Content uploaded by Aida Mujkić
Author content
All content in this area was uploaded by Aida Mujkić
Content may be subject to copyright.
43(4):396-402,2002
PUBLIC HEALTH
Impact of War on Health Related Quality of Life in Croatia: Population Study
Andreja Babiæ-Banaszak, Luka Kovaèiæ, Lana Kovaèeviæ, Gorka Vuletiæ, Aida Mujkiæ, Zdravko
Ebling1
Andrija Štampar School of Public Health, Zagreb University School of Medicine, Zagreb; and 1Osijek Health Care
Center, Osijek, Croatia
Aim. To present health-related quality of life in post-war Croatia, focusing on the population as a whole rather than on
the specific group of people.
Method. The study was conducted in six Croatian counties in the 1997-1999 period. Three of those counties had been
directly affected by the 1991-1995 war. The sample consisted of 1,297 randomly selected respondents aged 18 years
and older. The questionnaire was anonymous, consisting of questions on sociodemographic characteristics of respon-
dents and Medical Outcome Study 36-item short-form health survey (SF-36). SF-36 comprised the following nine
subscales: physical functioning (PF), role-physical (RP), bodily pain (BP), general health (GH), vitality (VT), social func-
tioning (SF), role-emotional (RE), mental health (MH), and health transition (HT).
Results. Mean subscale scores for the areas directly affected by war were PF 64.21; RP 52.70; BP 59.35; GH 49.02; VT
49.52; SF 68.29; RE 63.02; MH 57.95; HT 41.28; and for the areas not affected by war were PF 65.35; RP 62.01; BP
61.79; GH 50.45; VT 49.40; SF 71.41; RE 74.11; MH 60.33; HT 45.14. The two areas differed significantly in RP
(p<0.001), SF (p=0.035), RE (p<0.001), MH (p=0.038), and HT (p=0.003). Respondents living in the areas directly
affected by war achieved lower total health-related quality of life scores. Younger respondents, respondents with sec-
ondary education, and those with lower income were the groups mostly affected by war.
Conclusion. War affects self-perceived health, physical ability, and emotional and mental health of the entire popula-
tion affected by war, especially younger age groups, those with lower education, and lower income.
Key words: Croatia; education; income; quality of life; war
Ways in which war affects human existence are
constantly explored. Health consequences of war
traumas on soldiers and war veterans (1-7) and post-
traumatic stress disorder in war victims (8-12) were
extensively investigated, as well as problems of refu-
gees and displaced persons (13-18). However, these
studies were oriented towards specific groups of peo-
ple affected by war.
The 1991-1995 war against Croatia caused de-
mographic losses and left deep psychosocial scars.
War damages were estimated at US$37.4 billion, up
to 20,000 persons have been reported killed or miss-
ing, and more than 30,000 people have been dis-
abled as a result of the war (19). Approximately
27,000 square kilometers (or 47.5% of Croatian con-
tinental territory) with approximately 1.5 million in-
habitants were affected by war (20). At the end of
1991, as much as 11.5% of population lived in partly
or completely occupied area (21). The country was
flooded by displaced persons and refugees from
neighboring Bosnia and Herzegovina. During the pe-
riod from 1992 to 1998 the number of refuges and
displaced persons was between 430,000 and 700,
000 (22).
The World Health Organization’s definition of
health as a state of complete physical, social, and
mental well-being, and not merely the absence of dis-
ease or infirmity is generally regarded as a definition
of health-related quality of life (23,24). The popula-
tion quality of life is a highly vulnerable parameter,
crucial in health care policy. Since the population of
Croatia has experienced war in recent past, we de-
cided to investigate health-related quality of life in
post-war Croatia, starting with the hypothesis that the
quality of life was lower in the areas that had been
more severely affected by war. We focused on the
population as a whole, not on the specific group of
people, and explored the effect of war experience on
their quality of life, using the Medical Outcome Study
36-item short-form health survey (SF-36). SF-36 is the
model widely used as a generic short-form measure of
functional health and well-being of different popula-
tion groups and has been applied in hundreds of stud-
ies (25).
Subjects and Methods
The study was conducted in Croatia in the 1997-1999 pe-
riod, and covered six counties: three affected and three non-af-
396 www.cmj.hr
fected by war (Table 1). A county was defined as war-affected if it
was under occupation and/or constant armed attacks during the
war period, and as non-affected if it was not under occupation
and was under armed attacks sporadically.
Subjects
Study sample consisted of 1,297 respondents: 602 from
war-affected and 695 from non-affected area (Table 2). The age
range was 18-93 years. The mean age of respondents was
50.8±17.3 years for war-affected area, and 51.5±18.6 years for
non-affected area. Respondents were divided into three income
groups, as follows: low (less than 1,500 HRK per household
monthly), medium (1,500-3,500 HRK per household monthly),
and high (more than 3,500 HRK per household monthly).
The sampling strategy was a stratified multistage sampling.
In the first stage, a random sample of health care centers within
the counties was made. Health care centers were chosen because
every citizen in Croatia is registered with a general practitioner
working in a health care center. In the second stage, a 10% ran-
dom sample of the population under care of general practitioners
within health care centers was taken. Only persons aged 18 years
and over were included in the study. If a younger person was to
be selected according to the list of random numbers, they were
skipped and the next referred person was interviewed. The se-
lected inhabitants who gave their informed consent were inter-
viewed in their homes. Each interview took about 30 minutes
and was done by trained interviewers, students of the Zagreb
University Medical School. The response rate was 96.0%.
Questionnaire
The questionnaire was anonymous, consisting of a series of
questions on socio-demographic characteristics of respondents,
including age, sex, education, and income per household, and
the Medical Outcome Study 36-item short-form health survey
(SF-36). The SF-36 survey contains 36 questions and yields a
nine-subscale profile of scores (25): Physical Functioning, Role-
Physical, Bodily Pain, General Health, Vitality, Social Func-
tioning, Role-Emotional, Mental Health, and Health Transition.
Each scale describes a certain aspect of functional health (Table
3). The Croatian version of the SF-36 questionnaire was licensed
to Andrija Štampar School of Public Health in 1992 as a part of
the project “Tipping the Balance towards Primary Healthcare
Network”, and validated on the general population in Croatia
(26,27).
Statistics
Each SF-36 scale score was transformed toa0to100scale.
The transformation converted the lowest and highest possible
scores to 0 and 100, respectively. A score between those values
represented the percentage of the total possible score achieved.
Data on SF-36 scales were expressed as a mean value and inter-
quartile range. Since the data distribution was very skewed, the
standard deviation was not a good measure of variability and the
inter-quartile range was used as an alternative.
For comparison of the two areas, independent samples
t-test was used. Differences in socio-demographic characteristics
(sex, age, education, and income) of respondents in war-affected
and non-affected area were tested by a chi-square test. A proba-
bility value of p<0.05 (two-tailed) indicated a statistically signifi-
cant difference.
Results
There were no significant differences between
the war-affected and non-affected areas regarding
sex, age, and educational level of respondents. How-
ever, significant difference between war-affected and
war non-affected areas was found regarding income.
There were more respondents in low-income group
in war-affected area (chi-square=15.001, df=2,
p<0.001).
On the SF-36 subscales, the mean scores ranged
from 41.28 for Health Transition to 68.29 for Social
Functioning in war-affected area and from 45.14 for
Health Transition to 74.11 for Role-Emotional in
non-affected area. The differences in scores between
the two groups were analyzed by a t-test. War-af-
fected and war non-affected areas differed signifi-
cantly in five items: Role-Physical, Social Functio-
Babiæ-Banaszak et al: Impact of War on Quality of Life in Croatia Croat Med J 2002;43:396-402
397
Table 1. Respondents included in the health survey by coun-
ties in war-affected or non-affected areas of Croatia,
1997-1999
County No. (%) of respondents
War-affected:
Šibensko-kninska 220 (16.9)
Vukovarsko-srijemska 180 (13.9)
Osjeèko-baranjska 202 (15.6)
subtotal 602 (46.4)
War non-affected:
Zagrebaèka 220 (16.9)
Koprivnièko-kri´evaèka 216 (16.7)
Primorsko-goranska 259 (20.0)
subtotal 695 (53.6)
Total 1,297 (100.0)
Table 2. Sociodemographic characteristics of the sample of
respondents in Croatia, 1997-1999 (N=1,297)
No. (%) of respondents in area
Parameter war-affected non-affected total
Sex:
men 232 (17.9) 255 (19.7) 487 (37.6)
women 370 (28.5) 440 (33.9) 810 (62.4)
Age (years):
18-24 49 (3.8) 71 (5.5) 120 (9.3)
25-44 186 (14.3) 174 (13.4) 360 (27.7)
45-64 209 (16.1) 244 (18.8) 453 (34.9)
³65 158 (12.2) 206 (15.9) 364 (28.1)
Years of education:
£8 243 (18.7) 251 (19.4) 494 (38.1)
11-12 277 (21.4) 341 (26.3) 618 (47.7)
³14 82 (6.3) 103 (7.9) 185 (14.2)
Income:
low 205 (15.8) 174 (13.4) 379 (29.2)
medium 254 (19.6) 305 (23.5) 559 (43.1)
high 143 (11.0) 216 (16.7) 359 (27.7)
Total 602 (46.4) 695 (53.6) 1,297 (100.0)
Table 3. Description of subscales of Medical Outcome Study
36-item short-form used to assess the health of Croatian pop-
ulation, 1997-1999
Subscale Description
Physical functioning Ability to perform vigorous and moderate
activities, lift or carry groceries, climb the
stairs, bend, kneel, or stoop, walk, bathe or
dress
Role-physical Cutting down the amount of time spent on
activities, accomplishing less, feeling limited in
physical activities due to physical difficulties
Bodily pain Intensity of bodily pain and interference of
pain with usual activities
General health Self-perceived health, perception of getting ill
more easily than others, expectance of health
worsening, feeling healthy as others
Vitality Feeling full of energy, or worn out and tired
Social functioning Extent and frequency of interference of
physical health and emotional problems with
social activities
Role-emotional Cutting down the amount of time spent on
activities or accomplishing less due to
emotional problems
Mental health Feeling nervous, downhearted and sad or
calm, peaceful and happy
Health transition Health at the present moment compared to
health a year ago
ning, Role-Emotional, Mental Health, and Health
Transition, with scores being higher in the area not af-
fected by war (Table 4).
Sex Differences
Mean SF-36 subscale scores for men ranged from
41.70 for Health Transition to 70.50 for Social Func-
tioning in war-affected area, and from 45.29 for
Health Transition to 76.21 for Role-Emotional in
non-affected area. Mean SF-36 subscale scores for
women ranged from 41.01 for Health Transition to
66.91 for Social Functioning in war-affected area, and
from 45.06 for Health Transition to 72.89 for Role-
Emotional in war non-affected area.
Significant differences in scores between men in
war-affected and non-affected areas were found for
Role-Physical, Role-Emotional, and Mental Health
subscales. The scores were higher in war non-affected
area. Significant differences in scores between
women in war-affected and non-affected areas were
found for Role-Physical, Role-Emotional, and Health
Transition subscales. The scores were higher for
women in area not affected by war (Table 5).
Age Differences
Mean SF-36 subscale scores for respondents who
were 18 to 24 years old ranged from 52.04 for Health
Transition to 93.90 for Role-Emotional in war-affec-
ted area, and from 56.69 for Health Transition to
92.25 for Physical Functioning in non-affected area.
Mean SF-36 subscale scores for respondents between
25 and 44 years of age ranged from 45.56 for Health
Transition to 76.37 for Physical Functioning in war-af-
fected area, and from 51.44 for Health Transition to
81.03 for Role-Emotional subscale in war non-af-
fected area. Mean subscale scores for respondents be-
tween 45 and 64 years of age ranged from 38.04 for
Health Transition to 66.81 for Social Functioning in
war-affected area, and from 45.90 for Health Transi-
tion to 73.80 for Role-Emotional in non-affected area.
Mean subscale scores for respondents older than 65
ranged from 37.18 for Health Transition to 60.64 for
Social Functioning in war-affected area, and from
34.95 for Health Transition to 63.58 for Role-Emo-
tional in non-affected area.
In the 18-24 year-old age group, significant dif-
ferences between respondents in war-affected and
non-affected areas were found in Bodily Pain and Vi-
tality subscale scores. Respondents in war-affected
area had lower scores in Bodily Pain, but higher in Vi-
tality. Respondents from war-affected and non-af-
fected areas, who were between 25 and 44 years of
age, showed significant differences in Role-Physical,
Bodily Pain, General Health, Social Functioning,
Role-Emotional, Mental Health, and Health Transi-
tion subscale scores. The scores for all seven scales
were lower in war-affected area. In the 45-64 year-old
age group, significant differences were found be-
Babiæ-Banaszak et al: Impact of War on Quality of Life in Croatia Croat Med J 2002;43:396-402
398
Table 5. Health status of men and women in war-affected and non-affected areas of Croatia, according to their mean scores on
Medical Outcome Study 36-item short-form subscales, 1997-1999
Men Women
mean score (Q1-Q3) in areaamean score (Q1-Q3) in areaa
Subscale war-affected non-affected t p war-affected non-affected t p
Physical functioning 67.00
(45.00-95.00) 68.20
(40.00-100.00) 0.426 0.670 62.46
(35.00-95.00) 63.69
(40.00-95.00) 0.554 0.580
Role-physical 52.02
(0.00-100.00) 65.98
(25.00-100.00) 3.281 0.001 52.50
(0.00-100.00) 59.71
(0.00-100.00) 2.327 0.020
Bodily pain 61.19
(33.00-100.00) 65.56
(44.00-100.00) 1.580 0.115 58.19
(33.00-78.00) 59.61
(33.00-89.00) 0.665 0.506
General health 49.03
(30.00-65.00) 51.14
(35.00-65.00) 0.996 0.320 49.01
(35.00-65.00) 50.05
(35.00-65.00) 0.668 0.504
Vitality 51.72
(35.00-70.00) 51.00
(40.00-65.00) -0.358 0.721 48.14
(30.00-65.00) 48.48
(35.00-65.00) 0.218 0.827
Social functioning 70.50
(50.00-100.00) 73.26
(50.00-100.00) 1.172 0.242 66.91
(50.00-88.00) 70.34
(50.00-88.00) 1.813 0.070
Role-emotional 64.60
(0.00-100.00) 76.21
(67.00-100.00) 3.109 0.002 61.99
(0.00-100.00) 72.89
(33.00-100.00) 3.589 <0.001
Mental health 58.59
(44.00-76.00) 62.15
(52.00-76.00) 1.972 0.049 57.56
(44.00-72.00) 59.27
(48.00-72.00) 1.163 0.245
Health transition 41.70
(25.00-50.00) 45.29
(25.00-50.00) 1.822 0.069 41.01
(25.00-50.00) 45.06
(25.00-50.00) 2.338 0.020
aScore range: 0-100. The interquartile range (Q1-Q3) is a measure of dispersion. It is the difference between the 75th percentile (Q3) and the 25th percentile (Q1).
Table 4. Health status of Croatian population in war-affected and non-affected areas, according to their mean scores on Medical
Outcome Study 36-item short-form subscales, 1997-1999
Mean score (Q1-Q3) in areaa
Scale war-affected non-affected t p
Physical functioning 64.21 (40.00-95.00) 65.35 (40.00-95.00) 0.649 0.517
Role-physical 52.70 (0.00-100.00) 62.01 (0.00-100.00) 3.810 <0.001
Bodily pain 59.35 (33.00-89.00) 61.79 (44.00-89.00) 1.447 0.148
General health 49.02 (30.00-65.00) 50.45 (35.00-65.00) 1.138 0.255
Vitality 49.52 (35.00-65.00) 49.40 (35.00-65.00) -0.093 0.926
Social functioning 68.29 (50.00-100.00) 71.41 (50.00-100.00) 2.111 0.035
Role-emotional 63.02 (0.00-100.00) 74.11 (33.00-100.00) 4.711 <0.001
Mental health 57.95 (44.00-76.00) 60.33 (48.00-72.00) 2.080 0.038
Health transition 41.28 (25.00-50.00) 45.14 (25.00-50.00) 2.955 0.003
aScore range: 0-100. The interquartile range (Q1-Q3) is a measure of dispersion. It is the difference between the 75th percentile (Q3) and the 25th percentile (Q1).
tween respondents from war-affected and non-affe-
cted areas in Role-Physical, General Health, Role-
Emotional, Mental Health, and Health Transition sub-
scale scores. The scores were lower in war-affected
area. There were no significant differences in scores
between respondents 65 years of age and older in
war-affected and non-affected areas (Table 6).
Educational Differences
Mean SF-36 subscale scores for respondents with
complete or incomplete elementary education in
war-affected area ranged from 37.55 for Health Tran-
sition to 63.78 for Social Functioning, and in non-af-
fected area from 39.44 for Health Transition to 66.13
for Role-Emotional. Respondents with vocational/tra-
de or high school education achieved scores that
ranged from 43.59 for Health Transition to 72.38 for
Social Functioning in war-affected area, and from
49.05 for Health Transition to 80.36 for Role-Emo-
tional in non-affected area. Mean SF-36 subscale
scores for respondents with 2-year college or univer-
sity education ranged from 44.51 for Health Transi-
tion to 74.09 for Physical Functioning in war-affected
area, and from 46.12 for Health Transition to 73.83
for Social Functioning in non-affected area.
Respondents with primary school education in
war non-affected area achieved significantly higher
score in Role-Emotional than those in war-affected
area. Significant differences between respondents with
vocational/trade or high school education in war-af-
fected and non-affected areas were found in Role-Physi-
cal, Bodily Pain, Role-Emotional, and Health Transition
Babiæ-Banaszak et al: Impact of War on Quality of Life in Croatia Croat Med J 2002;43:396-402
399
Table 6. Health status of Croatian population in war-affected and non-affected areas by age, according to their mean scores on
Medical Outcome Study 36-item short-form subscales, 1997-1999
Age groups (years)
18-24 25-44 45-64 ³65
mean score
(Q1-Q3) in areaamean score
(Q1-Q3) in areaamean score
(Q1-Q3) in areaamean score
(Q1-Q3) in areaa
Subscale war-
affected non-
affected t p war-
affected non-
affected t p war-
affected non-
affected t p war-
affected non-
affected t p
Physical
functioning 87.65
(95.00-
100.00)
92.25
(95.00-
100.00)
1.091 0.279 76.37
(65.00-
100.00)
80.32
(70.00-
100.00)
1.347 0.179 59.81
(35.00-
85.00)
64.04
(40.00-
90.00)
1.547 0.123 48.45
(30.00-
70.00)
44.98
(20.00-
70.00)
-1.153 0.250
Role-
physical 84.18
(75.00-
100.00)
86.97
(100.00-
100.00)
0.511 0.610 65.05
(25.00-
100.00)
78.74
(75.00-
100.00)
3.322 0.001 43.18
(0.00-
100.00)
57.27
(0.00-
100.00)
3.388 0.001 40.98
(0.00-
100.00)
44.90
(0.00-
100.00)
0.860 0.390
Bodily
pain 79.67
(67.00-
100.00)
87.86
(78.00-
100.00)
2.041 0.045 66.53
(44.00-
100.00)
75.31
(56.00-
100.00)
3.062 0.002 54.30
(33.00-
78.00)
57.63
(44.00-
78.00)
1.205 0.229 51.27
(27.00-
78.00)
46.32
(22.00-
67.00)
-1.607 0.109
General health 73.67
(65.00-
85.00)
70.07
(60.00-
80.00)
-1.134 0.259 57.37
(40.00-
75.00)
61.72
(55.00-
75.00)
2.082 0.038 43.13
(25.00-
60.00)
47.19
(35.00-
60.00)
2.048 0.041 39.34
(25.00-
55.00)
38.01
(25.00-
50.00)
-0.666 0.506
Vitality 74.18
(60.00-
90.00)
63.60
(55.00-
75.00)
-3.130 0.002 52.71
(40.00-
65.00)
56.09
(45.00-
70.00)
1.625 0.105 46.67
(30.00-
60.00)
48.38
(35.00-
60.00)
0.858 0.392 41.87
(25.00-
55.00)
40.05
(25.00-
55.00)
-0.773 0.440
Social
functioning 87.3
(75.00-
100.00)
83.62
(75.00-
100.00)
-1.116 0.267 71.44
(50.00-
100.00)
79.15
(63.00-
100.00)
2.914 0.004 66.81
(50.00-
88.00)
70.07
(50.00-
88.00)
1.347 0.179 60.64
(38.00-
88.00)
62.27
(50.00-
88.00)
0.565 0.572
Role-
emotional 93.90
(100.00-
100.00)
88.75
(100.00-
100.00)
-1.170 0.244 65.24
(0.00-
100.00)
81.03
(67.00-
100.00)
3.818 <0.001 59.18
(0.00-
100.00)
73.80
(67.00-
100.00)
3.655 <0.001 55.91
(0.00-
100.00)
63.58
(0.00-
100.00)
1.592 0.112
Mental health 74.37
(64.00-
84.00)
70.48
(64.00-
80.00)
-1.303 0.195 60.28
(48.00-
76.00)
65.33
(56.00-
76.00)
2.595 0.010 55.69
(40.00-
72.00)
60.43
(48.00-
76.00)
2.454 0.015 53.11
(40.00-
68.00)
52.49
(40.00-
68.00)
-0.284 0.776
Health
transition 52.04
(50.00-
50.00)
56.69
(50.00-
75.00)
1.409 0.161 45.56
(25.00-
50.00)
51.44
(50.00-
50.00)
2.534 0.012 38.04
(25.00-
50.00)
45.90
(25.00-
50.00)
3.677 <0.001 37.18
(25.00-
50.00)
34.95
(25.00-
50.00)
-0.855 0.393
aScore range: 0-100. The interquartile range (Q1-Q3) is a measure of dispersion. It is the difference between the 75th percentile (Q3) and the 25th percentile (Q1).
Table 7. Health status of Croatian population in war-affected and non-affected areas by years of education, according to their
mean scores on Medical Outcome Study 36-item short-form subscales, 1997-1999
Years of education
£8 11-12 ³14
mean score (Q1-Q3) in areaamean score (Q1-Q3) in areaamean score (Q1-Q3) in areaa
Subscale war-affected non-affected t p war-affected non-affected t p war-affected non-affected t p
Physical
functioning 54.88
(30.00-80.00) 53.21
(25.00-80.00) -0.613 0.540 69.48
(45.00-100.00) 71.74
(50.00-100.00) 0.927 0.354 74.09
(50.00-100.00) 73.74
(60.00-100.00) -0.077 0.938
Role-physical 43.52
(0.00-100.00) 49.30
(0.00-100.00) 1.453 0.147 56.77
(0.00-100.00) 69.57
(25.00-100.00) 3.700 <0.001 66.16
(25.00-100.00) 67.96
(25.00-100.00) 0.297 0.767
Bodily pain 53.36
(33.00-78.00) 51.04
(22.00-78.00) -0.848 0.397 62.44
(44.00-100.00) 68.26
(44.00-100.00) 2.434 0.015 66.67
(44.00-89.00) 66.58
(44.00-89.00) -0.022 0.982
General health 41.83
(25.00-55.00) 42.71
(25.00-55.00) 0.465 0.645 53.09
(35.00-70.00) 54.49
(40.00-70.00) 0.768 0.443 56.59
(40.00-75.00) 55.92
(40.00-70.00) -0.214 0.831
Vitality 42.86
(25.00-60.00) 41.93
(25.00-55.00) -0.477 0.634 53.77
(40.00-70.00) 53.84
(40.00-70.00) 0.039 0.969 54.88
(40.00-70.00) 52.91
(40.00-65.00) -0.609 0.543
Social functioning 63.78
(38.00-88.00) 64.35
(50.00-88.00) 0.237 0.813 72.38
(50.00-100.00) 75.88
(63.00-100.00) 1.705 0.089 67.88
(50.00-100.00) 73.83
(50.00-100.00) 1.562 0.120
Role-emotional 52.81
(0.00-100.00) 66.13
(0.00-100.00) 3.266 0.001 70.41
(33.00-100.00) 80.36
(67.00-100.00) 3.149 0.002 68.30
(33.00-100.00) 72.83
(33.00-100.00) 0.768 0.443
Mental health 52.41
(36.00-68.00) 54.34
(44.00-68.00) 1.026 0.305 61.33
(48.00-76.00) 63.64
(52.00-76.00) 1.443 0.149 62.98
(52.00-80.00) 63.96
(52.00-76.00) 0.348 0.728
Health transition 37.55
(25.00-50.00) 39.44
(25.00-50.00) 0.819 0.413 43.59
(25.00-50.00) 49.05
(50.00-50.00) 3.136 0.002 44.51
(25.00-50.00) 46.12
(25.00-50.00) 0.993 0.623
aScore range: 0-100. The interquartile range (Q1-Q3) is a measure of dispersion. It is the difference between the 75th percentile (Q3) and the 25th percentile (Q1).
subscale scores. The scores were higher in war non-af-
fected area. There were no significant differences in
mean subscale scores between respondents with
2-year college or university education in war-affected
and non-affected areas (Table 7).
Income Differences
Mean scores for respondents in low-income
group ranged from 41.10 for Health Transition to
68.67 for Social Functioning subscale in war-affected
area, and from 50.14 for Health Transition to 77.41
for Role-Emotional subscale in war non-affected area.
Mean scores for respondents in medium-income
group ranged from 40.45 for Health Transition to
68.91 for Social Functioning subscale in war-affected
area and from 43.93 for Health Transition to 71.25 for
Role-Emotional subscale in war non-affected area.
High-income group’s scores ranged from 43.01 for
Health Transition to 66.65 for Social Functioning
subscale in war-affected area, and from 42.82 for
Health Transition to 75.48 for Role-Emotional subsca-
le in non-affected area.
Significant differences between respondents in
war-affected and non-affected areas in low-income
group were found for Role-Physical, Bodily Pain, Vi-
tality, Social Functioning, Role-Emotional, Mental
Health, and Health Transition subscales. Scores were
lower for all seven scales in war-affected area. Re-
spondents in medium-income group in war-affected
area had significantly lower scores in Role-Physical
and Role-Emotional subscales, and respondents in
high-income group in war-affected area had signifi-
cantly lower score in Role-Emotional subscale (Table 8).
Discussion
Respondents living in areas affected by war gen-
erally achieved lower scores for the following five
subscales: Role-Physical, Social Functioning, Role-
Emotional, Mental Health, and Health Transition.
Wherever there were differences between the scale
scores, they pointed at the same direction – scores
were lower in area directly affected by war. Role-
Emotional and Role-Physical subscales almost always
showed the difference between the two areas,
whether the differences were observed regarding sex,
age, education or income. Health Transition and
Mental Health subscales also commonly revealed the
difference between the areas. The subscale that did
not differ between the two areas was Physical Func-
tioning.
Physical Functioning is a relatively “objective”
indicator of physical health because it describes phys-
ical ability. Lack of difference in Physical Functioning
between the two areas could be interpreted as a non-
existence of real differences in physical health in
those areas. On the other hand, Role-Physical de-
scribes limitations and difficulties in performing every
day duties as well as cutting down the amount of time
spent on activities. Having in mind the fact that there
were no differences in Physical Functioning, this
could be attributable to self-perceived inability to
function in everyday activities. Role-Emotional is a
subscale that is in a way an analogue of Role-Physi-
cal. It comprises cutting down the amount of time
spent on activities, inability to work carefully, and ac-
complishing less (25). It is related to feeling incapable
for daily activities. Health Transition is a subscale that
describes health at the present moment compared
with health a year ago (25). It shows the dynamics of
change in self-perceived health. In this study, the
change in negative direction was more prominent in
war-affected area. Mental Health subscale describes
feelings of happiness, calmness, and peacefulness or
feelings of downheartedness and nervousness. This
scale measures depression and feeling of anxiety. The
research work on psychiatric illness and mental disor-
ders by Kajeviæ and Klein (28) showed an increase
and worsening of psychiatric morbidity due to the
war stress, but the study was oriented towards the
specific group of exposed population. Results of our
study showed that mental health is worse in the
whole population directly affected by war.
The finding that draws attention is that in the
18-24 age group, respondents from war-affected and
non-affected areas differed in Bodily Pain and Vital-
ity. The score for Bodily Pain was lower in war-af-
Babiæ-Banaszak et al: Impact of War on Quality of Life in Croatia Croat Med J 2002;43:396-402
400
Table 8. Health status of Croatian population in war-affected and non-affected areas by their income, according to their mean
scores on Medical Outcome Study 36-item short-form subscales, 1997-1999
Income
low medium high
mean score (Q1-Q3) in areaamean score (Q1-Q3) in areaamean score (Q1-Q3) in areaa
Subscale war-affected non-affected t p war-affected non-affected t p war-affected non-affected t p
Physical functioning 65.54
(40.00-95.00) 70.20
(50.00-100.00) 1.783 0.075 64.11
(40.00-95.00) 63.10
(35.00-95.00) -0.370 0.712 63.92
(45.00-90.00) 64.61
(45.00-95.00) 0.211 0.833
Role-physical 57.56
(0.00-100.00) 71.84
(50.00-100.00) 3.309 0.001 50.49
(0.00-100.00) 58.61
(0.00-100.00) 2.153 0.032 49.65
(0.00-100.00) 58.91
(0.00-100.00) 1.967 0.050
Bodily pain 60.39
(33.00-89.00) 69.50
(44.00-100.00) 2.996 0.003 59.49
(33.00-89.00) 57.15
(33.00-78.00) -0.890 0.374 57.62
(38.50-83.50) 62.13
(44.00-89.00) 1.420 0.156
General health 50.88
(35.00-65.00) 54.48
(40.00-70.00) 1.592 0.112 48.96
(30.00-65.00) 49.18
(30.00-65.00) 0.113 0.910 46.47
(30.00-65.00) 48.98
(35.00-65.00) 1.092 0.275
Vitality 49.71
(35.00-65.00) 54.17
(40.00-70.00) 1.993 0.047 50.69
(35.00-70.00) 47.57
(35.00-60.00) -1.597 0.111 47.17
(30.00-60.00) 48.15
(35.00-60.00) 0.423 0.673
Social functioning 68.67
(50.00-88.00) 76.63
(63.00-100.00) 3.146 0.002 68.91
(50.00-100.00) 69.99
(50.00-100.00) 0.471 0.638 66.65
(50.00-88.00) 69.23
(50.00-88.00) 0.882 0.378
Role-emotional 60.96
(0.00-100.00) 77.41
(67.00-100.00) 3.911 <0.001 63.14
(0.00-100.00) 71.25
(33.00-100.00) 2.192 0.029 65.76
(0.00-100.00) 75.48
(67.00-100.00) 2.177 0.030
Mental health 56.72
(44.00-72.00) 64.44
(52.00-80.00) 3.744 <0.001 59.07
(44.00-76.00) 59.15
(48.00-72.00) 0.043 0.966 57.73
(46.00-72.00) 58.69
(48.00-72.00) 0.435 0.663
Health transition 41.10
(25.00-50.00) 50.14
(50.00-50.00) 3.752 <0.001 40.45
(25.00-50.00) 43.93
(25.00-50.00) 1.804 0.072 43.01
(25.00-50.00) 42.82
(25.00-50.00) -0.069 0.945
aScore range: 0-100. The interquartile range (Q1-Q3) is a measure of It is the difference between the 75th percentile (Q3) and the 25th percentile (Q1).
fected area, but the score for Vitality was higher in
war-affected area, which is inconsistent with the other
findings from this study.
Respondents older than 65 years of age did not
differ in the quality of life parameters. However, com-
parison of our results with the results from the recent
study from eight health districts in six European coun-
tries showed that the scores in our study were lower
(29). Heslin et al (29) compared Mental Health, Physi-
cal Functioning, and General Health of elderly peo-
ple in Sweden (Jämtland), Finland (Porvoo), the
United Kingdom (North Staffordshire, Dudley, and
Morgannwg), Ireland (South East), Spain (Andalusia),
and Croatia (Istria). Our results were lower in older re-
spondents for all three scales.
When analyzing the respondents according to
their educational level, it became clear that there
were no differences in health-related quality of life in
the two areas between the respondents with univer-
sity education. Those with elementary school educa-
tion differed in only one scale – Role-Emotional. The
mostly affected were those with secondary and voca-
tional school education – they scored lower for Role-
Physical, Role-Emotional, Bodily Pain, and Health
Transition.
Income was another parameter that made a dif-
ference between respondents in the two observed ar-
eas – the lower the income, the larger the difference
between the two areas. Respondents in the low-in-
come group in the area affected by war had lower
scores than respondents in area not affected by war in
as many as seven subscales: Role-Physical, Bodily
Pain, Vitality, Social Functioning, Role-Emotional,
Mental Health, and Health Transition. Respondents
with high income differed only in Role-Emotional
subscale scores.
The scales that marked the difference in health-
related quality of life between the two areas – Role-
Physical, Social Functioning, Role-Emotional, Mental
Health, and Health Transition, are quality of life indi-
cators that are susceptible to external influences and,
as such, could indicate inadequate living environ-
ment due to war consequences. Younger respon-
dents, respondents with secondary and vocational
school education, and those with lower income rep-
resented the groups that were mostly affected by war.
This study is only a description of health-related
quality of life and did not take morbidity into account.
It was not our intention to focus on morbidity; we
merely wanted to present summary quality of life in
two areas of the country affected by war. We did not
collect information on migrations and direct war in-
volvement of respondents, although these additional
data could shed more light on causes of differences
presented in this study. Having in mind these limita-
tions, we feel that this research nevertheless shows
the affect of war on population health in general.
Findings from our study lead to a conclusion that war
affects self-perceived health and physical ability as
well as emotional and mental health of entire popula-
tion striken by war. Younger persons with lower in-
come and vocational or secondary school education
are identified as the most vulnerable groups of popu-
lation. This information gives a small contribution to a
large body of evidence on effect of war on different
population groups. It could be useful in public health
efforts to recognize specific needs of population
groups in order to reduce morbidity and improve
population quality of life.
Acknowledgment
We thank Professor Silvije Vuletiæ for his advice and com-
ments. This study was a part of the project “Analysis of Transition
of Healthcare System in Croatia” supported by the Ministry of Sci-
ence and Technology of the Republic of Croatia (No. 108035).
References
1 Unwin C, Blatchley N, Coker W, Ferry S, Hotopf M,
Hull L, et al. Health of UK servicemen who served in
Persian Gulf War. Lancet 1999;353:169-78.
2 Gray GC, Coate BD, Anderson CM, Kang HK, Berg SW,
Wignall FS, et al. The postwar hospitalization experi-
ence of U.S. veterans of the Persian Gulf War. N Engl J
Med 1996;335:1505-13.
3 Fukuda K, Nisenbaum R, Stewart G, Thompson WW,
Robin L, Washko RM, et al. Chronic multisymptom ill-
ness affecting Air Force veterans of the Gulf War. JAMA
1998;280:981-8.
4 The Iowa Persian Gulf Study Group. Self-reported ill-
ness and health status among Gulf War veterans. A pop-
ulation-based study. JAMA 1997;277:238-45.
5Penaviæ S. Case-mix analysis in veterans’ care institu-
tions in Croatia. Croat Med J 1995;36:134-7.
6 Health status of Vietnam veterans II: physical health.
Centers for Disease Control Vietnam Experience Study.
JAMA 1988;259:2708-14.
7 O’Toole B, Marshall R, Grayson D, Schureck RJ,
Dobson M, Ffrench M, et al. The Australian Vietnam
veterans health study, II: self-reported health of veterans
compared with the Australian population. Int J
Epidemiol 1996;25:319-30.
8 Orr SP, Pitman RK, Lasko NB, Herz LR. Psychophysio-
logical assessment of post-traumatic stress disorder im-
agery in World War II and Korean combat veterans. J
Abnorm Psychol 1993;102:152-9.
9 Pitman RK, Orr SP, Forgue DF, Altman B, de Jong JB,
Herz LR. Psychophysiological responses to combat im-
agery of Vietnam veterans with post-traumatic stress
disorder versus other anxiety disorders. J Abnorm
Psychol 1990;99:49-54.
10 Roemer L, Litz BT, Orsillo SM, Ehlich PJ, Friedman MJ.
Increases in retrospective accounts of war-zone expo-
sure over time: the role of PTSD symptoms severity. J
Trauma Stress 1998;11:597-605.
11 Kozariæ-Kovaèiæ D, Kocijan-Hercigonja D, Jambrošiæ A.
Psychiatric help to psychotraumatized persons during
and after war in Croatia. Croat Med J 2002;43:221-8.
12 Fontana A, Rosenheck R. A causal model of the etiology
of war-related PTSD. J Trauma Stress 1993;6:475-500.
13 Toole MJ, Waldman RJ. The public health aspects of
complex emergencies and refugees situations. Annu
Rev Public Health 1997;18:283-312.
14 Richter D. Croatian experience on the care for the dis-
placed and refugee children. Croat Med J 1994;35:8-11.
15 Tocilj Šimunkoviæ G, Urliæ I. War trauma: emotional re-
sponses and psychological defenses of displaced per-
sons. Croat Med J 1995;36:253-61.
Babiæ-Banaszak et al: Impact of War on Quality of Life in Croatia Croat Med J 2002;43:396-402
401
16 Kujund´iæ M, Kern J, Ivankoviæ D, Budak A, Dragun D,
Vuletiæ S. Person displacement pattern in Croatia. Croat
Med J 1992;33 War suppl 1:55-60.
17 Havelka M, Despot Luèanin J, Luèanin D. Psychologi-
cal reactions to war stressors among elderly displaced
persons in Croatia. Croat Med J 1995;36:262-5.
18 Mariæ Lj, Car T, Betica-Radiæ Lj, Radoniæ M. Mental
health status, psychosocial symptoms and migration
tendencies of displaced persons in Dubrovnik during
1991/92 siege. Croat Med J 1994;35:100-4.
19 Government of the Republic of Croatia. War damage in
the Republic of Croatia. Croatian Government Bulletin
1999. Available from: http://www.vlada.hr/bulletin/19
99/sep-oct/documents.html. Accessed: March 5, 2002.
20 ªiviæ D. Izravni demografski gubitci (ratne ´rtve)
Hrvatske (1990.-1998.) uzrokovani velikosrpskom agre-
sijom i neke njihove posljedice. Društvena istra´ivanja
2001;10:451-84.
21 Šterc S, Pokos N. Demografski uzroci i posljedice rata
protiv Hrvatske. Društvena istra´ivanja 1993;2:305-33.
22 ªiviæ D. Promjene u dinamici i razmještaju progna-
nièko-izbjeglièkog kontingenta u Republici Hrvatskoj
od sredine 1991. do sredine 1998. godine. Druš istra´
1999;8:767-91.
23 World Health Organization. Terminology of the World
Health Report 2000: health systems: improving perfor-
mance. Available from: www.who.int/terminology/ter/
index.html. Accessed: May 23, 2002.
24 Shah S, Sesti AM, Chopra T, McLaughlin-Miley C, Cop-
ley-Merriman K. Quality of Life terminology docu-
mented in package inserts. Quality of Life Newsletter
2001;(27):1-3.
25 Ware JE Jr, Gandek B. Overview of the SF-36 health sur-
vey and the International Quality of Life Assessment
(IQOLA) Project. J Clin Epidemiol 1998;51:903-12.
26 Buttanshaw C, editor. Tipping the balance towards
primary healthcare network. Proceedings of the 10th
anniversary conference of the Tipping the Balance to-
wards Primary Healthcare Network; 1997 Nov 13-16;
Göteborg, Sweden. Göteborg: The Nordic School of
Public Health; 1999.
27 Jureša V, Ivankoviæ D, Vuletiæ G, Babiæ-Banaszak A,
Srèek I, Mastilica M, et al. The Croatian Health Survey –
SF-36: I. General Quality of Life Assessment. Coll
Antropol 2000;24:69-78.
28 Kajeviæ M, Klain E. Psychological reactions in the large
group of medical personnel in a psychiatric hospital
near the war zone. Croat Med J 1995;36:120-5.
29 Heslin JM, Soveri PJ, Winoy JB, Lyons RA, Buttanshaw
AC, Kovaèiæ L, et al. Health status and service utiliza-
tion of older people in different European countries.
Scand J Prim Health Care 2001;19:218-22.
Received: March 27, 2002
Accepted: May 29, 2002
Correspondence to:
Andreja Babiæ-Banaszak
Andrija Štampar School of Public Health
Zagreb University School of Medicine
Rockefellerova 4
10000 Zagreb, Croatia
abana@andrija.snz.hr
Babiæ-Banaszak et al: Impact of War on Quality of Life in Croatia Croat Med J 2002;43:396-402
402
