https://orcid.org/0000-0002-0035-3194
Figures
Abstract
Background
Given the high prevalence of trauma-related mental health problems in a growing population of refugees, identifying robust predictors of treatment outcome can optimize treatment strategies and resource allocation. The current study aimed to examine the robustness of predictors from seven previous predictor studies at the Competence Centre for Transcultural Psychiatry (CTP).
Methods
We studied if predictors from previous studies could be replicated in a sample of 219 refugees with PTSD who participated in a new distinct study conducted at CTP. We included all variables previously found to be significantly related to treatment response, as measured via the Harvard Trauma Questionnaire (HTQ) and the Hamilton Depression Scale (HAM-D). We categorised the variables as sociodemographic factors, CTP predictor index items, and baseline assessment scores. We performed regression analyses with one exposure variable at a time, controlling for baseline assessment scores corresponding to the dependent variable. We also conducted multiple linear regression analyses to examine the variance explained in total by the identified significant predictor variables.
Results
The study successfully replicated several predictors of treatment response for PTSD and depression, including a high understanding of therapy concepts, acceptability of psychological treatment, ability to reflect, motivation for active participation and cognitive resources, as well as low baseline pain levels. The findings replicated a U-shaped relationship between age and treatment response for depression, indicating better outcomes for younger participants and those over 60. Multiple regression analyses explained 28% and 39% of the variance in the PTSD (HTQ) and depression (HAM-D) treatment responses, respectively.
Conclusions
This study contributes to the understanding of treatment response predictors in trauma-affected refugees by replicating findings from previous research in a new sample. It highlights the importance of modifiable factors, such as psychotherapy readiness, and underscores the necessity for tailored interventions to enhance treatment efficacy. As the global refugee crisis grows, the insights from this research can inform mental health strategies, ultimately improving care for this vulnerable population. Future research should continue to investigate modifiable predictors to further enhance treatment outcomes.
Citation: Carlsson J, Bruhn M, Poschmann IS, Silove D, Westergaard ML, Mortensen EL, et al. (2025) Identifying robust predictors of treatment response in trauma-affected refugees: Results from a randomised controlled trial. PLoS One 20(9): e0324935. https://doi.org/10.1371/journal.pone.0324935
Editor: Mu-Hong Chen,, Taipei Veterans General Hospital, TAIWAN
Received: May 1, 2025; Accepted: August 30, 2025; Published: September 26, 2025
Copyright: © 2025 Carlsson et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data that support the findings of this study are available upon request from the Competence Centre for Transcultural Psychiatry www.ctp-net.dk. The data are not publicly available due to privacy and ethical restrictions. Public deposition would breach compliance with the protocol approved by the Ethics Committee of the Capital Region of Denmark and the Danish Data Protection Agency. Furthermore, public availability would compromise patient privacy.
Funding: The RCT was supported by grants from TrygFonden (JC, grant number 120354) and Fonden til Lægevidenskabens Fremme (HS, grant number 16-319). There was no additional external funding received for this study. All funders did not play a role in the study design, data collection, analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Identifying predictors of treatment response is an important focus of research aimed at developing better treatment options. With an increasing number of trauma-affected refugees, this avenue of research is called for due to a pressing need to effectively utilise scarce resources for this vulnerable population. As of mid-2024, the number of refugees reached 37.8 million; compared with 14.4 million refugees 10 years ago [1,2]. Trauma-affected refugees are at high risk of developing mental health problems, with an estimated prevalence of 30% for PTSD and depression [3,4]. Both pre- and postmigration factors contribute to these mental health problems [5–7]. Psychosocial interventions for trauma-affected refugees have resulted in small- to medium-sized improvements [8–10] and most often need to be comprehensive and multidisciplinary to address the complex health conditions and challenges of trauma-affected refugees [9–11]. Hence, the growing number of refugees, combined with the costly nature of mental health interventions and lack of improvement during these interventions, emphasise the need to optimise treatment strategies to improve treatment response and make more effective use of resources [12].
Need for studies on predictors of treatment response
Importantly, robust predictors of treatment outcome can assist healthcare personnel in assigning interventions that are more likely to be successful, thereby optimising the use of resources.
Efforts have been made to determine which factors might predict treatment response in trauma-affected refugees, given that not all individuals show a positive response to best-practice treatments. In previous reseach, several possible predictors of treatment response have been identified; including sociodemographic factors; exposure to traumatic events; and factors directly related to mental and physical health, such as the severity of symptoms and comorbidities [12–15].
Past studies examining predictors of treatment response are nevertheless limited by small sample sizes, differences in the factors considered as possible predictors, and variations in outcome measures, all of which contribute to inconsistent findings [13,14,16–20]. Thus, although several predictors of treatment response have been identified, knowledge on robust predictors of treatment response in psychosocial interventions remains limited. Studies reproducing previous findings are necessary but nevertheless lacking [21,22].
The identification of modifiable predictors in treatment, such as sleep or motivation, presents an opportunity for providing treatment interventions aimed specifically at improving these areas, thereby increasing the likelihood of success of the intervention [23]. Conversely, identifying non-modifiable factors such as offender status or age can provide insights into the necessity of early identification and may assist in assigning interventions that are more likely to be successful [24]. While such factors cannot be changed, they can still inform clinical decision-making; for example, individuals experiencing persistent guilt or rumination may benefit from trauma-focused interventions that address these manifestations of distress.
Replicating findings from previous predictor studies
At the Competence Center for Transcultural Psychiatry (CTP), a specialised mental health outpatient clinic in Denmark, we have performed a total of five randomised controlled trials (RCTs) with trauma-affected refugees. On the basis of data from the first four trials, we have previously published seven papers identifying different predictors of treatment response [12,24–29]. However, we have not examined whether the predictors of treatment response identified across these four trials can be replicated in a similar but new and distinct sample from our fifth randomized controlled trial [30,31]. Predictors from the first four trials are summarised in Table 1.
The standard treatment in all five RCTs was a multidisciplinary psychosocial intervention. Sociodemographic and mental health data for the populations included in the five RCTs show only unsystematic differences, allowing for comparisons between studies [32–36].
The aim of the current study was to explore whether the findings of treatment outcome predictors from previous CTP studies are robust and can be replicated in this study’s population. The current study examines the robustness of these factors by repeating the analysis with the identified possible predictors in a new sample of 219 trauma-affected refugees who participated in an RCT conducted at CTP. Furthermore, we examine the variance explained in total by the identified significant predictor variables.
Improved knowledge on factors influencing treatment response will enable offering improved treatment.
Materials and methods
Study design
Data for the current study derived from a a randomized controlled superiority trial with an allocation ratio of 1:1:1:1 and an allocation sequence with block size unknown to the investigator. The randomization was stratified by gender.
The patients were randomised into four groups. All four groups received treatment as usual (TAU), which consisted of a 10-month intervention that included manual psychotherapy, physiotherapy, psychopharmacotherapy, psychoeducation, and social counselling. One group received solely TAU, serving as a control group; the three remaining groups were active treatment groups receiving add-on treatment with either mianserin, IRT, or a combination of both.
The primary outcome in the RCT was the Pittsburgh Sleep Quality Index. Based on previous studies, a minimal clinically important difference of 2.5 points and a standard deviation of 3 were assumed. With 90% power and alpha = 0.05, the required sample size was 128. To account for an expected dropout of 25%, the target sample was increased to 228 participants. Please see study protocol and paper reporting main results for in-depth description of methods and results of the RCT [30,36]. The first patient was enrolled on the 12th of May 2016 the primary completion date was the 1st of June 2019.
Ethics approval and consent to participate
All the participants provided signed informed consent. The trial was approved by the Danish Medicines Agency, the Ethics Committee of the Capital Region of Denmark, and the Danish Data Protection Agency. The study was registered at ClinicalTrials.gov, ID: NCT02761161, on 27 April 2016 (https://clinicaltrials.gov/study/NCT02761161?id=NCT02761161&rank=1). The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.
Participants
The study sample consisted of 219 trauma-affected refugees, the intention-to-treat sample of the RCT. Given that there were no differences in outcomes across the intervention groups [36], the current study examined the complete sample across intervention groups. The participants were adult refugees (or persons whose families reunified with refugees) diagnosed with PTSD according to the ICD-10.
Measures
Variable selection.
Outcome measures: For the present study, PTSD symptoms were measured via self-reports on the Harvard Trauma Questionnaire (HTQ) [37,38], and symptoms of depression were measured via observer ratings on the Hamilton Depression Scale (HAM-D) [39].
Exposure variables: We selected candidate predictor variables on the basis of an assessment of the seven previous CTP predictor studies [12,24–28,40]. We included all variables previously found to be significantly related to the HTQ or HAM-D via correlational, bivariate, or multiple regression analyses [12,24–28,40]. We categorised the variables as sociodemographic factors, CTP predictor index items, and baseline assessment scores. The variables are presented in Table 1.
The CTP predictor index (S1 Appendix) was developed in 2012 to assess predictors of treatment response on the basis of clinical experience and available literature on trauma-affected refugees [12,41]. The index initially consists of 15 items rated on a 0–4 Likert scale (4 being the best score) according to predefined criteria [12]. A medical doctor rated five items, a psychologist rated five items, and a social counsellor rated five items. The index was later modified on the basis of experiences from previous CTP predictor index studies and the needs of ongoing RCTs [26,27,30,42].
The following candidate variables were included in the analyses:
- Sociodemographic factors included were age, years since arrival in Denmark, gender, migrant status refugee, combat experience, need for an interpreter in psychologist sessions, education above ten years in the country of origin, stable occupation/not on financial support, and duration of functional impairment.
- The CTP predictor index items included were upbringing, previous psychiatric treatment attempts, chronic pain, chronicity of the mental condition, understanding of concepts of therapy, acceptability of psychological treatment, ability to reflect, motivation for active participation, cognitive resources, daily activity, employment status, and economy (self-perceived).
- Baseline assessment scores included were Global Assessment of Functioning (GAF-F) [43], the Brief Pain Inventory-Short Form (BPI) [44], and headache measured on item 8 of the HSCL-25.
Statistics
To describe the characteristics of the participants, means and standard deviations were calculated for continuous variables, and percentages were calculated for categorical variables. The mean rating scores were calculated pre- and posttreatment, and pre- and posttreatment differences were tested with paired t-tests.
We tested our main hypothesis that previous treatment predictors could be replicated in this study population by performing regression analyses with one exposure variable at a time, controlling for baseline assessment scores corresponding to the dependent variable. In previous predictor studies, correlation analyses, bivariate regression analyses, and hierarchical regression analyses were performed, and different variables were controlled for in the hierarchical regression analyses. In the present study, we chose to perform regression analyses that only controlled for baseline scores to compare variables across studies.
We used pre- to posttreatment score differences in symptoms of PTSD (HTQ) and depression (HAM-D) as dependent variables and included the variables mentioned above as exposure variables.
To supplement the regression analyses, we performed multiple linear regression analyses in two models to examine the variance explained in total by the identified significant predictor variables. In Model 1, we included all the sociodemographic factors, baseline assessment scores identified as significant in the regression analyses, and the baseline assessment scores corresponding to the dependent variable to adjust for baseline symptoms. In Model 2, CTP predictor index items identified as significant in regression analyses were added to the model. Based on the results of the initial multiple linear regression analyses, we tested an alternative (Model 2b) regression analysis to avoid the effects of possible collinearity among the five CTP predictor index items rated by a psychologist. Instead of the individual predictor items, this analysis included the first principal component derived from these five items.
In the initial analyses, we included an indicator of the intervention group in the RCT. However, we did not identify significant differences between the intervention groups, and the variable was nonsignificant in the regression analyses. Hence, the presented analyses do not include an indicator of the intervention group.
To assess assumptions of linearity, a scatterplot of the included continuous variables was plotted with a superimposed regression line. Visual inspection of these plots revealed a linear relationship between the variables and no outliers of importance. For the variable age, the relationship with treatment response was nonlinear for several dependent variables; thus, both linear and quadratic terms for age were included.
Assumptions of independence of errors, homoscedasticity, and normality of residuals were met to a reasonable degree based on visual inspection of residual-versus-fitted plots and Q-Q plots, as well as rvfplot commands in Stata. Multicollinearity was assessed using variance inflation factors (VIF), and no problematic levels were observed.
All analyses were conducted via STATA 17. Linear regression analyses were performed via STATA’s SEM procedure and Full Information Maximum Likelihood analyses to include all available information, including data for participants with missing posttreatment outcome scores.
Results
The study included 219 participants. The majority came from Syria, Afghanistan, Iraq, and Iran. Trauma histories were diverse: 98% had experienced war, 42% had been imprisoned, and 37% had experienced violence from relatives. Further characteristics and the included exposure variables are presented in Table 2.
The mean scores before and after treatment and the changes from pre- to posttreatment for the HTQ and HAM-D are presented in Table 3. We found significant improvement in scores on the HTQ (p = 0.000) but not on the HAM-D (p = 0.6316).
Table 4 presents the results of regression analyses between candidate predictors and treatment response for symptoms of PTSD and depression controlled for baseline values in the current sample.
Replicating findings from previous predictor studies
We identified the following significant variables as predictors of greater symptom improvement in symptoms of PTSD (HTQ): younger age (p = 0.017), migrant status (refugee) (p = 0.034), greater understanding of the concepts of therapy (p = 0.000), higher acceptability of psychological treatment (p = 0.000), stronger ability to reflect (p = 0.000), higher motivation for active participation (p = 0.017), high cognitive resources (p = 0.001), and low pain score on BPI at baseline (p = 0.001). The variable with the greatest association was understanding the concepts of therapy (β = 0.32).
For depression (HAM-D), the following variables significantly predictored greater symptom improvement: greater understanding of the concepts of therapy (p = 0.016), higher acceptability of psychological treatment (p = 0.003), stronger ability to reflect (p = 0.001), higher motivation for active participation (p = 0.008), high cognitive resources (p = 0.007), abscence of chronic pain (p = 0.034) and low pain score on BPI at baseline (p = 0.034). In contrast, medium duration of functional impairment was associated with less symptom improvement (i.e., poorer treatment outcome) (p = 0.005). The variable with the strongest association was the ability to reflect (β = 0.27).
Variance explained by the identified significant predictor variables
Table 5 presents the results of the multiple linear regression analyses and shows the variance explained in total by the identified significant predictor variables. Model 1 included all the sociodemographic factors and baseline assessment scores identified as significant in the regression analyses and the baseline assessment scores corresponding to the dependent variable, to adjust for baseline symptoms. Model 1 accounted for 19% and 32% of the variance in improvement in symptoms for PTSD and depression, respectively.
In Model 2, the CTP predictor index items found to be significant in the regression analyses were added to Model 1, increasing the explained variance by 9% and 5%. The model now accounted for 28% and 39% of the variance in symptom improvement for PTSD and depression, respectively. In Model 2, age added significantly to the prediction of depression symptom improvement (HAM-D) (p = 0.000). Furthermore, the quadratic term included for age was significantly associated with treatment response for depressive symptoms (HAM-D) (p = 0.000), reflecting a nonlinear relationship. For HAM-D, the association between age and treatment response showed a U-shaped relationship, indicating a greater treatment response, primarily for participants younger than 40 years but also for the oldest group above 60 years.
In Model 2, which included all the CTP predictor index variables, none of these variables were significantly associated with any of the dependent variables. Given the significance observed in analyses of the five individual CTP predictor items rated by a psychologist, the lack of significance in Model 2 might reflect collinearity among the predictor variables. Consequently, a principal component analysis was conducted on the five predictor variables. Only one eigenvalue was greater than 1, and the first principal component explained 69% of the variance. All the variables had high loadings on the first principal component, ranging from 0.77 (cognitive resources) to 0.88 (ability to reflect). When scores on the first principal component were included in alternative Model 2b together with the variables from Model 2, the component was a highly significant predictor of symptom improvement for both PTSD (HTQ) (β = 0.30 p = 0.000) and depression (HAM-D) (β = 0.25 p = 0.001).
Furthermore, owing to the similarly high loadings on the first principal component for the five CTP predictor items rated by a psychologist, we calculated the mean score for the five items. The correlation coefficient between the mean score and the first principal component was close to 1, and essentially the same results were observed when the mean score replaced the first principal component in a regression model.
Discussion
Repeated studies have shown only small- to medium-sized improvements in the treatment of trauma-affected refugees [8–10]. Thus, we still need to explore how we can improve the support offered to this group. With several possible predictors of treatment response identified in previous studies in the same setting, the current study examined the robustness of these factors by repeating the analysis with the identified possible predictors in a new sample of 219 trauma-affected refugees. The overall aim was to enhance the understanding of factors influencing treatment response in this target group and, ultimately, be able to offer improved treatment.
The replicated predictors for a positive treatment response regarding both PTSD (HTQ) and depression (HAM-D) were age, low pain score (BPI baseline), and all five psychotherapy readiness factors from the CTP predictor index. Furthermore, migrant status was identified as a predictor of PTSD (HTQ), whereas a medium duration of functional impairment and chronic pain were identified as predictors of depression (HAM-D).
Age, chronicity and migration status replicated as predictors of treatment response
Age was replicated as a predictor of treatment response for both PTSD (HTQ) and depression (HAM-D) patients. For depression (HAM-D), we identified a U-shaped relationship, indicating a greater treatment response, primarily for participants younger than 40 years but also for the oldest group above 60 years. This finding is similar to those of previous studies, where we reported better treatment responses for PTSD and depression among patients aged under 40 years [12,24]. There also seems to be a group of the oldest participants with better treatment responses. However, this latter finding is more uncertain because of the smaller sample size in the older group.
Age was not identified as a predictor in a review from 2020 on factors associated with treatment response from psychological treatments for PTSD, where eight of the 25 included studies were on refugees [45]. Although not found in the review, Resick et al. have identified younger age as a predictor of better treatment response in a study on war veterans with PTSD. Several factors could be related to younger age and have also been found in other populations with PTSD [46]. These factors are related to less stigma attached to accepting mental health treatment [47]; fewer social stressors; less comorbidity, including somatic comorbidity; and psychotherapy readiness factors, such as cognitive flexibility and openness to engage in psychotherapy [46]. The younger group might have been distressed for a shorter time and had less time since they experienced trauma, which might be associated with fewer chronic conditions that generally respond better to treatment.
In the present study, a medium duration of functional impairment was replicated as a predictor of negative treatment response for depression (HAM-D), likely reflecting the same group of middle-aged patients not responding to treatment [24]. However, we did not find that other factors related to illness chronicity or complexity, such as previous hospitalisation, time since trauma or arrival at the new country [12], or comorbid depression [13], predicted outcomes. This could be explained by the high level of symptoms experienced by a substantial proportion of the participants and the very high frequency of comorbid depression (>95%) [33,34,48].
Migration status, i.e., having arrived in Denmark as a refugee versus being family-reunified with a refugee, was identified as a predictor for positive treatment response [12], However, in the previous predictor study, the identified relation was the opposite – having arrived in Denmark as a refugee was a predictor for a negative treatment outcome [49]. Furthermore, in a previous study focusing specifically on differences in trauma, psychopathology, and treatment response between refugees and family-reunified, no differences were identified in treatment response, and differences were primarily found regarding traumatic exposure [50]. Based on the current analysis, migration status is not identified as a robust predictor of treatment outcome.
Pain replicated as a predictor of treatment response
In the present study, pain measurements were replicated as predictors of treatment response for both PTSD (BPI) and depressive symptoms (BPI, chronic pain from the CTP predictor index) [24,34]. Those with higher baseline levels of pain improved less during treatment. A study on a similar population of 276 trauma-affected refugees revealed that pain interference (BPI) predicts poorer outcomes for PTSD, depression and anxiety [19]. Research on comorbid mental illness and complaints of pain in refugees is still scarce. There is a need to explore how a focus on relieving pain or the impact of pain can be effectively integrated into the treatment of trauma-affected refugees.
Factors related to readiness for psychotherapy as predictors of treatment response
The current study points to several factors included in the CTP predictor index assessed by psychologists before the start of psychotherapy as predictors of treatment response across outcomes of PTSD and depression. Although these are subjective evaluations (in contrast to demographic data), the replication shows that these factors include information that seems essential in predicting treatment outcomes. Assessment of readiness for psychotherapy may contribute to identifying patients who are likely to benefit from psychotherapy and may help plan treatment for individual patients [51].
According to the multiple regression model, none of the single psychotherapy readiness factors were significant individual predictors. Nevertheless, the multiple regression analyses revealed an explained variance for the different outcomes of 28--39%, which is rather large in this type of study. However, in the bivariate analyses, they all together contributed to explaining the treatment response variance. In Model 2b, the first principal component of the psychotherapy readiness factors in the CTP predictor index was a highly significant predictor across all five outcome measurements. This finding suggests that we cannot choose one of these factors as more important than the others. There is a need to further understand and explore the content of single factors and their interrelatedness, and further develop and search for robust psychotherapy readiness factors. There is a large amount of research on readiness to change and the importance of motivation, as well as studies on interventions trying to improve treatment response, with interventions focusing on these aspects in other populations [52,53]. Several factors evaluated as part of the CTP predictor index are potentially modifiable [52,54]. However, this field is still unexplored in studies on trauma-affected refugees and requires attention.
Strengths & limitations
One strength of this study is that the participants in current and previous trials were included in pragmatic randomised trials; thus, the participants are comparable to those referred to similar clinics treating refugees. Another strength is the use of the same data collection methods (variables and outcome measures) across studies. The CTP predictor index developed and utilised in the CTP is not a validated scale. However, in the current study, we assume no specific psychometric properties by using single items and the first principal component as a composite score. Some items in the scale depend highly on a subjective evaluation of qualities that are difficult to assess, such as the ability to reflect. It is, therefore, interesting that although some of the items leave room for individual interpretation of subjective quality, the results regarding associations with outcomes are rather robust. For future studies, when the items included in the index are reviewed, some could be improved to ensure that different clinicians rate them consistently, and continuous use of the index would require regular training to increase interrater reliability. In the current study, all clinicians received standardized training, but no formal interrater reliability checks were performed, and agreement between raters was not systematically monitored. Future studies should include procedures for evaluating and reporting interrater agreement to strengthen the quality of clinician-rated data.
While all tested predictors were theory-driven and based on previous findings, multiple comparisons increase the risk of false positives. Although no formal adjustment was applied, findings should be interpreted with caution, particularly where p-values approach the conventional significance threshold.
Future research: Possible modifiable factors and factors relevant to clinical practice
In striving to improve services, it is crucial to identify both predictors of patients who benefit from treatment and the barriers to being able to benefit [15]. Among the factors discussed above, potentially modifiable factors are of particular interest in our efforts to improve treatment response. Age is not modifiable per se, but earlier referral could improve treatment response for some patients. The participants’ migrant status is not modifiable, but the treatment may be increasingly adapted to specific stressors and needs of daily living. Currently, we are investigating postmigration stressors and cross-sectoral collaboration in an RCT [55].
In the current study, factors related to readiness for psychotherapy could be a target for modification. This should be explored in studies examining the extent to which these factors are modifiable as well as further developing a reliable measure for this concept in trauma-affected refugees.
The group of refugees with PTSD referred to the CTP often have long-lasting severe distress and impaired functioning, and many have been in treatment previously. Thus, the group as a whole is considered challenging to treat. Despite this, we observed statistically significant improvements in mental health and well-being from pre- to posttreatment in the current and previous studies. This finding is rather stable across studies carried out in the past 13 years. These results suggest that further improvement can be obtained by addressing modifiable factors influencing treatment response.
Conclusions
In conclusion, the findings of this study underscore the importance of identifying and understanding predictors of treatment response in trauma-affected refugees. By replicating and validating previous findings, we have established a clearer picture of the factors that influence treatment efficacy in this vulnerable population.
The findings from our research, highlight key predictors of treatment response that can inform clinical practice and guide future research efforts. First, the identification of modifiable predictors, such as readiness for psychotherapy, suggests that clinicians can enhance treatment outcomes by focusing on these areas. For instance, interventions aimed at increasing motivation and understanding of therapy could be integrated into treatment protocols to improve overall efficacy. Moreover, the recognition of chronic pain as a significant barrier to recovery emphasizes the need for a multidisciplinary approach to treatment. Integrating pain management strategies into mental health interventions may improve overall treatment efficacy for trauma-affected refugees. This integration could involve collaboration with physical therapists and other healthcare professionals to develop comprehensive care plans that address both psychological and physical health needs.
The U-shaped relationship between age and treatment response also suggests that clinicians should consider age-specific factors when designing treatment plans. Younger patients may benefit from interventions that enhance cognitive flexibility and reduce stigma around mental health treatment, while older patients may require strategies that acknowledge their unique experiences and resilience factors.
As the global refugee crisis continues to escalate, the need for effective mental health treatment strategies becomes increasingly urgent. By focusing on modifiable predictors and integrating multidisciplinary approaches, we can enhance the quality of care for trauma-affected refugees, ultimately leading to improved mental health outcomes and a better quality of life for this vulnerable population. Future research must continue to explore these areas to build upon our understanding and improve the effectiveness of interventions for trauma-affected refugees.
To ensure a more comprehensive and robust evaluation of these predictors, future studies should be conducted across multiple regions and countries, using larger and more diverse samples.
References
- 1.
Mid-Year Trends 2024 | UNHCR [Internet]. [cited 2025 Mar 4]. Available from: https://www.unhcr.org/mid-year-trends-report-2024
- 2.
UNHCR - UNHCR Global Trends 2014 [Internet]. [cited 2015 Jul 10]. Available from: http://www.unhcr.org/556725e69.html
- 3. Steel Z, Chey T, Silove D, Marnane C, Bryant RA, van Ommeren M. Association of torture and other potentially traumatic events with mental health outcomes among populations exposed to mass conflict and displacement: a systematic review and meta-analysis. JAMA. 2009;302(5):537–49. pmid:19654388
- 4. Blackmore R, Boyle JA, Fazel M, Ranasinha S, Gray KM, Fitzgerald G. The prevalence of mental illness in refugees and asylum seekers: A systematic review and meta-analysis. PLoS Med. 2020;17(9).
- 5. Fazel M, Wheeler J, Danesh J. Prevalence of serious mental disorder in 7000 refugees resettled in western countries: a systematic review. Lancet. 2005;365(1474-547X (Electronic)):1309–14.
- 6. Miller KE, Rasmussen A. The mental health of civilians displaced by armed conflict: an ecological model of refugee distress. Epidemiol Psychiatr Sci. 2017;26(2):129–38. pmid:27040595
- 7. Gleeson C, Frost R, Sherwood L, Shevlin M, Hyland P, Halpin R, et al. Post-migration factors and mental health outcomes in asylum-seeking and refugee populations: a systematic review. Eur J Psychotraumatol. 2020;11(1):1793567. pmid:33312450
- 8. Thompson CT, Vidgen A, Roberts NP. Psychological interventions for post-traumatic stress disorder in refugees and asylum seekers: A systematic review and meta-analysis. Clin Psychol Rev. 2018;63:66–79. pmid:29936342
- 9. Nosè M, Ballette F, Bighelli I, Turrini G, Purgato M, Tol W, et al. Psychosocial interventions for post-traumatic stress disorder in refugees and asylum seekers resettled in high-income countries: Systematic review and meta-analysis. PLoS One. 2017;12(2):e0171030. pmid:28151992
- 10. Turrini G, Purgato M, Acarturk C, Anttila M, Au T, Ballette F, et al. Efficacy and acceptability of psychosocial interventions in asylum seekers and refugees: systematic review and meta-analysis. Epidemiol Psychiatr Sci. 2019;28(4):376–88. pmid:30739625
- 11. Uphoff E, Robertson L, Cabieses B, Villalón FJ, Purgato M, Churchill R. An overview of systematic reviews on mental health promotion, prevention, and treatment of common mental disorders for refugees, asylum seekers, and internally displaced persons. Cochrane Database Syst Rev. 2020;2020(9).
- 12. Sonne C, Mortensen EL, Silove D, Palic S, Carlsson J. Predictors of treatment outcomes for trauma-affected refugees - results from two randomised trials. J Affect Disord. 2021;282:194–202. pmid:33418367
- 13. Haagen JFG, Ter Heide FJJ, Mooren TM, Knipscheer JW, Kleber RJ. Predicting post-traumatic stress disorder treatment response in refugees: Multilevel analysis. Br J Clin Psychol. 2017;56(1):69–83. pmid:27900778
- 14. Tran A, Brickman S, Jordan J, Samuelson KW. Predictors of Posttraumatic Stress Disorder, Anxiety, and Depression Symptoms in Survivors of Torture Living in the United States. J Nerv Ment Dis. 2020;208(8):593–9. pmid:32541397
- 15. Opaas M, Wentzel-Larsen T, Varvin S. Predictors of the 10 year course of mental health and quality of life for trauma-affected refugees after psychological treatment. Eur J Psychotraumatol. 2022;13(1):2068910. pmid:35957631
- 16. Halvorsen JØ, Stenmark H, Neuner F, Nordahl HM. Does dissociation moderate treatment outcomes of narrative exposure therapy for PTSD? A secondary analysis from a randomized controlled clinical trial. Behav Res Ther. 2014;57:21–8. pmid:24762779
- 17. Stenmark H, Catani C, Neuner F, Elbert T, Holen A. Treating PTSD in refugees and asylum seekers within the general health care system. A randomized controlled multicenter study. Behav Res Ther. 2013;51(10):641–7. pmid:23916633
- 18. Stammel N, Knaevelsrud C, Schock K, Walther LCS, Wenk-Ansohn M, Böttche M. Multidisciplinary treatment for traumatized refugees in a naturalistic setting: symptom courses and predictors. Eur J Psychotraumatol. 2017;8(sup2):1377552. pmid:29163866
- 19. Nordin L, Perrin S. Pre-treatment pain predicts outcomes in multimodal treatment for tortured and traumatized refugees: a pilot investigation. Eur J Psychotraumatol. 2019;10(1):1686807. pmid:31762955
- 20. van Wyk S, Schweitzer R, Brough M, Vromans L, Murray K. A longitudinal study of mental health in refugees from Burma: the impact of therapeutic interventions. Aust N Z J Psychiatry. 2012;46(10):995–1003. pmid:22431841
- 21. Begley CG, Buchan AM, Dirnagl U. Robust research: Institutions must do their part for reproducibility. Nature. 2015;525(7567):25–7. pmid:26333454
- 22. Baker M, Penny D. Is there a reproducibility crisis? Nature. 2016;533(7604):452–4.
- 23. Morris LS, Grehl MM, Rutter SB, Mehta M, Westwater ML. On what motivates us: a detailed review of intrinsic v. extrinsic motivation. Psychol Med. 2022;52(10):1801–16. pmid:35796023
- 24. Nordbrandt MS, Vindbjerg E, Mortensen EL, Carlsson J. Chronicity of posttraumatic stress disorder and comorbid pain as predictors of treatment response for trauma-affected refugees. J Trauma Stress. 2022;35(5):1393–404. pmid:35446986
- 25. Buhmann C, Mortensen EL, Nordentoft M, Ryberg J, Ekstrøm M. Follow-up study of the treatment outcomes at a psychiatric trauma clinic for refugees. Torture. 2015;25(1):1–16. pmid:26021344
- 26.
Nordbrandt MS. Dissertation: Treatment for trauma-affected refugees A randomised controlled trial of physical activity as add-on to treatment as usual. University of Copenhagen; 2020.
- 27. Sonne C, Carlsson J, Bech P, Vindbjerg E, Mortensen EL, Elklit A. Psychosocial predictors of treatment outcome for trauma-affected refugees. Eur J Psychotraumatol. 2016;1:1–11.
- 28. Sander R, Laugesen H, Skammeritz S, Mortensen EL, Carlsson J. Interpreter-mediated psychotherapy with trauma-affected refugees - A retrospective cohort study. Psychiatry Res. 2019;271:684–92. pmid:30791342
- 29. Westergaard M, Jensen R, Carlsson J. Headache comorbidity in refugees and migrants with post-traumatic stress disorder. Cephalalgia. 2023.
- 30. Sandahl H, Jennum P, Baandrup L, Poschmann IS, Carlsson J. Treatment of sleep disturbances in trauma-affected refugees: Study protocol for a randomised controlled trial. Trials. 2017;18(1):520. pmid:29110681
- 31. Sandahl H, Jennum P, Baandrup L, Lykke Mortensen E, Carlsson J. Imagery rehearsal therapy and/or mianserin in treatment of refugees diagnosed with PTSD: Results from a randomized controlled trial. J Sleep Res. 2021;30(4).
- 32. Buhmann CB, Nordentoft M, Ekstroem M, Carlsson J, Mortensen EL. The effect of flexible cognitive-behavioural therapy and medical treatment, including antidepressants on post-traumatic stress disorder and depression in traumatised refugees: pragmatic randomised controlled clinical trial. Br J Psychiatry. 2016;208(3):252–9.
- 33. Carlsson J, Sonne C, Vindbjerg E, Mortensen EL. Stress management versus cognitive restructuring in trauma-affected refugees-A pragmatic randomised study. Psychiatry Res. 2018;266:116–23. pmid:29859498
- 34. Sonne C, Carlsson J, Bech P, Elklit A, Mortensen EL. Treatment of trauma-affected refugees with venlafaxine versus sertraline combined with psychotherapy - a randomised study. BMC Psychiatry. 2016;16(1):383. pmid:27825327
- 35. Nordbrandt MS, Sonne C, Mortensen EL, Carlsson J. Trauma-affected refugees treated with basic body awareness therapy or mixed physical activity as augmentation to treatment as usual-A pragmatic randomised controlled trial. PLoS One. 2020;15(3):e0230300. pmid:32163509
- 36. Sandahl H, Jennum P, Baandrup L, Lykke Mortensen E, Carlsson J. Imagery rehearsal therapy and/or mianserin in treatment of refugees diagnosed with PTSD: Results from a randomized controlled trial. J Sleep Res. 2021;30(4):e13276. pmid:33529449
- 37. Hollifield M, Warner TD, Lian N, Krakow B, Jenkins JH, Kesler J, et al. Measuring trauma and health status in refugees: a critical review. JAMA. 2002;288(5):611–21. pmid:12150673
- 38. Mollica RF, Caspi-Yavin Y, Bollini P, Truong T, Tor S, Lavelle J. The Harvard Trauma Questionnaire. Validating a cross-cultural instrument for measuring torture, trauma, and posttraumatic stress disorder in Indochinese refugees. J Nerv Ment Dis. 1992;180:111–6.
- 39. Bech P, Kastrup M, Rafaelsen OJ. Mini-compendium of rating scales for states of anxiety depression mania schizophrenia with corresponding DSM-III syndromes. Acta Psychiatr Scand Suppl. 1986;326:1–37. pmid:3458353
- 40. Westergaard ML, Jensen RH, Carlsson J. Headache comorbidity in refugees and migrants with post-traumatic stress disorder. Cephalalgia. 2023;43(3).
- 41. Sonne C, Carlsson J, Elklit A, Mortensen EL, Ekstrøm M. Treatment of traumatized refugees with sertraline versus venlafaxine in combination with psychotherapy - study protocol for a randomized clinical trial. Trials. 2013;14:137. pmid:23663588
- 42. Nordbrandt MS, Carlsson J, Lindberg LG, Sandahl H, Mortensen EL. Treatment of traumatised refugees with basic body awareness therapy versus mixed physical activity as add-on treatment: Study protocol of a randomised controlled trial. Trials. 2015;16:477. pmid:26492879
- 43. Grootenboer EMV, Giltay EJ, van der Lem R, van Veen T, van der Wee NJA, Zitman FG. Reliability and validity of the Global Assessment of Functioning Scale in clinical outpatients with depressive disorders. J Eval Clin Pract. 2012;18(2):502–7. pmid:21223457
- 44. Atkinson TM, Rosenfeld BD, Sit L, Mendoza TR, Fruscione M, Lavene D, et al. Using confirmatory factor analysis to evaluate construct validity of the Brief Pain Inventory (BPI). J Pain Symptom Manage. 2011;41(3):558–65.
- 45. Barawi KS, Lewis C, Simon N, Bisson JI. A systematic review of factors associated with outcome of psychological treatments for post-traumatic stress disorder. Eur J Psychotraumatol. 2020;11(01):1774240.
- 46. Resick PA, LoSavio ST, Wachen JS, Dillon KH, Nason EE, Dondanville KA, et al. Predictors of Treatment Outcome in Group or Individual Cognitive Processing Therapy for Posttraumatic Stress Disorder Among Active Duty Military. Cognit Ther Res. 2020;44(3):611–20. pmid:35431370
- 47. Bradbury A. Mental health stigma: The impact of age and gender on attitudes. Community Mental Health Journal. 2020;56(5):933–8.
- 48. Nordbrandt MS, Sonne C, Mortensen EL, Carlsson J. Trauma-affected refugees treated with basic body awareness therapy or mixed physical activity as augmentation to treatment as usual - a pragmatic randomised controlled trial. PLoS One. 2020;15(3).
- 49. Sonne C, Mortensen EL, Silove D, Palic S, Carlsson J. Predictors of treatment outcomes for trauma-affected refugees - results from two randomised trials. Journal of Affective Disorders. 2023.
- 50. Lange AB, Mortensen EL, Palic S, Sonne CK, Carlsson J. Refugees versus persons family reunified to a refugee: A comparison of sociodemographic factors, trauma history, baseline psychopathology and treatment effectiveness. J Affect Disord Rep. 2023;13(10).
- 51. Riley RD, Hayden JA, Steyerberg EW, Moons KGM, Abrams K, Kyzas PA. Prognosis Research Strategy (PROGRESS) 2: Prognostic Factor Research. PLoS Med. 2013;10(2).
- 52. Marker I, Norton PJ. The efficacy of incorporating motivational interviewing to cognitive behavior therapy for anxiety disorders: A review and meta-analysis. Clin Psychol Rev. 2018;62:1–10. pmid:29727863
- 53. Dean S, Britt E, Bell E, Stanley J, Collings S. Motivational interviewing to enhance adolescent mental health treatment engagement: a randomized clinical trial. Psychol Med. 2016;46(9):1961–9.
- 54. Simoneau H, Bergeron J. Factors affecting motivation during the first six weeks of treatment. Addict Behav. 2003;28(7):1219–41. pmid:12915165
- 55. Bruhn M, Laugesen H, Kromann-Larsen M, Trevino CS, Eplov L, Hjorthøj C, et al. The effect of an integrated care intervention of multidisciplinary mental health treatment and employment services for trauma-affected refugees: study protocol for a randomised controlled trial. Trials. 2022;23(1):859. pmid:36209104