
14
3
3
Never
1
1
Are you hopeful about the future?
Always
Often
Sometimes
Seldom
44
21
20
10
5
42
23
22
8
Never
5
Do you enjoy getting up in the morning?
Always
Often
Sometimes
Seldom
60
19
13
5
54
20
17
5
Never
3
4
Negatively Worded Items and Response Options
Do you feel your life is empty?
Never
48
22
21
7
50
19
24
5
2
Seldom
Sometimes
Often
Always
2
(Continued)
262

TABLE 1
(Continued)
Endorsement (%)
Malesa Femalesb
Negatively Worded Items and Response Options
How often do you get bored?
Never
32
47
Seldom
Sometimes
Often
Always
31
24
11
2
26
20
6
1
How many activities or interests have you dropped?
None
49
17
14
14
6
65
18
7
8
2
1
or 2
Some
Most
All
How often do you feel helpless?
Never
Seldom
Sometimes
Often
35
25
28
9
31
30
30
7
Always
3
2
How often do you prefer to stay at home rather than go out and do
new things?
Never
Seldom
Sometimes
Often
7
12
36
31
14
15
20
40
18
7
Always
Are you bothered by thoughts you can’t get out of your head?
Never
Seldom
Sometimes
Often
42
21
26
8
49
19
24
7
Always
3
1
How often do you worry about things that happened during the war?
Never
38
17
27
13
5
52
20
20
6
Seldom
Sometimes
Often
Always
2
a
b
n = 876. n = 744.
263

264
BYLES ET AL.
All the items were retained in the scale. Cronbach’s alpha indicated high levels
of internal consistency (.85 for men and .83 for women).
Exploratory factor analysis revealed one main factor and a smaller secondary
factor related to negatively worded items (eigenvalues = 4.65, 1.27). These two
factors explained 48% of the varience.
The distribution of the total scores is presented in Figure 1 for 1,542 veterans
and widows who provided responses to all items. The percentage of missing data
was low; endorsement of iems ranged from 100% response (“How often do you
get bored?”) to 97.23% (“Are you hopeful about the future?”). The scores ranged
from a minimum of 7.4 to a maximum of 33.6 (60 was the highest possible score).
The mean score was 14.75 (SD = 5.5). The total score demonstrated a high correla-
tion with MOS SF–36 subscale scores, particularly the Mental Health subscale
(–.72) and the Mental Health Component Summary Score (–.68) (see Table 2).
These correlations are negative because a higher score on the GDS–Veterans scale
indicates more depressive thoughts and feelings, whereas a higher score on the
MOS SF–36 subscales indicates a more favorable health-related quality of life
profile.
Univariate analyses revealed that the GDS–Veterans was significantly associ-
ated with urban or rural zone (p = .002), marital status (p = .01), sex (p = .001), total
DSSI score (p = .0001), and use of community services (p = .001). Multivariate
analyses found that the GDS–Veterans score was significantly associated with the
use of community services (p = .02) and negatively associated with social support
(those with more support had less depression; p = .001). Urban versus rural zone of
residence, age, sex, marital status, and past occupation were retained in the model
FIGURE 1 GDS–Veterans’ scores (N = 1,542).

DEPRESSION AMONG WAR VETERANS AND WIDOWS
265
TABLE 2
Correlation Between Modified GDS Scores and SF–36 Subscales
Subscale
Pearson Correlation Coefficient
Mental Health
–.72
Role Emotional
Social Functioning
Vitality
Mental Health Component Summary Score
General Health
Bodily pain
–.48
–.52
–.61
–.68
–.57
–.4
Role Limitations Physical
Physical Functioning
Physical Health Components Summary Score
–.42
–.45
–.29
but were not significantly associated with the GDS–Veterans score. The full model
explained 31% of the variance in scores.
DISCUSSION
The GDS–Veterans scale applied in this study has demonstrated acceptable levels
of validity for use as a measure of depressive symptoms within the veteran popula-
tion. Evidence of scale validity is shown by the high correlation between the
GDS–Veterans scores and the MOS SF–36 Mental Health subscale. The Mental
Health subscale has been reported to be useful in measuring depression in elderly
populations (Beusterien et al., 1996) and is therefore a suitable comparison for as-
sessing validity. In our analysis, the GDS–Veterans scores correlated most highly
with the MOS SF–36 Mental Health subscale and the mental health component
summary score. These correlations were of higher magnitude than Beusterien et al.
(1996) observed in a comparison of the MPS SF–36 Mental Health subscale and the
Hamilton Depression Scale score.
The distribution of depression scores for participants in the Preventive Care
Trial shows that most people scored at the lower end of the scale (not depressed).
Participants in this study, however, are likely to underrepresent depressed elderly,
who may be less likely to participate in research of this type. The consent rate of
3
7% limits the extent to which the results can be generalized to all Australian vet-
erans and widows.
A small number of veterans (n = 78) taking part in the survey did not complete
all items comprising the GDS–Veterans scale. Using a case-wise deletion method,
we dropped these respondents from further analysis. To increase the validity of the
scale for these people, other standard approaches to missing data, such as mean
substitution, may need to be employed. Otherwise the validity of the scale is lim-
ited to those people who are able to answer all items.

266
BYLES ET AL.
For veterans within the trial, the scale seems able to discriminate between vari-
ous levels of depressive symptomatology. Although there is no clear cutoff for de-
ciding if an individual is depressed, the higher the score the greater the reporting of
depressive symptoms. In the absence of other indicators, the 90th percentile
(22/60) provides one useful benchmark for those at the extreme end of this contin-
uum who should be more closely evaluated for depression.
Although several groups such as the U.S. Preventive Services Task Force
(1996) and the Canadian Task Force on the Periodic Health Examination (1994)
conclude there is insufficient evidence to support screening for depression, eval-
uation of depression is explicitly listed as a component of comprehensive geriat-
ric assessment (Carpenter, 1996). Other researchers have shown that the use of
screening tests can increase detection of depression (Linn & Yager, 1980;
Magruder-Habib & Zung, 1990; Moore, Lilimperi, & Bobula, 1978; Zung et al.,
1
983), and treatment of depression in the elderly by using cognitive therapy as
well as medication has been shown to be effective in randomized controlled tri-
als (e.g., Old Age Depression Interest Group, 1993; Shea et al., 1992) Also, be-
cause depression can be a side effect of other therapies (e.g., beta-blocker
medication), a thorough clinical assessment and medication review form the
foundation of therapy.
Mental symptoms experienced by elderly populations are more likely to result
in medical rather than mental health treatment (Druss et al., 1999), suggesting that
assistance for the elderly presently is focused in a medical–biological framework
that bypasses the psychological cause for the complaints. The potential contribu-
tion of war-related events to current psychological functioning needs to be consid-
ered further and may provide important avenues for effective psychotherapeutic
strategies for this population.
Our modification of the GDS scale arose from recognition of the relevance of
war-related experience to veterans’ current mental state. Focus group discussions
with older Australian veterans and war widows suggested that thoughts and feel-
ings about the war may feature prominently in veterans’ daily lives. Of the men in
the study, 11% were often or always bothered by thoughts they “could not get out
of (their) head,” and 18% often or always worried about “things that happened dur-
ing the war” (13.5% of participants overall). The additional item regarding the war
was significantly correlated with the depression total, although it was the most
weakly correlated of all 12 items in the scale.
In Australia, people currently in their 70s and 80s belong to a culture that de-
rived a strong part of its identity from the national contribution to two world wars
(Walker-Birckhead & Davison, 1995), and these events retain their significance
particularly for those who personally experienced the conflict.
It is important to understand the current physical and emotional state of older
people in this context (Walker-Birckhead & Davison, 1995) and to acknowledge
that war-related issues may affect the current lives of older veterans and war wid-

DEPRESSION AMONG WAR VETERANS AND WIDOWS
267
ows. To date, very little research has been done to understand how these experi-
ences affect people as they age (Walker-Birckhead & Davison, 1995).
War-related psychiatric symptomatology has been shown to be significant and
persistent over long periods of time (Archibald & Tuddenham, 1965). Sutker and
Allain (1996) reported prevalence of major depression over a lifetime for World
War II and Korean Conflict prisoner of war survivors and combat veterans as high
as 42%. Within the Australian population of Vietnam veterans, about half of those
veterans who have ever had posttraumatic stress disorder (PTSD) still have it today
(Creamer, Morris, Biddle, & Elliott, 1999).
More recently a phenomenon has been described whereby war-related condi-
tions such as anxiety, depression, insomnia, alcoholism, and PTSD are experi-
enced for the first time in older adult life (Elder & Clipp, 1989). These symptoms
and conditions may be triggered by traumatic current events including the onset of
ill health or retirement (Van Dyke, Zilberg, & McKinnon, 1985). Retirement also
may mean the loss of occupation and distraction previously used to fend off intru-
sive thoughts and depressive symptoms. Furthermore, memories are known to
play an important role in the process of aging (Coleman, 1986), and traumatic
memories may be more prominent at this time of life.
In European countries, not only military veterans have been observed to suffer
late-onset symptoms related to their war experience but also concentration camp
survivors, resistance members, and civilians whose lives and homelands were dis-
rupted by combat and occupation (Hovens, Falger, Opden Velde, et al., 1992). For
example, Hovens et al. (1992) found that 56% of resistance veterans suffered from
PTSD. Similarly, Robinson et al (1990) reported that 75% of holocaust survivors
suffered from one or more symptoms such as hypermnesia, depressive moods,
nightmares, and anxiety. Thus, although the prevalence of war-related psychiatric
symptoms among older people is unknown, it is likely that a very large number of
older people are affected.
Although this study concentrated on depression, other psychiatric symptoms
also may be important for this population. Other studies have reported on the prev-
alence of PTSD and other symptoms of depression and anxiety (Long, MacDon-
ald, & Chamberlain, 1996), and worry about “things that happened during the
war” may be better correlated with PTSD. However, Hyer, Stanger, and
Boudewyns (1999) reported that depression, not PTSD, was more related to health
status and social support in a group of elderly male combat veterans, hence our se-
lection of depression as the most relevant psychiatric condition to assess in this
population. In terms of the screening methodologies employed in the Preventive
Care Trial, the modified 12-item GDS–Veterans proved adequate in identifying
veterans with war-related concerns for further follow-up and referral.
If screening for depressive symptoms among the elderly is to be adopted, then
specific questioning about war-related experience should be included, particularly
among war veterans. Without more general acknowledgment of the impact of

268
BYLES ET AL.
symptoms and the importance of the social context and life experience, these peo-
ple’s symptoms are likely to be unrecognized, misdiagnosed, and poorly treated.
More accurate assessment could lead to better treatment and support for a large
proportion of older people experiencing depressive symptoms, thus significantly
improving the quality of their experience of aging.
REFERENCES
American Psychiatric Association. (1980). Diagnostic and statistical manual of mental disorders (Xth
ed.). Washington, DC: American Psychiatric Association.
Archibald, H. C., & Tuddenham, R. D. (1965). Persistent stress reaction after combat: A 20-year
follow-up. Archives of General Psychiatry, 12, 475–481.
Australian Bureau of Statistics. (1995). National health survey: SF–36 population norms, Australia.
Canberra: Australian Bureau of Statistics.
Australian Bureau of Statistics. (1996). Selected social and housing characteristics: 1996 census of
population and housing (Catalogue No. 2015). Canberra: Australian Bureau of Statistics.
Australian Commonwealth Department of Veterans’ Affairs. (1999). [Description of data]. Unpub-
lished data.
Beusterien, K. M., Steinwald, B., & Ware, J. (1996). Usefulness of the SF–36 Health survey in measur-
ing health outcomes in the depressed elderly. Journal of Geriatric Psychiatry and Neurology, 9,
1
3–21.
Burdekin, B. (1993). Human rights and mental illness –Report of the national inquiry into human rights
of people with mental illness. Canberra, Australia: AGPS.
Blazer, D. G., Burchett, B., Service, C., & George, L. K. (1991). The association of age and depression
among the elderly: An epidemiologic exploration. Journal of Gerontology, 46, 210–215.
Byles, J. E., Harris, M. A., Nair, B. R., & Butler, J. (1996). Preventive health programs for older Austra-
lians. Health Promotion Journal of Australia, 6, 37–43.
Carpenter, I. (1996). Value of screening old age. In S. Ebrahim & A. Kaleche (Eds.), Epidemiology in
old age. London: BMJ.
Canadian Task Force on the Periodic Health Examination. (1994). Canadian guide to clinical preven-
tive health care. Ottowa, ON: Canada Communication Group.
Coleman, P. G. (1986). Ageing and reminiscence processes: Social and clinical implications. New
York: Wiley.
Creamer, M., Morris, P., Biddle, D., & Elliott, P. (1999). Treatment outcome in Australian veterans with
combat-related post-traumatic stress disorder: A cause for cautious optimism? Journal of Traumatic
Stress, 12, 545–558.
Druss, B. G., Rohrbaugh, R. M., & Rosenheck, R. A. (1999). Depressive symptoms and health costs in
older medical patients. American Journal of Psychiatry, 156, 477–479.
Elder, G. H., Jr., & Clipp, E. C. (1989). Combat experience and emotional health: Impairment and resil-
ience in later life. Journal of Personality, 57, 311–341.
Garrard, J., Rolnick, S. J., Nitz, N. M., Luepke, L., Jackson, J., Fischer, L. R., Leibson, C., Bland, P. C.,
Heinrich, R., & Waller, L. A. (1998). Clinical detection of depression among community-based
elderly people with self-reported symptoms of depression. Journal of Gerontology: Series A,
Biological Sciences and Medical Sciences, 53, M92–101.
Goodger, B., Byles, J. E., Higginbotham, N., & Mishra, G. (1999). Assessment of a short scale to mea-
sures social support among older people. Australian and New Zealand Journal of Public Health, 23,
2
60–265.

DEPRESSION AMONG WAR VETERANS AND WIDOWS
269
Gurland, B., Copeland, J., Kuriansky, J., Kelleher, M., Sharpe, L., & Dean, L. (1983). The mind and
mood of ageing. London: Croom Helm.
Henderson, A. S., Jorm, A. F., MacKinnon, A., Christensen, H., Scott, L. R., Korten, A. E., & Doyle, C.
(1993). The prevalence of depressive disorders and the distribution of depressive symptoms in later
life: A survey using draft ICD–10 and DSM–IIIR. Psychological Medicine, 23, 719–724.
Hovens, J. E., Falger, P. R. J., Opden Velde, W., Schouter, E. G. W., DeGroer, J. H. M. & Van Duling, H.
(1992). Occurrence of posttraumatic stress disorder among Dutch World War II resistance veterans
according to the SCID. Journal of Anxiety Disorders, 6, 147–157.
Hyer, L., Stanger, E., & Boudewyns, P. (1999). The interaction of post-traumatic stress disorder and de-
pression among older combat veterans. Journal of Clinical Psychology, 55, 1073–1083.
Kay, D., Henderson, A., Scott, R., Wilson, J., Rickwood, D., Blazer, D. G., & Hybels, C. (1985). Dementia
and depression amongst the elderly living in the Hobart community: The effect of the diagnostic
criteria on the prevalence rates. Psychological Medicine, 15, 771–788.
Kendall, R. (1976). The classification of depressions: A review of contemporary confusion. British
Journal of Psychiatry, 129, 15–28.
Koenig, H. G., Westlund, R., George, L. K., Hughes, D. C., Blazer, D. G., & Hybels, C. (1993). Abbrevi-
atingtheDukesocialsupportindexforuseinchronicallyillelderlyindividuals. Psychosomatics, 34,
6
1–69.
Lichtenberg, P. A., Marcopulos, B. A., Steiner, D. A., & Tabscott, J. A. (1992). Comparison of the Ham-
ilton Depression Rating Scale and the Geriatric Depression Scale: Detection of depression in de-
mentia patients. Psychological Reports, 70, 515–521.
Linn, L. S., & Yager, J. (1980). The effect of screening, sensitization, and feedback on notation of de-
pression. Journal of Medical Education, 55, 942–949.
Livingston, G., Manela, M., &Katona, C. (1997). Costofcommunitycareforolderpeople. BritishJour-
nal of Psychiatry, 171, 56–59.
Long, N., MacDonald, C., & Chamberlain, K. (1996). Prevalence of post-traumatic stress disorder, de-
pression and anxiety in a community sample of New Zealand Vietnam War veterans. Australian and
New Zealand Journal of Psychiatry, 30, 253–256.
Lyness, J. M., Noel, T. K., Cox, C., King, D. A., Conwell, Y., & Caine, E.D. (1997). Screening for
depression in elderly primary care patients. A comparison of the Center for Epidemiologic
Studies–Depression Scale and the Geriatric Depression Scale. Archives of Internal Medicine, 157,
4
49–454.
McCallum, J. (1995). The SF–36 in an Australian sample: Validating a new generic health status mea-
sure. Australian Journal of Public Health, 19, 160–166.
McDowell, I., & Newell, C. (1996). Measuring health: A guide to rating scales and questionnaires.
New York: Oxford University Press.
McHorney, C. A., Ware, J. E., Lu, R., & Sherbourne, C. D. (1994). The MOS 36-Item Short-Form
Health survey (SF–36): III. Tests of data quality, scaling assumptions, and reliability across diverse
patient groups. Medical Care, 2, 40–66.
Magruder-Habib, K., & Zung, W. W. K. (1990). Improving physicians’ recognition and treatment of de-
pression in general medical care: Results from a randomized clinical trial. Medical Care, 28,
2
39–250.
Moore, J. T., Lilimperi, D. R., & Bobula, J. A. (1978). Recognition of depression by family medicine
residents: The impact of screening. Journal of Family Practice, 7, 509–513.
Old Age Depression Interest Group. (1993). How long should the elderly take anti-depressants? A dou-
ble-blind placebo-controlled study of continuation/prophylaxis therapy with dothiepin. British
Journal of Psychiatry, 162, 175–182.
Pitt, S. W., Schurink, J., Nair, B. R., Byles, J. E., & Heller, R. F. (1996). Use of the Short-Form-36 health
survey to assess quality of life among Australian elderly. Australian Journal on Ageing, 15,
1
32–135.

270
BYLES ET AL.
Robinson, S., Rapaport, J., Durst, R., Rapaport, M., Rosca, P., Metzer, S., & Zilberman, L. (1990). The
late effects of Nazi-persecution among elderly Holocaust survivors. Acta Psychiatrica
Scandinavica, 82, 311–315.
Shea, M., Elkin, I., Imber, S., Sotsky, S. M., Watkins, J. T., Collins, J. F., Pilkonis, P. A., Beckham, E.,
Glass, D. R., Dolan, R. T. (1992). Course of depressive symptoms over follow-up. Archives of Gen-
eral Psychiatry, 49, 782–787.
Simon, G. E., von Korff, M., & Barlow, W. (1995). Health care costs of primary care patients with rec-
ognized depression. Archives of General Psychiatry, 52, 850–856.
Snaith, R. P. (1987). The concepts of mild depression. British Journal of Psychiatry, 150, 387–393.
Sutker, P. B., & Allain, A. N. (1996). Assessment of PTSD and other mental disorders in World War II
and Korean conflict POW survivors and combat veterans. Psychological Australian Journal on
Ageing, 14, 124–127.
Ware, J. E., Jr., & Sherbourne, C. D. (1994). The MOS 36-Item Short-Form Health Survey (SF–36): I.
Conceptual framework and item selection. Medical Care, 30, 473–483.
Ware, J. E., Kosinski, M., & Keller, S. D. (1994). SF–36 Physical and Mental Health Summary Scales:
A users manual. Boston: The Health Institute.
Weissman, M. M., Meyers, J. K., Tischler, G. L., Holzer, C. E., III, Leaf, P. J., Orvaschel, H., & Brody, J.
A. (1985). Psychiatric disorders (DSM–III) and cognitive impairment among the elderly in a US ur-
ban community. Acta Psychiatrica Scandinavica, 71, 366–379.
Woods, R. T. (1996). Handbook of the clinical psychology of ageing. Chichester, NY: Wiley.
Yesavage, J. A., Brink, T. L., Rose, T. L., Lum, O., Huang, V., Adey, M., & Leirer, V. O. (1983). Devel-
opment and validation of a geriatric depression screening scale: A preliminary report. Journal of
Psychiatric Research, 17, 37–49.
Zung, W. W. K., Magill, M., Moore, J. T., & George, D. T. (1983). Recognition and treatment of depres-
sion in a family medicine practice. Journal of Clinical Psychiatry, 44, 3–6.