P1: GXB

Journal of Psychopathology and Behavioral Assessment (JOBA) PP1052-JOBA-476509 November 22, 2003 15:35 Style file version June 25th, 2002

Journal of Psychopathology and Behavioral Assessment, Vol. 26, No. 1, March 2004 (°
C 2004)

Clinical Validation of the Swedish Version of the Quality of

Life Inventory in Crime Victims With Posttraumatic Stress
Disorder and a Nonclinical Sample

Nenad Paunović1,2 and Lars-Göran Öst1

Accepted June 24, 2003

The Swedish version of the Quality of Life Inventory (QOLI) was cross-culturally validated with a
crime victim sample (N = 53) with posttraumatic stress disorder (PTSD) and a nonclinical group
(N = 100) with no lifetime and current psychiatric disorder from the general population in the
Stockholm county of Sweden. The QOLI showed excellent internal consistencies in the clinical and
the nonclinical sample. In addition, PTSD participants displayed a significantly lower self-perceived
quality of life in 13 out of 16 domains of life than the nonclinical individuals. Furthermore, the QOLI
was inversely correlated with interview and self-report measures of PTSD symptoms, depression, and
anxiety. Overall, the results support the utility of the Swedish QOLI version as a potentially useful
measure of self-perceived quality of life in Swedish clinical and nonclinical samples. The results are
discussed in relation to the Swedish cross-cultural validation of the QOLI, methodological limitations
and future directions.

KEY WORDS: PTSD; crime victims; nonclinical sample; cross-cultural validation; quality of life inventory.

INTRODUCTION civilian-related traumatic events that have been associ-

ated with the highest current and lifetime prevalence rates
Posttraumatic stress disorder (PTSD; American Psy- of PTSD in representative samples (Breslau et al., 1998;
chiatric Association, 1994) has been associated with sev- Kilpatrick & Resnick, 1993; Resnick, Kilpatrick, Dansky,
eral indices of deteriorated mental health functioning. Saunders, & Best, 1993). Resnick et al. (1993) found that
Zatzick et al. (1997) found that PTSD was related to im- 32% of rape victims and 39% of physical assault victims
pairments in role functioning, subjective well-being, self- fulfilled criteria for PTSD at one point in their lifetime,
reported physical health status, current physical function- and that the current prevalence rates were 12 and 18%
ing, and perpetration of violent interpersonal acts in the respectively.
past year. Also, individuals with PTSD are much more Assessment measures of psychiatric disorders usu-
likely to report marital, parental, and family adjustment ally focus on negative symptoms and distress. But from
problems than those without PTSD (Jordan et al., 1992). a client’s perspective such measures do not capture posi-
The subjective distress that is generated by PTSD symp- tive mental health functioning. One aspect of the latter is
toms and the associated functional impairment suggests how an individual evaluates his or her own life satisfac-
that this patient group may be characterized by a deterio- tion or quality of life. Even though negative and positive
rated self-perceived quality of life. Furthermore, the high affects are inversely related to each other they also repre-
incidence of PTSD may lead to a deteriorated quality of sent independent aspects of peoples’ mental health status
life for many people. Assaultive violence and rape are the (Frisch, Cornell, Villanueva, & Retzlaff, 1992). In psychi-
atrically disordered individuals self-perceived quality of
1 Department of Psychology, Stockholm University, Stockholm, Sweden.
life may tap the clinical significance of negative symptoms
2 To whom correspondence should be addressed at Traumacenter
and affects.
(house 27), Danderyds hospital, 182 88 Danderyd, Sweden; e-mail: There are few validated measures of quality of life
sonepaunovic@hotmail.com. that cover both a broad range of life domains and that

15
0882-2689/04/0300-0015/0 °
C 2004 Plenum Publishing Corporation
P1: GXB
Journal of Psychopathology and Behavioral Assessment (JOBA) PP1052-JOBA-476509 November 22, 2003 15:35 Style file version June 25th, 2002

16 Paunović and Öst

are applicable to clinical and nondisordered groups alike edge, there is as yet no published study that has cross-
(Mendlowicz & Stein, 2000). In addition, only the Qual- culturally validated the QOLI with clinical and nonclini-
ity of Life Inventory (QOLI; Frisch et al., 1992) includes cal samples in other countries. Validated measures in the
an assessment of the subjective importance of various life United States and elsewhere need to be cross-culturally
areas and their relative contribution to overall life satisfac- validated before they can be internationally applied in
tion. If a life area is deemed unimportant, then the degree various research and clinical contexts. Also, the concept
of quality of life in such a life domain should not have an of quality of life that QOLI taps is particularly relevant
impact upon the individual’s overall quality of life. The to the study of PTSD since it may indicate the clinical
QOLI is based on an individual’s subjective evaluation of significance of psychopathology symptoms. Because no
his or her quality of life which is defined as a person’s study has evaluated the psychometric properties of the
judgment of whether his or her goals, needs, and wishes QOLI in Sweden there was a great need for such an eval-
have been fulfilled. This instrument is based on a theoreti- uation to provide Swedish researchers with reliable and
cal model that assumes that (1) an individual’s overall life valid tools for the assessment of quality of life. Further-
satisfaction consists largely of the sum of satisfactions in more, because the QOLI covers a large variety of life do-
particular areas of life deemed important, (2) a person’s mains, has been validated in the United States, and in-
satisfaction with a particular area of life is weighted ac- cludes both importance and satisfaction ratings, it may be
cording to its importance or value before it enters into a particularly useful quality of life measure to validate
the “equation” of overall life satisfaction, and (3) a broad cross-culturally.
range of life areas of human aspiration and fulfillment The purpose of this study was to conduct a cross-
are identified that are applicable to both psychiatric and cultural validation study on the Swedish version of the
nonpsychiatric populations. The QOLI is the only measure QOLI with a group of crime victims with PTSD and a
of quality of life that fulfills all these three characteristics. group of nonclinical individuals’ with no lifetime and cur-
Furthermore, the QOLI is a convenient measure since it rent psychological disorder. First, the internal consistency
takes only 10 min to complete, is easily scored, and inter- was computed for each group separately. Second, to as-
pretable by laymen (Frisch et al., 1992). sess the discriminative validity of the QOLI the PTSD
The reliability and validity of the QOLI have been group was matched on gender and age with individuals
examined in two studies. Frisch et al. (1992) included sam- from the nonclinical group. These two groups were com-
ples of undergraduates, undergraduate counseling center pared on the 16 weighted life satisfaction items of the
patients, alcohol dependence or abuse patients, recovered QOLI. Third, the construct validity of the QOLI was in-
patients, and criminal offenders. Test–retest and inter- vestigated by exploring its relationships with interview
nal consistency reliabilities were high. The QOLI cor- and self-report measures of PTSD symptoms, depression,
related positively with various interview, self-report and and anxiety.
peer-rating measures of subjective well-being across all
samples. Also, the QOLI was negatively correlated with
psychometric measures of general psychopathology, de- METHOD
pression, and anxiety. Furthermore, the clinical samples
had significantly lower QOLI scores than the nonclinical Samples
groups. In another study the subjective quality of life was
significantly lower in social phobics than in nonclinical The Crime Victims With PTSD
samples from Frisch et al.’s study (Frisch et al., 1992;
Safren, Heimberg, Brown, & Holle, 1996–97). Also, the Crime victims were recruited by the first author from
QOLI was inversely correlated with various measures of police department records of violent crimes and psychi-
social phobia, functional impairment, and depression. To atric units in the Stockholm County of Sweden. Police
summarize, the QOLI has in general shown good psycho- department records of violent crimes were scanned 3–6
metric properties in empirical studies. Furthermore, since times a year with the permission from police authorities to
the QOLI is the only validated measure of self-perceived obtain the names and addresses of victims 1 month after
quality of life that includes both importance and satisfac- the traumatic event. These people were sent a letter de-
tion ratings of various life areas, it may be considered as a scribing a treatment outcome study of cognitive-behavior
first hand choice in evaluating the quality of life in various therapy (CBT) for PTSD that was conducted at the Depart-
psychiatric disorders. ment of Psychology, Stockholm University, Stockholm,
The QOLI has been validated with U.S. clinical and Sweden. Persons who felt that they were distressed be-
nonclinical samples. To the best of the authors’ knowl- cause of memories of the crime situation were invited to
P1: GXB
Journal of Psychopathology and Behavioral Assessment (JOBA) PP1052-JOBA-476509 November 22, 2003 15:35 Style file version June 25th, 2002

Swedish Cross-Cultural Validation of the QOLI 17

contact the research assistant (the first author) by phone The Nonclinical Sample
or letter. Potential participants for the CBT study were
first briefly screened through telephone calls for the pres- The nonclinical sample was recruited by two un-
ence of PTSD according to the Diagnostic and Statisti- dergraduate students at the Department of Psychology,
cal Manual of Mental Disorders-IV (DSM-IV; American Stockholm University, Stockholm, Sweden. One thou-
Psychiatric Association, 1994) criteria. Those who might sand individuals born between 1938 and 1980 were ran-
fulfil the DSM-IV criteria for PTSD and who wanted to domly selected from the whole population in the Stock-
participate were scheduled for a screening interview. The holm County of Sweden through an order to the Swedish
purpose was to assess PTSD and disorders with similar Government’s Person and Address Register (SPAR). A
psychopathology according to the DSM-IV criteria. One request for participation in an evaluation of psychiatric
hundred and thirty-eight crime victims were administered disorders in psychologically healthy individuals that had
the Clinician Administered PTSD Scale-IV (CAPS-IV; never suffered from psychological problems nor received
Blake et al., 1997) and parts of the Anxiety Disorders Inter- medication for such disturbances was sent out by mail to
view Schedule-IV (ADIS-IV; Brown, diNardo, & Barlow, 324 randomly selected individuals from all age categories
1994) that were relevant in determining the inclusion cri- of both genders that had a phone number. It was stated
teria. The CAPS-IV scale gives information for making that the evaluation consisted of an interview and filling
a DSM-IV PTSD diagnosis and ADIS-IV was used to out self-report questionnaires and that they would receive
assess other DSM-IV anxiety disorder diagnoses and to a fee of 250 Swedish kronor upon completion of the as-
differentiate between anxiety disorders and other disor- sessments. Four to seven days after the mail was sent out
ders with similar symptomatology. The test–retest reli- these individuals were contacted by phone and a screening
ability of the CAPS for diagnosing PTSD (DSM-III-R) interview with the “mini-ADIS-IV” (Brown et al., 1994)
was reported from 0.77 to 0.96 for the three symptom was scheduled with those who wanted to participate. The
clusters, and 0.90–0.98 for all 17 items (Blake et al., inclusion criteria were: age between 18 and 60 years; no
1995). Interrater reliabilities of the ADIS-IV in the assess- present or lifetime anxiety, affective or psychotic disor-
ment of anxiety disorders and other disorders with similar der; or substance or drugs abuse; speaking and writing the
symptomatology have in general been good or excellent Swedish language fluently; and no current psychological
(κ ≥ .60–.75; Brown, Di Nardo, Lehman, & Campbell, or pharmacological treatment for psychiatric disorders.
2001). One hundred individuals fulfilled the inclusion criteria.
Fifty-three crime victims that fulfilled the follow- They were administered the “mini-ADIS” and filled out
ing inclusion criteria served as clinical participants in the self-report questionnaires. The nonclinical participants re-
cross-cultural validation of the Swedish QOLI version: ceived a fee of 250 Swedish kronor after completing all
DSM-IV criteria for PTSD; having been a victim of vio- the assessments. There were 45 men and 55 women, and
lent crime 1–3 months prior to the screening interview; the mean age was 38.2 years (SD = 10.4).
18–60 years of age; a CAPS-IV global severity rating
of at least 2; no psychotic or organic mental disorder;
and no current drug or alcohol abuse. The PTSD partic- Matching
ipants had personally experienced the following types of
traumatic events: aggravated assault 15, physical assault The PTSD participants (N = 53) were matched on
13, completed rape 11, witnessed a murder 4, manslaugh- gender and age (±2 years) with nonclinical participants
ter/homicide attempt 3, other sexual assault 2, robbery and (n = 53) from the whole nonclinical sample (N = 100).
manslaughter/homicide attempt 2, robbery 2, and homi- The purpose was to compare the groups on the weighted
cide attempt 1. At the end of the screening interview crime life satisfaction items of the QOLI. Each group included
victims that fulfilled the criteria were provided with infor- 22 men (41%) and 31 women (59%). The mean age of the
mation about the research project. They read through the PTSD group was 35.9 years (SD = 9.4) and of the non-
informed consent form and decided if they wanted to par- clinical group 36.5 years (SD = 9.4). Twenty-two (41%)
ticipate. The victims who wanted to participate signed the of the PTSD participants were married or lived in a steady
form and were given self-report questionnaires to fill out relationship, 18 (34%) were single, and 13 (25%) were
until the second appointment, which was scheduled 1 week divorced. Twenty-nine (55%) of the nonclinical partici-
after the first screening interview. During this interview pants were married or lived in a steady relationship, 18
the rest of the ADIS-IV was completed. After the sec- (34%) were single, and 6 (11%) were divorced. The occu-
ond interview the participants completed the self-report pational status of the PTSD group was as follows: Twenty-
assessments. one (40%) worked full-time (including full-time students),
P1: GXB
Journal of Psychopathology and Behavioral Assessment (JOBA) PP1052-JOBA-476509 November 22, 2003 15:35 Style file version June 25th, 2002

18 Paunović and Öst

9 (17%) were part-time employees, and 23 (43%) were ety was assessed with the Beck Anxiety Inventory (BAI;
unemployed (including those on sick leave). The occu- Beck, Epstein, Brown, & Steer, 1988) that consists of 21
pational status of the nonclinical group was: 39 (74%) items rated on a 0–3 scale. State and trait anxiety was
worked full-time (including full-time students), 8 (15%) assessed with the State–Trait Anxiety Inventory (STAI-S
were part-time employees, and 6 (11%) were unemployed. + T; Spielberger, Gorsuch, & Lushene, 1970), consisting
Educational status of the PTSD participants was as fol- of two forms with 20 items (1–4 scale) each. Depression
lows: 8 (15%) finished elementary school, 41 (77%) had a was assessed with the Beck Depression Inventory (BDI;
high-school education, and 4 (8%) had a university de- Beck, Steer, & Garbin, 1988; Beck, Ward, Mendelsohn,
gree. Educational status of the nonclinical participants Mock, & Erbaugh, 1961), which has 21 items rated on a
was: 9 (17%) finished elementary school, 36 (68%) had a 0–3 scale.
high-school education, and 8 (15%) had a university de-
gree. Because of limited cell size, subjects were classified
into two groups based on employment. Those who were Quality of Life
employed full-time comprised one group (including full-
time students), and those who were either unemployed The QOLI that was developed by Frisch et al. (1992)
or employed part-time comprised the second. Chi-square was translated into Swedish by a doctoral student at the
tests showed no significant differences between the two Department of Psychology, Stockholm University, Stock-
groups on marital status and educational level. However, holm, Sweden. Then the second author, who is a profes-
the PTSD participants had a lower employment rate than sor in clinical psychology with an extensive experience
the nonclinical control group (χ 2 = 15.93, p < .001). in writing and speaking in American English, back-
translated each sentence at a time of the QOLI and made
correction notes. After back-translation only a few correc-
Assessments tions were necessary to make.
The Swedish self-report version of the QOLI con-
Clinical Interviews sists of 16 weighted life satisfaction items that measures a
person’s self-perceived quality of life in 16 life areas (see
The CAPS-IV (Blake et al., 1997) was used to as- Table I). Each of the 16 areas of life deemed potentially
sess the frequency and intensity of the 17 symptoms of relevant to overall life satisfaction was rated by subjects
PTSD outlined in the DSM-IV (each scored 0–4). A total in terms of its importance to their overall happiness and
CAPS-IV severity score was computed by adding both the satisfaction (0 = not at all important, 1 = important, 2
frequency and intensity ratings of the 17 PTSD symptoms. = extremely important) and in terms of their satisfaction
A global severity rating of PTSD symptoms was based on with the area (−3 = very dissatisfied to 3 = very satisfied).
an estimation of the overall subjective distress, functional Overall life satisfaction was calculated by first computing
impairment, and observed behaviors during the interview the product of the satisfaction and importance ratings for
(including reporting style). As an integrated part of the each area of life. Next, all weighted satisfaction ratings
ADIS-IV the Hamilton Anxiety Scale (HAS; Hamilton, that had nonzero importance ratings were averaged. Items
1959, 14 items, 0–4 scale) and the Hamilton Depression or areas that were deemed irrelevant to an individual’s
Scale (HDS; Hamilton, 1960, 25 items, 1–5 scale) were overall life satisfaction were omitted.
administered.

Data Analysis
Self-Report Questionnaires
Internal consistency (Cronbach’s alpha) of the QOLI
PTSD symptom measures included the PTSD Symp- was calculated for the PTSD (N = 53) and the nonclini-
tom Scale (PSS-SR; Foa, Riggs, Dancu, & Rothbaum, cal group (N = 100) respectively. A MANOVA was per-
1993), and the Impact of Event Scale-Revised (IES-R; formed with the 16 weighted life satisfaction items of the
Weiss & Marmar, 1997). The PSS consists of 17 items QOLI as dependent variables and clinical status (crime
(0–3 scale) that correspond to the 17 symptoms included in victims with PTSD vs. nonclinical participants) as the
the DSM-III-R (American Psychiatric Association, 1987) factor (see Table I). Significance levels were adjusted
as diagnostic criteria for PTSD. The IES-R consists of with Bonferroni corrections (0.05/16). Pearsson product
22 items (0–4 scale), measuring 8 intrusion, 8 avoidance, moment correlations between the QOLI total score and
and 6 hyperarousal PTSD symptoms. Generalized anxi- psychopathology measures of PTSD symptoms, depres-
P1: GXB
Journal of Psychopathology and Behavioral Assessment (JOBA) PP1052-JOBA-476509 November 22, 2003 15:35 Style file version June 25th, 2002

Swedish Cross-Cultural Validation of the QOLI 19

Table I. Means, Standard Deviations, and F-Values of Weighted Items on the Swedish Version of the Quality of Life
Inventory (QOLI) Among the Gender and Age Matched Crime Victims With PTSD and Nonclinical Participants

PTSD Nonclinical participants

Weighted item M SD M SD Difference F(1, 104)

1. Health −0.23 3.71 3.38 2.94 30.70∗∗

2. Self-regard 0.08 3.04 3.45 1.61 50.96∗∗
3. Philosophy of life 0.13 3.24 2.94 2.00 28.85∗∗
4. Standard-of-living −0.94 2.90 1.75 1.77 33.40∗∗
5. Work −0.19 3.70 3.08 2.05 31.65∗∗
6. Recreation 0.09 3.49 3.19 2.34 28.73∗∗
7. Learning −0.49 2.82 2.43 2.22 35.17∗∗
8. Creativity −0.51 3.23 2.74 2.04 38.21∗∗
9. Social service 0.15 3.17 1.15 2.26 3.50
10. Love relationships 0.49 3.61 4.51 2.05 49.65∗∗
11. Friendships 1.77 2.78 4.09 2.11 23.42∗∗
12. Relationships with children 1.43 3.00 2.49 2.88 3.42
13. Relationships with relatives 1.28 3.15 3.06 2.55 10.16∗
14. Home 2.15 3.24 3.13 2.88 2.72
15. Neighborhood 0.57 2.54 2.23 2.87 9.97∗
16. Community −0.49 2.97 2.38 2.54 28.63∗∗

∗p < .0031. ∗∗ p < .001 (Bonferroni adjusted).

sion, and anxiety were calculated for the PTSD group (see The QOLI was significantly inversely correlated with
Table II). the intrusion measures of the CAPS, the PSS, and the IES-
R (see Table II). The intrusion scales of the CAPS and
the PSS were somewhat less correlated with the QOLI
RESULTS

Cronbach’s coefficients alpha of the total QOLI score Table II. Pearson Correlations Between Psychopathology
for the PTSD and the nonclinical group were 0.84 and 0.80 Measures and the Swedish Quality of Life Inventory (QOLI)
Among Crime Victims With PTSD
respectively. The MANOVA revealed a significant effect
for clinical status, Wilks’ 3 = 0.35, F(16, 89) = 10.28, Measure QOLI
p < .001. Subsequent univariate F tests adjusted with
Bonferroni corrections showed that the PTSD group had CAPS
Intrusion −.34∗
a significantly lower quality of life than the nonclinical
Avoidance/numbing −.49∗∗
group in the life areas of health, self-regard, philosophy Arousal −.32∗
of life, standard-of-living, work, recreation, learning, cre- Total severity −.41∗∗
ativity, love relationships, friendships, relationships with Global severity −.52∗∗
relatives, neighborhood and community (see Table I). No PSS
Intrusion −.38∗
significant differences were found in the life domains of
Avoidance/numbing −.62∗∗
social service, relationships with children, and home. In- Arousal −.36∗∗
spection of the effect sizes was conducted in the life ar- Total −.51∗∗
eas for which significant differences were found between IES-R
the groups. Large differences emerged between the PTSD Intrusion −.43∗∗
Avoidance −.29∗
group and the nonclinical matched group in the life areas of
Arousal −.38∗∗
self-regard (ε2 = .33), love relationships (ε2 = .32), cre- Total −.39∗∗
ativity (ε2 = .27), learning (ε2 = .25), standard-of-living HDS −.55∗∗
(ε2 = .24), work (ε2 = .23), health (ε2 = .23), philoso- BDI −.65∗∗
phy of life (ε 2 = .23), recreation (ε2 = .22), community HAS −.52∗∗
BAI −.44∗∗
(ε2 = .22), and friendships (ε2 = .18). Small differences
STAI-S −.43∗∗
were found between the groups in the life domains of STAI-T −.52∗∗
relationships with relatives (ε2 = .09) and neighborhood
(ε 2 = .09). ∗p < .05. ∗∗ p < .01.
P1: GXB
Journal of Psychopathology and Behavioral Assessment (JOBA) PP1052-JOBA-476509 November 22, 2003 15:35 Style file version June 25th, 2002

20 Paunović and Öst

than the intrusion scale of the IES-R. Significant nega- matched nonclinical group in 13 out of 16 domains of life.
tive correlations were found between the QOLI and the Large differences between the two groups were found in
avoidance/numbing symptom measures of the CAPS and the life areas of self-regard, love relationships, creativ-
the PSS, and between the QOLI and the avoidance symp- ity, learning, standard-of-living, work, health, philosophy
tom scale of the IES-R. The avoidance/numbing symptom of life, recreation, community, and friendships. Smaller
scales of the CAPS and the PSS showed larger correlations differences were found in the life areas of relationships
with the QOLI than did the avoidance scale of the IES- with relatives and neighborhood. There were no signifi-
R. The correlations between the QOLI and the arousal cant differences between the groups in the life domains of
symptom measures of the CAPS, the PSS, and the IES-R social service, relationships with children, and home. A
were of a similar magnitude. Furthermore, the QOLI was possible explanation for the lack of differences between
significantly correlated with the total scores of the CAPS, the groups in the latter life areas may be that the present
the PSS, and the IES-R, and with the global severity score PTSD group consisted of recently traumatized individu-
of the CAPS. The correlations between the QOLI and the als. This possibility is supported by empirical studies that
interview and self-report PTSD symptom measures were showed that significant proportions of crime victims with
of a low–moderate magnitude. PTSD recover during the first months after a traumatic
The QOLI was significantly negatively correlated event (Riggs, Rothbaum, & Foa, 1995; Rothbaum, Foa,
with the depression symptom measures of the HDS and Riggs, Murdock, & Walsh, 1992). Crime victims with
the BDI. Significant inverse correlations were also found chronic PTSD that has on average persisted for several
between the QOLI and the anxiety symptom measures of years may have a more severe deteriorated self-perceived
the HAS, the BAI, the STAI-S, and the STAI-T. Further- quality of life across the life areas measured by the QOLI
more, all the inverse correlations between the QOLI and than the recently traumatized sample in this study. Future
the interview and self-report measures of depression and research should explore this issue.
anxiety symptoms were of a moderate magnitude. The construct validity of the Swedish QOLI version
was supported by its significant negative correlations with
all interview and self-report measures of PTSD symptoms,
DISCUSSION depression, and anxiety. The concept of life satisfaction
seems to be clinically important to crime victims with
This study validated the Swedish version of the QOLI PTSD. However, it is not possible to conclude that life
using a group of crime victims with PTSD and a group satisfaction represents an inverse concept to PTSD, de-
of nonclinical individuals from the general population in pression, and anxiety. Because the correlations were of
the Stockholm County of Sweden with no lifetime and a low–moderate magnitude most of the variance in psy-
current psychiatric disorder. The internal consistency of chopathology measures was accounted for by factors other
the QOLI was excellent in both groups. Also, the PTSD than life satisfaction. In conclusion, it may be useful to in-
group showed a significantly lower self-perceived quality clude the QOLI as an assessment measure of life satisfac-
of life, in 13 out of 16 domains of life, than the gender and tion of psychiatrically disordered individuals to measure
age matched nonclinical group. Furthermore, the QOLI positive mental health that may capture some unique as-
was significantly inversely correlated with all interview pects of clients’ functioning.
and self-report measures of PTSD symptoms, depression, There are several methodological limitations of this
and anxiety. Overall, the Swedish QOLI version showed study. First, a traumatized group without PTSD was not
satisfactory psychometric properties in this study. included. Therefore, it is not possible to conclude whether
The Swedish QOLI version displayed a strong in- the lower quality of life in the PTSD group relative to
ternal consistency in both the PTSD and the nonclinical the nonclinical participants and the negative correlations
group. These results can be compared to Frisch et al.’s between the QOLI and psychopathology measures were
result who found that the QOLI showed strong internal due to PTSD or to the traumatic experience. To explore
consistencies in two clinical and two nonclinical samples whether a deteriorated self-perceived quality of life
(Frisch et al., 1992; .77–.89). Thus, the cross-cultural val- in crime victims is due to PTSD or to the traumatic
idation of the Swedish QOLI version displayed similar experience a traumatized control group without PTSD
high internal consistencies as the clinical and nonclinical must be included in future studies. Second, the inclusion
U.S. samples. criteria were strict, which may limit the generalizability
The Swedish QOLI version showed a satisfactory of the results. Third, since the PTSD participants were
discriminative validity in this study. The PTSD group crime victims that had experienced heterogeneous types
displayed a significantly lower quality of life than the of violent events, the results cannot be generalized to
P1: GXB
Journal of Psychopathology and Behavioral Assessment (JOBA) PP1052-JOBA-476509 November 22, 2003 15:35 Style file version June 25th, 2002

Swedish Cross-Cultural Validation of the QOLI 21

PTSD groups that have experienced homogeneous kinds Blake, D. D., Weathers, F. W., Nagy, L. M., Kaloupek, D. G., Gusman,
of violent events (e.g. rape) or other types of traumatic F. D., Charney, D. S., & Keane, T. M. (1995). The development of
a clinician-administered PTSD scale. Journal of Traumatic Stress,
events (e.g. motor vehicle accidents). The Swedish QOLI 8, 75–90.
version should be evaluated with PTSD samples that Breslau, N., Kessler, R. C., Chilcoat, H. D., Schultz, L. R., Davis, G. C.,
have experienced homogeneous and accidental-related & Andreski, P. (1998). Trauma and posttraumatic stress disorder in
the community: The 1996 Detroit area survey of trauma. Archives
traumatic events to conclude whether it is a useful of General Psychiatry, 55, 626–632.
quality of life measure in PTSD in general in Swedish Brown, T. A., Di Nardo, P. A., & Barlow, D. H. (1994). Anxiety disorders
samples. Fourth, because self-perceived quality of life in interview schedule for DSM-IV. Albany, NY: Center for Stress and
Anxiety Disorders, University at Albany.
various life domains was only assessed with a self-report Brown, T. A., Di Nardo, P. A., Lehman, C. L., & Campbell, L. A. (2001).
questionnaire the results may have been due to common Reliability of DSM-IV anxiety and mood disorders: Implications
method variance. That is, the systematic variance may for the classification of emotional disorders. Journal of Abnormal
Psychology, 110, 49–58.
have, in large part, been due to the method employed rather Foa, E. B., Riggs, D. S., Dancu, C. V., & Rothbaum, B. O. (1993).
than real differences between the clinical and nonclinical Reliability and validity of a brief instrument for assessing post-
groups in self-perceived quality of life or true correlations traumatic stress disorder. Journal of Traumatic Stress, 6, 459–473.
Frisch, M. B., Cornell, J., Villanueva, M., & Retzlaff, P. J. (1992). Clin-
between the QOLI and the psychopathology measures. To ical validation of the quality of life inventory. A measure of life
minimize this potential confound it is necessary to include satisfaction for use in treatment planning and outcome assessment.
an interview version of the QOLI, administer additional Psychological Assessment, 4, 92–101.
Hamilton, M. (1959). The assessment of anxiety states by rating. British
validated measures of quality of life and assess the Journal of Medical Psychology, 32, 50–55.
convergent validity of the QOLI, and replicate the present Hamilton, M. (1960). A rating scale for depression. Journal of Neurology,
findings with other Swedish clinical and nonclinical Neurosurgery and Psychiatry, 23, 56–61.
Jordan, B. K., Marmar, C. R., Fairbank, J. A., Schlenger, W. E., Kulka,
samples. R. A., Hough, R. L., & Weiss, D. S. (1992). Problems with fam-
ilies of male Vietnam veterans with posttraumatic stress disorder.
Journal of Consulting and Clinical Psychology, 60, 916–926.
ACKNOWLEDGMENTS Kilpatrick, D. G., & Resnick, H. S. (1993). Posttraumatic stress disorder
associated with exposure to criminal victimization in clinical and
community populations. In J. R. T. Davidson & E. B. Foa (Eds.),
We thank Peter Csatlos and Aina Lindgren for their Posttraumatic stress disorder: DSM-IV and beyond (pp. 113–143).
help with the data collection from the nonclinical partici- Washington, DC: American Psychiatric Press.
Mendlowicz, M. V., & Stein, M. B. (2000). Quality of life in individuals
pants in this study. with anxiety disorders. American Journal of Psychiatry, 157, 669–
682.
Resnick, H. S., Kilpatrick, D. G., Dansky, B. S., Saunders, B. E., & Best,
REFERENCES C. (1993). Prevalence of civilian trauma and post-traumatic stress
disorder in a representative national sample of women. Journal of
Consulting and Clinical Psychology, 61, 984–991.
American Psychiatric Association. (1987). Diagnostic and statistical Riggs, D. S., Rothbaum, B. O., & Foa, E. B. (1995). A prospective exam-
manual of mental disorders (3rd ed.). Washington, DC: Author. ination of symptoms of post-traumatic stress disorder in victims of
American Psychiatric Association. (1994). Diagnostic and statistical non-sexual assault. Journal of Interpersonal Violence, 2, 201–214.
manual of mental disorders (4th ed.). Washington, DC: Author. Rothbaum, B. O., Foa, E. B., Riggs, D., Murdock, T., & Walsh, W. (1992).
Beck, A. T., Epstein, N., Brown, G., & Steer, R. A. (1988). An inventory A prospective examination of post-traumatic stress disorder in rape
for measuring clinical anxiety. Journal of Consulting and Clinical victims. Journal of Traumatic Stress, 5, 455–475.
Psychology, 56, 893–897. Safren, S. A., Heimberg, R. G., Brown, E. J., & Holle, C. (1996/1997).
Beck, A. T., Steer, R. A., & Garbin, M. G. (1988). Psychometric proper- Quality of life in social phobia. Depression and Anxiety, 4, 126–133.
ties of the Beck Depression Inventory: Twenty-five years of evalu- Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1970). Manual
ation. Clinical Psychology Review, 8, 77–100. for the Stait–Trait Anxiety Inventory. Palo Alto, CA: Consulting
Beck, A. T., Ward, C. H., Mendelsohn, M., Mock, J., & Erbaugh, J. Psychologists Press.
(1961). An inventory for measuring depression. Archives of General Weiss, D. S., & Marmar, C. R. (1997). The impact of event scale—
Psychiatry, 4, 561–571. Revised. In J.P. Wilson & T.M. Keane (Ed.), Assessing psycholog-
Blake, D. D., Weathers, F. W., Nagy, L. M., Kaloupek, D. G., ical trauma and PTSD (pp. 399–411). New York: Guilford Press.
Charney, D. S., & Keane, T. M. (1997). Boston: Clinician- Zatzick, D. F., Marmar, C. R., Weiss, D. S., Browner, W. S., Metzler, T.
administered PTSD scale for DSM-IV. National center for posttrau- J., Golding, J. M., Stewart, A., Schlenger, W. E., & Wells, K. B.
matic stress disorder, behavioral science division at Boston VA med- (1997). Posttraumatic stress disorder and functioning and quality of
ical center and neurosciences division at West Haven VA medical life outcomes in a nationally representative sample of male Vietnam
center. veterans. American Journal of Psychiatry, 154, 1690–1695.