Content uploaded by Etzel Cardeña
Author content
All content in this area was uploaded by Etzel Cardeña on Mar 11, 2025
Content may be subject to copyright.
Nordic Journal of Psychiatry
ISSN: (Print) (Online) Journal homepage: www.tandfonline.com/journals/ipsc20
Development and validation of a new measure for forced
migrants/refugees: the Screening of Refugees Self-Report
(ScoRE-SR)
Etzel Cardeña, Charlotte Sonne, Erik Vindbjerg, Jessica Carlsson Lohmann,
Matti Cervin, Derrick Silove, Joshua Hall, Mariano Coello, Shakeh Momartin,
Sabina Gušić & Jorge Aroche
To cite this article: Etzel Cardeña, Charlotte Sonne, Erik Vindbjerg, Jessica Carlsson Lohmann,
Matti Cervin, Derrick Silove, Joshua Hall, Mariano Coello, Shakeh Momartin, Sabina Gušić
& Jorge Aroche (10 Mar 2025): Development and validation of a new measure for forced
migrants/refugees: the Screening of Refugees Self-Report (ScoRE-SR), Nordic Journal of
Psychiatry, DOI: 10.1080/08039488.2025.2475940
To link to this article: https://doi.org/10.1080/08039488.2025.2475940
© 2025 The Author(s). Published by Informa
UK Limited, trading as Taylor & Francis
Group
View supplementary material
Published online: 10 Mar 2025.
Submit your article to this journal
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=ipsc20
ARTICLE
NORDIC JOURNAL OF PSYCHIATRY
Development and validation of a new measure for forced migrants/refugees:
the Screening of Refugees Self-Report (ScoRE-SR)
Etzel Cardeñaa, Charlotte Sonneb, Erik Vindbjergb, Jessica Carlsson Lohmannb,c, Matti Cervina,
Derrick Siloved, Joshua Halle, Mariano Coelloe, Shakeh Momartine, Sabina Gušića and Jorge Arochee
aDepartment of Psychology, Lund University, Sweden; bMental Health Center Ballerup, Copenhagen University Hospital – Mental Health
Services CPH, Copenhagen, Denmark; cDepartment of Clinical Medicine, University of Copenhagen, Denmark; dPsychiatry Research and
Teaching Unit at Liverpool Hospital (PRTU), Australia; eNSW Service for the Treatment and Rehabilitation of Torture and Trauma Survivors
(STARTTS), Australia
ABSTRACT
Purpose: To remedy the lack of a measure that jointly assesses the psychological status and impact of
migration stressors among forced migrants, we developed and tested the Screening of Refugees
Self-Report (ScoRE-SR) Questionnaire.
Method: Four institutions with expertise on posttraumatic symptomatology and/or migration stressors
developed the ScoRE-SR through an iterative process also involving consultation with migrants and
experts. The measure consists of 54 items on functioning, migration stressors, and posttraumatic
symptomatology. It was administered in English, Danish, Arabic, Farsi, Dari, and Tamil to migrants
(N = 149) from different countries and residency types (citizen, permanent resident, holder of temporary
protection visa, and Asylum Seeker) at refugee assistance institutions in Australia and Denmark.
Analyses: The measure’s internal psychometric properties were evaluated through confirmatory factor
analyses and test-retest (at 2–3 months) intraclass correlations. Its convergent validity was evaluated
through correlating the measure with questionnaires expected to relate to it.
Results: The original 13 theoretical factors of the Score-SR showed adequate model-data fit, but CFAs
indicated that a factor structure with 6 broader factors was more appropriate. These factors (emotional
distress, anger, concerns about family/friends in other countries, concerns about family/friends in country
of resettlement, adjustment/resettlement/practical difficulties, and impairment) were internally consistent,
showed adequate temporal stability, and correlated as expected with measures of posttraumatic
symptomatology and well-being.
Conclusions: The ScoRE-SR is a valid measure of different types of migrants’ distress and/or adjustment,
filling a current gap and providing essential information for migrants’ guidance and treatment.
Introduction: the need for a comprehensive new
refugee assessment tool
In this paper we present a new comprehensive measure to
evaluate the psychological health and migration stressors
among forced migrants. There were more than 117 million
displaced people worldwide at the beginning of 2024 [1],
typically fleeing from various types of conflicts and violence.
In addition to the serious distress and trauma associated with
the causes for displacement, when migrants arrive to a new
country they are likely to face barriers and stressors that may
exacerbate their distress. Potentially traumatizing events,
including conflicts that force the displacement of millions of
people [2], can bring about or exacerbate post-traumatic
stress (PTSD), dissociation, anxiety, depression, somatization,
and other symptomatology (e.g [3,4]). In addition, post-
migration variables such as the length of the asylum process,
residency status, and social integration can impact the men-
tal health and treatment of migrants. For instance,
post-migration stressors have been found to predict treat-
ment adherence [5] and outcomes in longitudinal studies
with outpatient torture survivors [6,7].
A narrative systematic review of 22 studies examining the
impact of post-migration variables on mental health showed
that in 9 of them the length of the asylum process predicted
psychiatric morbidity [8]. Other factors of relevance to
migrants’ mental health include a distressing family situation,
poor social integration and weak social networks, financial
difficulties, and housing and accommodation status, with less
clear effects of unemployment, language proficiency, educa-
tion, and gender. A similar narrative review with 21 studies
mentioned additional factors potentially impacting mental
health, including discrimination and unemployment [9].
© 2025 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
CONTACT Etzel Cardeña etzel.cardena@psychology.lu.se Department of Psychology, Lund University, Sweden
Supplemental data for this article can be accessed online at https://doi.org/10.1080/08039488.2025.2475940.
https://doi.org/10.1080/08039488.2025.2475940
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted
Manuscript in a repository by the author(s) or with their consent.
ARTICLE HISTORY
Received 12 November
2024
Revised 12 February 2025
Accepted 3 March 2025
KEYWORDS
Forced migrants;
refugees; measure;
migration stressor;
psychological health
2 E. CARDEÑA ETAL.
Some refugee groups have many sources of potentially trau-
matizing events, including high early childhood trauma (e.g.
physical and sexual abuse), trauma related to armed conflicts,
persecution, imprisonment, and torture (both as victim and per-
petrator), leaving the conflict area (with related loss of goods
and lack of resources), and post-migratory stressors, including
acculturation problems and stressors related to the asylum pro-
cess and application for visa/residency status [10]. There are
many measures of posttraumatic stress disorder symptomatology
(e.g [11].), depression, anxiety, somatization, and dissociation,
some of which have been adapted to refugees (e.g. the Refugee
Health Screener 15, see chrome-extension://efaidnbmnnnibpcajp-
cglclefindmkaj/https://www.refugeehealthta.org/wp-content/
uploads/2012/09/RHS15_Packet_PathwaysToWellness-1.pdf).
A few questionnaires of post-migration stressors have been
developed, including the Post-Migration Living Difficulties
Checklist [12] and the Refugee Post-Migration Stress Scale [13],
covering different psychosocial domains. Lacking so far has
been a measure that encompasses both potential migration-
related psychological symptoms and stressors, which might
interact. The effectiveness of psychosocial interventions for
migration posttraumatic symptomatology among adults is
impacted by migration stressors (e.g [14]), yet there has not
been a generally adopted measure of such stressors [15].
Furthermore, there is no measure that we know of that evalu-
ates common posttraumatic symptoms and postmigration
stressors together, which would not only save time and effort,
but allow for easier analyses of how each variable impacts
intervention outcomes.
Considering these gaps, the authors of this study collabo-
rated to develop and test the psychometric properties of a
self-report measurement tool for refugees: the Screening of
Refugees Self-Report (ScoRE-SR) questionnaire. Our aim was to
create and evaluate an instrument to evaluate different types
of symptoms and stressors commonly experienced by forced
migrants. The aims of the measure were to measure all major
areas of potential distress, function, and triggering migration
stressors, not be too lengthy for respondents/counselors/
translators, and have comparable response format for all
items. We also wanted an instrument with the potential to
assess intervention outcomes.
Materials and methods
Development of the Screening of Refugees Self-Report
(ScoRE-SR)
Four international institutions with expertise on posttrau-
matic symptomatology and/or migration stressors collabo-
rated in the development of the new measure: Australia’s
NSW Service for the Treatment and Rehabilitation of Torture
and Trauma Survivors (STARTTS) and the Psychiatry Research
and Teaching Unit at Liverpool Hospital (PRTU), Denmark’s
Competence Centre for Transcultural Psychiatry (CTP), and
Sweden’s Center for Research on Consciousness and
Anomalous Psychology at Lund University (CERCAP).
We initiated the development of the ScoRE-SR by review-
ing relevant measures of psychological health and stressor
issues and selecting the most relevant items for forced
migration. We identified four main types of refugee mea-
sures: those focusing on specific diagnoses, very brief rating
scales to screen potential adjustment problems, comprehen-
sive and long mental health instruments, and those focus-
ing on migration-related stressors. We concluded that rather
than using a variety of instruments with different format-
ting and accessibility, it would be preferable to develop a
new questionnaire targeted to refugee populations. The first
section would include items measuring symptoms of PTSD,
depression, anxiety, grief, somatic symptoms, dissociation,
and anger. The psychosocial subscale would include various
domains of psychosocial functioning related to the immi-
gration experience including safety/security; bonds/network;
justice; roles and identities; existential meaning; attachment
and a sense of belonging; and transitional issues.
Then, data from the STARTTS and CTP’s Danish database
on Refugees with Trauma (DART) databases were used to
identify areas of distress for forced migrants. Those areas
were discussed with different migrant groups in STARTTS to
adjust questions and phrasing depending on the groups’
input, which resulted in increasingly refined iterations of the
measure. All focus groups were recruited from existing
STARTTS eligible groups with the support of the counselor
working with them. A researcher explained the purpose of
the study and no group declined to participate. A preliminary
draft was also sent to international experts on trauma inquir-
ing as to the measure’s cultural appropriateness and relevant
items we might have missed. A final measure with multiple
items for each domain but not onerous to complete became
the final version to be tested.
Testing of the ScoRE-SR
The final version of ScoRE-SR questionnaire was administered
to clients at STARTTS and CTP. It was also going to be used
in Region Skåne, Sweden, but because of the COVID pan-
demic it could not be tested there. The ScoRE-SR consists of
54 items that evaluate: (a) General Functioning (4 items deal-
ing with inability to complete activities and substance abuse);
Concerns about family living in other countries (5 items);
Concerns about family or friends in country of resettlement
(6 items); Resettlement/adjustment concerns (7 items);
Practical needs (7 items); and all of the following with 3
items each: Anxiety; Depression; Grief and Loss; Post-Traumatic
Stress Disorder symptoms; Anger; Somatic Symptoms;
Dissociation, and Psychotic symptoms. We also included one
validity item (‘I felt happy with every aspect of my life’). The
score range is 1–4, with the anchors being ‘never,’ ‘sometimes,’
‘often,’ and ‘always’ or ‘no distress,’ ‘mild distress,’ ‘moderate
distress,’ and ‘extreme distress,’ depending on the question
(see Appendix A, supplementary material).
The ScoRE-SR and four other validating existing measures
used were translated in Australia by a known professional
company into Arabic, Farsi, Dari, and Tamil and evaluated by
counselors with linguistic and cultural expertise. In Denmark,
the ScoRE-SR was translated into Danish by two indepen-
dent translators, who then agreed on their final version,
which was back translated by a third translator and com-
pared with the original. All Danish translators then agreed
NORDIC JOURNAL OF PSYCHIATRY 3
on a final version, which was approved by a native speaker
layperson. All translations are available from the STARTTS
coauthors.
Other measures
The Harvard Trauma Questionnaire-Part 4 (HTQ [16]) is a
widely used self-administered 16-item measure of PTSD
symptomatology based on Diagnostic and Statistical Manual
of Mental Disorders (DSM-IV) criteria. It uses a four-point
scale from ‘not at all’ to ‘extremely’ and has been shown to
be reliable with refugee samples (Hollifield et al. 2002) [17].
The internal reliability was good in the present study: omega
= .87 at baseline.
The Sheehan Disability Scale (SDS [18]) is a self-report rat-
ing instrument that measures impairment in three domains:
work/everyday tasks, social network, and family obligations,
through a visual analogue scale asking if symptoms have dis-
rupted an activity, from 0 (not at all) to 10 (extremely). The
scale has good reliability and validity [19] and the internal
reliability at baseline scores was adequate in the present
study, omega = .71.
The WHO-5 [20] is a self-administered questionnaire that
evaluates quality of life/well-being, on a scale from 5 (all of
the time) to 0 (at no time). The instrument has well-supported
validity and sensitivity to change [21] and has been used in
trials with trauma-affected refugees (e.g [22]). Internal consis-
tency was good in our sample, omega = .88. In contrast with
the other measures, higher scores on this scale represent bet-
ter well-being.
The 25-item Hopkins Symptom Checklist (HSCL-25 [23]) is a
commonly used self-report questionnaire of symptoms of
anxiety and depression, rated on a four-point scale: ‘Not at
all’ to ‘Extremely.’ It has been translated into a range of lan-
guages [24] and has shown good convergent validity and
internal consistency with several non-western populations
[17,25]. In our study, the internal consistency was excellent,
omega = .93.
Procedure
Australia
STARTTS staff and community partners were given a flyer
(translated into the relevant languages) to enroll suitable
potential participants who met the study criteria and
expressed interest in the study. The referrer then obtained
permission to share contact details of potential participants
with the STARTTS research team, which invited them to a
focus group. Separate groups were held for each language,
facilitated by a researcher, and with an interpreter and a
researcher present. The initial focus groups commenced with
an informed consent process to reiterate the purpose and
process of the study. Then a semi-structured group interview
began by asking about the participants perceptions of psy-
chosocial problems experienced by people from refugee
backgrounds, and exploring how the problems identified
may impact day-to-day functioning and exploring the themes
of the questions.
A follow-up focus group was completed after approxi-
mately nine months. All participants of the initial focus
groups were invited to attend the follow-up focus group; of
the seven initial groups, three follow-up groups were com-
pleted. The semi-structured group interview presented the
items of the draft of the psychosocial subscale and asked for
feedback and opinions on how to improve it. In Australia,
ethics approval was secured from the South Western Sydney
Local Health District Human Research Ethics Committee
(HREC Reference: HREC/16/LPOOL/560; SSA Reference: SSA/16/
LPOOL/585; Local Project Number: HE16/300).
Denmark
Participants from Denmark were informed about the study
orally and in writing by the treating physician and asked if
they wished to take part in the study. For those who needed
it, a professional interpreter translated the general informa-
tion and consent form, while the written questionnaires were
in their mother tongue. The relevant Danish ethics commit-
tee concluded that data collection did not require additional
approval (Capital Region of Denmark Regional Committee,
VEK Journal H-18041753). Those who were willing to partici-
pate again (n = 85) were retested after about 2–3 months to
evaluate concordance with improvements detected by other
scales. However, because of uncertainty as to whether such
short treatment duration would produce clinically noticeable
improvements, we later decided to use this second testing
solely for test-retest analyses. For all data from Denmark,
double data entry with validation was carried out.
Both sites
Participants could answer the questionnaires in either English,
Danish, Arabic, Farsi, Dari, or Tamil, either alone or with the
assistance of a translator. Some of the participants in both
centers were also asked, and agreed to, fill out some of the
four additional questionnaires: HTQ, WHO-5, HSCL, and SDS.
The inclusion criteria were: (a) being at least 18 years old,
(b) a refugee seeking individual counseling, and (c) being
cognitively and emotionally able to give informed consent
and participate in the project. No other inclusion/exclusion
criteria were adopted so as to increase the generalizability of
results to migrants in general. All participants gave informed
consent to participate in the study.
Statistical analyses
The ScoRE was administered in English or, as needed, in the
other languages to which it was translated. There were only
few responses in alternate languages, so no statistical analy-
ses could be carried on these subgroups but they were rather
added to the whole sample. We conducted independent
samples t-tests to compare ScoRE ratings from STARTTS
(n = 39) and CTP (n = 110) at baseline. No predetermined ns
were selected, rather, all available responses were used
during the time of data collection. The only significant differ-
ence was for impairment (which was the only factor with
questionable reliability, see below), where those recruited at
STARTTS (M = 7.72, SD = 2.63) scored lower than those at CTP
4 E. CARDEÑA ETAL.
(M = 8.74, SD = 2.39; t[147] = −2.23, p = .03, Cohen’s d = 0.42).
We decided to combine both groups in the analyses that fol-
low. To evaluate the internal psychometric properties of the
ScoRE-SR, we used confirmatory factor analysis (CFA).
Exploratory factor analysis was not used because the size of
the sample precluded performing both exploratory and con-
firmatory analyses. Furthermore, considering that the mea-
sure is based on a clear theoretical structure with well-defined
domains, we deemed CFA to be appropriate.
Our CFA approach included two steps: (1) A-priori model
testing: Examining data/model fit of a theory-derived factor
structure and using fit indices, estimated parameters, and
modification indices to evaluate its adequacy. (2) Post-hoc
model testing: Fitting a refined model based on results from
step 1. Step 1 was performed by fitting a model that included
the 13 theoretically defined factors: anxiety, depression, func-
tional impairment, grief/loss, post-traumatic stress, anger,
somatic symptoms, dissociation, psychotic symptoms, con-
cerns about family/friends in other countries, concerns about
family/friends in country of resettlement, adjustment/resettle-
ment concerns, and difficulties regarding practical needs. The
model/data fit of this model and specific item loadings, mod-
ification indices, and covariance among latent factors
informed model improvement in step 2. Modification indices
indicate which parameters to relax to improve model fit and
can help identify sources of misfit and ways to improve over-
all model/data fit.
Because the items of the measure were ordinal, we used
the diagonally weighted least squares estimator and a global
evaluation of four fit indices to evaluate overall model/data
fit: Comparative Fit Index (CFI), Root Mean Square Error of
Approximation (RMSEA), Tucker-Lewis Index (TLI), and
Standardized Mean Square Residual (SRMR). An RMSEA below
0.06, an SRMR below 0.08, and CFI and TLI estimates greater
than 0.90 are indicative of acceptable model-data fit; CFI and
TLI estimates above 0.95 are indicative of good model-data
fit [26]. To inform model improvement in step 2, standardized
item loadings were inspected and modification indices evalu-
ated to identify sources of misfit. The covariance among
latent factors was inspected to see whether some factors
were strongly correlated and whether their items could be
collapsed and included under the same factor, with standard-
ized covariance coefficients above .70 indicating that collaps-
ing of items may be justified. Of note, to provide information
regarding the psychometric properties of the validity item
(item 10, ‘I felt happy with every aspect of my life’), it was
included under the depression factor. This decision was based
on its thematical relatedness to depression and our expecta-
tion was that it would load negatively onto the factor.
The internal consistency of the items of the latent factors
were examined by estimating McDonald’s omega (ω) coeffi-
cients and we considered coefficients above .70 to indicate
adequate internal consistency. The average variance extracted
(AVE) was computed to estimate the degree of item variance
explained by the latent factor. All CFA analyses were con-
ducted using the R library lavaan. Test-retest reliability was
examined by computing the intraclass correlation coefficient
(ICC [2,k]), with values above .50 being used as an indication
of adequate reliability [27]. Missing data at the item-level for
the ScoRE-SR measure at assessment 1 (2.3%, n = 150) and
assessment 2 (2.5%, n = 87) were low and handled using pair-
wise deletion. Convergent validity was examined by correlat-
ing the empirically derived scores of the ScoRE-SR and the
HTQ, HSCL-25, SDS, and WHO-5. The criterion for significance
was set at p < .05.
Results
Participants
At baseline, the ScoRE-SR was administered to forced migrant
clients at CTP (n = 116; Mage = 50.51; SD = 68.93; 50% female,
49% male, 1 participant undetermined) and STARTTS (n = 40;
Mage = 44.02; SD = 12.77; 42.5% female, 57.5% male), along
with four additional measures to evaluate its validity. Some
CTP participants were retested 2–3 months later. Participants
came from at least 23 different countries, and their residential
categories were Citizen, Permanent Resident, TVP (temporary
protection visas) Humanitarian, other TVP, and Asylum Seeker.
A priori model testing and adjustment of the model
The theoretical model with 13 factors exhibited acceptable
overall model/data fit (RMSEA = .038 [95% CI: .031 to .045],
SRMR = .091, CFI = .938, TLI = .931). Three items had stan-
dardized loadings below .30, indicating a weak loading: items
4 (impairment, factor, ‘I engaged in excessive alcohol and/or
drug use,’ loading = .23), 10 (the validity item, depression fac-
tor, ‘I felt happy with every aspect of my life,’ loading = − .10),
and 30 (concerns about family/friends in other countries fac-
tor, ‘Not knowing whether my family members or close
friends may get hurt,’ loading = .11). Additionally, item 22
(factor: somatic symptoms, ‘I had unexplained aches in my
arms, legs or joints,’) had a .35 loading, while all remaining
loadings were ≥ .49. Of note, the result for the validity item
indicates that some participants scored the item in a
non-expected way. We reviewed all individual scores and
around 20% of respondents scored the validity similar to
depression items. This suggests that the validity item can be
used to evaluate potential misunderstanding, acquiescence,
or response bias.
Modification indices indicated that model fit would be
improved by letting items 8 (‘I did not have enough energy
to carry out simple tasks’) and 9 (‘I felt sad and was unable
to find joy in anything’) be indicators of several other scales.
Correlated residuals between items 7 and 27 (‘I was scared
by sudden sounds or movements’ and ‘I heard sounds or
voices when there was no one around’) and 4 and 30 (‘I
engaged in excessive alcohol and/or drug use’ and ‘Not
knowing whether my family members or close friends may
get hurt’) were also suggestive but would result in much less
improvement of fit than to let items 8 and 9 have
cross-loadings onto several of the latent factors. Further, very
strong standardized covariance coefficients among all symp-
tom factors emerged (with most being > .70), except for the
anger factor that showed smaller, but still substantial, covari-
ance with other symptom factors (.37 to .74). The anxiety,
depression, and grief factors showed very high standardized
NORDIC JOURNAL OF PSYCHIATRY 5
covariance coefficients (> .84). and the standardized covari-
ance coefficient between the adjustment/resettlement con-
cerns factor and the difficulties regarding practical needs
factor was also high (.87) (see Appendix B, supplementary
material).
Post-Hoc model testing
Based on the results outlined above, we refined the model
by: (1) omitting items 4, 10, 22, and 30; (2) grouping the anx-
iety, depression, and grief items into a single factor, (3)
grouping the items of adjustment/resettlement concerns and
difficulties regarding practical needs factors into a single fac-
tor; (4) including a higher-order latent factor capturing all
symptom factors except anger, as an indicator of a broad
emotional distress factor; and (5) including correlated residu-
als for items 7 and 27 (correlated residuals for items 7 and 30
could not be included as item 30 was omitted, see above).
These adjustments resulted in a model with 10 first-order fac-
tors (see Table 1) in which the Anxiety/depression/grief,
Posttraumatic symptoms, Dissociation, Psychosis symptoms,
and Somatization factors were included in an overarching
factor that we call Emotional Distress (see Appendix C, sup-
plementary material).
The model-data fit of the refined model was good (RMSEA
= .035 [.027 to .043], SRMR = .085, CFI = .953, TLI = .949) and
the model was more parsimonious than the original model.
All items displayed a loading ≥ .48 and all symptom factors
displayed a significant loading > .73 onto the high-order
emotional distress factor. The latent factors showed correla-
tions between .13 (impairment & concerns about family/
friends in other countries) and .62 (concerns about family/
friends in country of resettlement & adjustment/resettlement/
practical difficulties), indicating that they measured partially
distinct constructs.
The means and standard deviations for all factors are in
Table 1. They show that participants reported high levels
(i.e. ratings close to ‘often’) of impairment, anxiety/depres-
sion/grief, and posttraumatic symptoms, and between
‘sometimes’ and ‘often’ levels of anger, dissociation, psycho-
sis symptoms, somatization, concern for friends/family in
the original country and the resettlement country, and
resettlement/adjustment and practical issues.
Reliability
The internal consistency (see also Table 2), based on the fit-
ted CFA model, of the items of each latent factor was good
to excellent, except for the somatization factor (ω = .48). The
AVEs were overall adequate: impairment = 58%, anxiety/
depression/grief = 44%, posttraumatic stress = 56%, anger =
76%, somatic symptoms= 38%, dissociation = 49%, psychotic
symptoms = 73%, concerns about family/friends in other
countries = 69%, concerns about family/friends in country of
resettlement = 51%), and adjustment/resettlement/practical
difficulties = 49%. We measured test-retest reliability for those
who were retested 2–3 months after baseline, with intra-class
correlations for each of the six single and composite factors
(Table 2). Except for impairment, all factors showed adequate
test-retest reliability.
Convergent validity
To evaluate the convergent validity of the ScoRE—SR, we
estimated product moment correlations between its factors
and questionnaires that conceptually should correlate posi-
tively (HTQ, HHSCL-25, and SDS) or negatively (the WHO-5,
which is scored in an opposite direction to the other mea-
sures) (Table 3). As expected, the impairment factor had
moderate positive correlations with measures of PTSD symp-
tomatology, anxiety/depression, and dysfunction, and a mod-
erate negative correlation with the measure of quality of life.
The factor of adjustment/practical difficulties had similar
Table 1. Descriptive statistics for single and grouped factors.
Category Mean SD
Impairment 8.47 2.49
Anxiety/depression/grief 28.29 4.95
Posttraumatic symptoms 9.02 2.25
Anger 7.51 2.80
Dissociation 7.20 2.32
Psychosis symptoms 6.27 2.97
Somatization 5.25 1.67
Concern family: old country 12.26 3.66
Concern family:new country 14.93 5.07
Adjustment diculties 35.25 11.38
Table 2. Test-retest reliability and internal consistency for the six SCORE-SR
factors (n = 85).
ICC ω
Impairment .46 .77
Emotional Distress* .74 .91
Anger .77 .87
Concerns about family/friends in
other countries
.77 .85
Concerns about family/friends in
resettlement site
.74 .90
Adjustment/resettlement/practical
diculties
.81 .64
Note: * = all items of the subfactors pooled.
Table 3. Correlations between the SCORE-SR factors and HTQ, WHO-5, HSCL, and SDS.
HTQ
n = 89
WHO-5
n = 84 to 85
HSCL
n = 58 to 59
SDS
n = 51 to 52
Impairment .43 −0.31 .30 .34
Emotional distress .71 −0.42 .72 .29
Anger .42 −0.34 .24† .23†
Concerns about family/friends in other countries .47 −0.30 .24† .10†
Concerns about family/friends in country of resettlement .32 −0.30 .38 .16†
Adjustment/resettlement/practical diculties .52 −42 .39 .21†
Note. † = not statistically signicant at p < .05.
6 E. CARDEÑA ETAL.
correlations except that it, surprisingly, did not correlate sig-
nificantly with dysfunction.
Similarly, the emotional distress factor correlated strongly
with measures of PTSD, depression and anxiety symptoms,
weakly with disability, and moderately (and negatively) with
quality of life. The anger factor correlated moderately and posi-
tively with PTSD symptoms (which includes an item about
being irritable or angry) and negatively with quality of life.
Concerns about family/friends both in other countries and the
country of resettlement had moderate positive correlations with
the measure of PTSD symptoms and a negative correlation with
quality of life. Finally, Concerns with those in resettlement coun-
try also corelated positively with depression/anxiety.
Discussion
The results show that the revised factors of the ScoRE-SR are
overall internally consistent and have adequate structural
validity. To improve its psychometric properties, three original
items were omitted from those factors: ‘I engaged in exces-
sive alcohol and/or drug use,’ ‘I had unexplained aches in my
arms, legs or joints,’ and ‘Not knowing whether my family
members or close friends may get hurt.’ Nonetheless, we are
retaining them in the measure as checklist items because
they can provide important clinical information, which may
be discussed further with the clinician. In addition, item 10 ‘I
felt happy with every aspect of my life’ is a validity item
excluded from the factors but should be taken into consider-
ation if the respondent answer with an unrealistic high score
(e.g. 3, 4) discrepant with most of the other items. In that
case, the respondent can be asked further about it.
The fact that the impairment factor of the ScoRE-SR was
less temporally stable than the others can be interpreted as
it being less reliable than the other factors Alternatively, with
respect to the re-testing, it should be borne in mind that it
occurred 2–3 months after participants were receiving treat-
ment, so it might indicate that they became less generally
distressed, although more specific symptomatology was not
as affected.
The ScoRE-SR showed overall convergent validity with
respect to the measures that it should correlate highly with:
the HTQ and the HSCL. The negative correlations between
the ScoRE-SR factors and the WHO-5, which is a measure of
well-being, also support the validity of the former. The
ScoRE-SR had low to moderate correlations with the SDS,
most of them not significant.
Thus, the internal consistency and convergent validity
analyses support the psychometric reliability and validity of
the ScoRE-SR to assess areas of concern for the respondents,
with the caveat that impairment was not temporally stable.
An important consideration is that a total score for the whole
measure would be a very general indicator. Rather, scores for
each of the factors (and for the four items not included in
them) provide more valuable information of what are areas
of concern (and strength) for respondents.
The strengths of this study include having used an itera-
tive approach with multiple sources, including professional
experts and community members, in the development of the
measure. Another strong point is that the data collected
comes from two international sources and the responses in
both sites were overall comparable, suggesting the generaliz-
ability of the measure. Furthermore, the ScoRE-SR can be cur-
rently administered to speakers of Arabic, Danish, English,
Farsi, Dari, and Tamil (contact the authors from STARTTS and
CTP to get those translations).
Nonetheless, the study has various limitations: One is that
participants received interventions during the interim period
between the two assessments used to evaluate test-retest
reliability. Such reliability assesses whether the trait or con-
struct remains stable over time and is ideally examined with-
out any interventions that could affect the outcome.
Nonetheless, the results suggest overall adequate test-retest
reliability. Another limitation is that the complexity of the
model relative to the sample size creates some uncertainty
around individual parameters, and the correlations between
the ScoRE-SR and validating measures had relatively modest
sample sizes.
A very important limitation is that the measure was not
tested as a measure of treatment efficacy, which would be of
great use to centers and clinicians. Additional research should
involve, if at all possible, larger groups followed longitudi-
nally so that specific hypotheses can be tested (for instance,
potential differences in distress of those with citizenship or
residence as compared with those with an uncertain status).
The ScoRE-SR as a measure of treatment efficacy should be
evaluated with repeated testing and long follow-ups. The full
potential of the measure as a measure indicating areas to
focus on in an intervention, and the efficacy of treatment,
should be evaluated in future studies.
The ScoRE-SR will be most useful in its subscale scores
to assess areas of distress and/or maladjustment (along
with the single non-factor items that may be of clinical
concern). Areas of difficult may then be targeted for
potential guidance and/or treatment. Considering the var-
ious dimensions covered by the measure, focusing on an
overall score would provide only a rough estimate of level
of distress/maladjustment.
Despite the limitations of this study, the current forced
migration crisis and consequent need for a comprehensive
valid and reliable measure of reactions to forced migration
amply justify offering the ScoRE-SR to institutions, clinicians,
and researchers alike, given the lack of any measure of both
symptomatology and migration stressors. We look forward to
future refinements of the measure as needed along with its
use to help vulnerable individuals.
Acknowledgments
We thank the contribution of Kit Bibby in the early stages of this project.
We also thank Roberto Lewis-Fernández, Ph. D. and Terry Keane, Ph. D.
for their valuable consultation on the measure.
Disclosure statement
The author(s) declare no potential conicts of interest with respect to
the research, authorship, and/or publication of this article.
NORDIC JOURNAL OF PSYCHIATRY 7
Data availability statement
Because of the sensitive nature of the patient data involved in this study,
the data cannot be shared publicly. Access to the data is restricted and
can only be processed by the authors under specic regulatory and eth-
ical guidelines.
References
[1] UNHCR. Global Trends; 2024. https://www.unhcr.org/globaltrends.
html.
[2] USA for UNHCR. Refugee statistics; 2023; [cited 2023 April 17].
Available from: https://www.unrefugees.org/refugee-facts/statistics/.
[3] Bryant RA, Nickerson A, Morina N, et al. Posttraumatic stress disor-
der in refugees. Annu Rev Clin Psychol. 2023;19(1):413–436.
doi:10.1146/annurev-clinpsy-080921-080359.
[4] Turrini G, Purgato M, Ballette F, etal. Common mental disorders in
asylum seekers and refugees: umbrella review of prevalence and
intervention studies. Int J Ment Health Syst. 2017;11(1):51.
doi:10.1186/s13033-017-0156-0.
[5] Djelantik AAAMJ, de Heus A, de Kuiper D, et al. Post-migration
stressors and their association with symptom reduction and
non-completion during treatment for traumatic grief in refugees.
Front Psychiatry. 2020;11:407. doi:10.3389/fpsyt.2020.00407.
[6] Schick M, Morina N, Mistridis P, et al. Changes in post-migration
living diculties predict treatment outcome in traumatized refu-
gees. Front Psychiatry. 2018;9:476. doi:10.3389/fpsyt.2018.00476.
[7] Sonne C, Mortensen EL, Silove D, et al. Predictors of treatment out-
comes for trauma-aected refugees – results from two randomised
trials. J Aect Disord. 2021;282:194–202. doi:10.1016/j.jad.2020.12.095.
[8] Gleeson C, Frost R, Sherwood L, et al. Post-migration factors and
mental health outcomes in asylum-seeking and refugee popula-
tions: a systematic review. Eur J Psychotraumatol. 2020;11(1):1793567.
doi:10.1080/20008198.2020.1793567.
[9] Jannesari S, Hatch S, Prina M, et al. Post-migration social–environ-
mental factors associated with mental health problems among asy-
lum seekers: a systematic review. J Immigr Minor Health.
2020;22(5):1055–1064. doi:10.1007/s10903-020-01025-2.
[10] Carlsson J, Sonne C. Mental health, pre-migratory trauma and
post-migratory stressors among adult refugees. In Morina N,
Nickerson A, editors. Mental health of refugee and conict aected
populations. Myrtle Beach, SC: Springer; 2018. p. 15–35.
[11] Wilson J. P, Keane T. M. Assessing psychological trauma and PTSD.
New York, NY: Guilford; 1997.
[12] Silove D, Sinnerbrink I, Field A, et al. Anxiety, depression and PTSD
in asylum-seekers: associations with pre-migration trauma and
post-migration stressors versus special characteristics. Br J
Psychiatry. 1997;170(4):351–357. doi:10.1192/bjp.170.4.351.
[13] Malm A, Tinghög P, Narusyte J, et al. The refugee post-migration
stress scale (RPMS) – development and validation among refugees
from Syria recently resettled in Sweden. Con Health. 2020;14(1):2.
doi:10.2190/T8EM-C8YH-373N-1UWD.
[14] Bruhn M, Rees S, Mohsin M, etal. The range and impact of postmigra-
tion stressors during treatment of trauma-aected refugees. J Nerv
Ment Dis. 2018;206(1):61–68. doi:10.1097/nmd.0000000000000774.
[15] Mässing A, Cardeña E. The relation between post-migration
stressors and trauma treatment outcomes: a scoping review;
under review.
[16] Mollica R, Caspi-Yavin Y, Bollini P, et al. The Harvard Trauma
Questionnaire. Validating a cross-cultural instrument for mea-
suring torture, trauma, and posttraumatic stress disorder in
Indochinese refugees. J Nerv Ment Dis. 1992;180(2):111–116.
doi:10.1097/00005053-199202000-00008.
[17] Renner W, Salem I, Ottomeyer K. Cross-cultural validation of mea-
sures of traumatic symptoms in groups of asylum seekers from
Chechnya, Afghanistan, and West Africa. Soc Behav Pers.
2006;34(9):1101–1114. doi:10.2224/sbp.2006.34.9.1101.
[18] Sheehan DV. The anxiety disease: new hope for the millions who
suer from anxiety. London: Bantam; 1986.
[19] Leon AC, Olfson M, Portera L, etal. Assessing psychiatric impairment
in primary care with the Sheehan Disability Scale. Int J Psychiatry
Med. 1997;27(2):93–105. doi:10.2190/T8EM-C8YH-373N-1UWD.
[20] World Health Organization: regional Oce for Europe. Use of
well-being measures in primary health care - the DepCare project.
Target 12. E60246. Geneva: WHO; 1998.
[21] Topp CW, Østergaard SD, Søndergaard S, et al. The WHO-5
Well-Being Index: a systematic review of the literature. Psychother
Psychosom. 2015;84(3):167–176. doi:10.1159/000376585.
[22] Buhmann C, Mortensen EL, Nordentoft M, etal. Follow-up study of
the treatment outcomes at a psychiatric trauma clinic for refugees.
torture. 2018;25(1):16. doi:10.7146/torture.v25i1.109505.
[23] Parlo MB, Kelman HC, Frank JD. Comfort, eectiveness, and
self-awareness as criteria for improvement in psychotherapy. AJP.
1954;111(5):343–352. doi:10.1176/ajp.111.5.343.
[24] Mollica R, McDonald LS, Masagli MP, et al. Measuring trauma.
Measuring torture; 2004. Harvard Program in Refugee Trauma.
https://healtorture.org/resources/harvard-trauma-questionnai
re-measuring-trauma-measuring-torture-htq/.
[25] Hollield M, Warner TD, Lian N, etal. Measuring trauma and health
status in refugees: a critical review. JAMA. 2002;288(5):611–621.
doi:10.1001/jama.288.5.611.
[26] Hu LT, Bentler PM. Cuto criteria for t indexes in covariance struc-
ture analysis: conventional criteria versus new alternatives. Struct
Equat Model. 1999;6(1):1–55. doi:10.1080/10705519909540118.
[27] Prinsen CA, Mokkink LB, Bouter LM, et al. COSMIN guideline for
systematic reviews of patient-reported outcome measures. Qual
Life Res. 2018;27(5):1147–1157. doi:10.1007/s11136-018-1798-3.
