Refugee mental health: screening and care

A 34-year-old woman who arrived in the country four months ago comes to the primary care clinic with a headache every day, insomnia and fatigue. She speaks little of the local language, came with two children and does not know where her husband is. The physical exam is normal. The appointment lasts fifteen minutes, the queue is long and the temptation is to order a test, prescribe an analgesic and move on. But the question that changes the case is another one: what has this person lived through, and how has she been sleeping, thinking and feeling since then?

Refugee mental health is one of those topics where clinicians err on two sides. They may let a treatable disorder hidden behind physical complaints go unnoticed, or they may label as psychiatric illness a suffering that is expected. This text shows what epidemiology says, how screening works, which instruments have been tested, what helps in care and which biases get in the way of the consultation.

How much mental suffering exists

The question seems simple, but the numbers vary so much that it is easy to get lost. The most cited systematic review of recent years, by Blackmore and colleagues, included only studies with a clinical interview and a validated instrument, with at least 50 participants (Blackmore and colleagues, 2020). There were 26 studies, with 5143 adult refugees and asylum seekers in 15 countries (Blackmore and colleagues, 2020).

The result gives the order of magnitude. The prevalence of post-traumatic stress disorder was about 31 in every 100 people, that of depression also about 31 in every 100, that of anxiety disorders 11 in every 100 and that of psychosis about 1.5 in every 100 (Blackmore and colleagues, 2020). The authors note that there was great heterogeneity between studies in the estimates for post-traumatic stress, depression and anxiety (Blackmore and colleagues, 2020).

This heterogeneity is not a detail. Morina and colleagues reviewed 38 studies with 39518 displaced adults and refugees from 21 countries affected by armed conflict (Morina and colleagues, 2018). The highest prevalences were for post-traumatic stress, which ranged from 3 to 88 in every 100, depression, from 5 to 80, and anxiety, from 1 to 81 (Morina and colleagues, 2018). Only 12 original articles carried data on other mental disorders (Morina and colleagues, 2018).

For those who think like a clinician, this range teaches more than the average. It says that prevalence depends on who was studied, where, when and with what method. It is the same reasoning as pretest probability: before interpreting a screening result, you need a reasonable idea of how common the problem is in that population.

The same instrument, applied in two different places, will produce very different proportions of true positives. In a shelter for new arrivals from an active war zone, a positive result has a high chance of corresponding to a real disorder. In a routine appointment with someone settled for years, with a job and a support network, the same result deserves more caution. This effect is discussed in the same test in different populations, and it applies as much to a mental health questionnaire as to a blood test.

Another point that epidemiology helps to remember is that psychosis is rare in this population, in the range of just over one in every hundred adults studied (Blackmore and colleagues, 2020). This does not mean ignoring it, but it does mean that delirium, hallucination or major disorganization call for a broad assessment, including organic causes, before being attributed to trauma.

Time does not resolve it on its own

A common belief is that suffering passes once the person stabilizes in the new country. The evidence suggests caution. Bogic and colleagues reviewed the mental health of war refugees five years or more after displacement (Bogic and colleagues, 2015). There were 29 studies with 16010 people (Bogic and colleagues, 2015).

The variation between studies was enormous for depression, post-traumatic stress and anxiety, but the estimates usually fell in the range of 20 per 100 or more (Bogic and colleagues, 2015). Clinical and methodological factors explained much of the difference, and the higher-quality studies generally reported lower rates (Bogic and colleagues, 2015). The prevalences also depended on the country of origin and the host country (Bogic and colleagues, 2015).

This has two practical implications. The first is that a refugee who arrived many years ago may still have an active disorder, and the question about the past does not lose its validity over time. The second is that low-quality studies tend to overestimate the problem, which reinforces the importance of not labeling by group membership.

For the woman in the opening case, this means something concrete. She arrived four months ago, is separated from her husband and cares for two children alone. These data raise the chance of significant distress, but they do not say what the problem is. It could be depression, it could be post-traumatic stress, it could be grief, it could be a combination. It could also be a headache from another cause that deserves its own investigation. The clinical conversation is what separates these hypotheses.

The umbrella review by Turrini and colleagues adds an important correction. Although attention usually goes to post-traumatic stress, depression and anxiety were at least as frequent as it, reaching up to 40 per 100 asylum seekers and refugees in some studies (Turrini and colleagues, 2017). Anyone who looks only for trauma may miss the depression that is right in front.

Why screen and what can go wrong

Screening means asking systematically, with a short instrument, people who have not yet sought help for that problem. In refugees, the argument in favor is strong: distress is frequent, it often arrives disguised as a physical complaint and access to specialized care is usually difficult.

Magwood and colleagues did a scoping review on mental health screening in refugees and asylum seekers before departure and after arrival in the resettlement country (Magwood and colleagues, 2022). Of 25862 citations, 70 publications met the criteria: 45 described screening programs and 25 were instrument validation studies, and the authors characterized 85 screening instruments (Magwood and colleagues, 2022).

The review raises warnings that every service should consider. The authors note that it is important to ensure that information from a prior assessment does not lead to the rejection of vulnerable refugees because of mental health, nor because of the services offered in the destination country (Magwood and colleagues, 2022). They also recommend defining in advance how the information will be shared and used, because in a pilot some professionals did not know what the data would be for or with whom they could share it (Magwood and colleagues, 2022).

This point is ethical, but also clinical. If the person fears that the answer will affect their asylum claim, screening loses validity, because they answer what they think is safe. Explaining what the question is for, who will see the answer and what happens next is part of the instrument as much as the items are.

In practice, one simple sentence helps: explain that the questions are asked of everyone who arrives, that they serve to offer help and that the answers stay in the health record. If there is any limit to that confidentiality, it must be stated beforehand. This transparency is also a form of respect, and it usually improves the quality of the answers.

There is also the risk of screening without having anywhere to refer people. The same review points out that refugee health assessment programs need to be linked to the national health system to make sense (Magwood and colleagues, 2022). A positive result with no care pathway becomes just a label, something close to what is discussed in what overdiagnosis is not and in how harm reaches the patient.

Instruments that have been tested

Two instruments appear most often in recent literature, and both were designed to be applied by professionals who are not mental health specialists.

The first is the Refugee Health Screener. Kaltenbach and colleagues tested the 15-item version in 86 refugees newly arrived in Germany, as a self-administered questionnaire, and then conducted a clinical interview with 56 of them (Kaltenbach and colleagues, 2017). About half of the refugees had a positive result, indicating current mental health problems (Kaltenbach and colleagues, 2017). The instrument showed good feasibility, reliability and validity, detected post-traumatic stress, depression, anxiety and somatization, and a shorter 13-item version was equally valid (Kaltenbach and colleagues, 2017).

Bjärtå and colleagues applied the 13-item version to 510 refugees living in accommodation for asylum seekers in Sweden (Bjärtå and colleagues, 2018). They proposed cutoffs of 11, 18 and 25 for mild, moderate and severe symptoms, and confirmed the cutoff of 11 already described before (Bjärtå and colleagues, 2018). The authors argue that additional cutoffs help guide clinical decisions, separating those who need follow-up from those who need urgent referral (Bjärtå and colleagues, 2018).

The second is STAR-MH, developed in Australia to screen for major depression and post-traumatic stress in asylum seekers and refugees (Hocking and colleagues, 2018). The final instrument has nine items, two of which are immediate inclusion items, and takes a median of six minutes to apply (Hocking and colleagues, 2018). In the sample tested, 32 out of every 100 people had post-traumatic stress, depression or both according to the structured interview, and almost all the other diagnosed disorders appeared together with one of these two (Hocking and colleagues, 2018). With a cutoff of two points, sensitivity was 93 out of every 100 and specificity was 75 out of every 100 (Hocking and colleagues, 2018).

Instrument	Format	What was shown	Point of attention
Refugee Health Screener, 15 items	Self-administered or interview	Good validity; about half positive in new arrivals (Kaltenbach and colleagues, 2017)	A positive result indicates distress, not diagnosis
Refugee Health Screener, 13 items	Short version	Cutoffs for mild, moderate and severe (Bjärtå and colleagues, 2018)	Real use depends on time and language (Hagström and colleagues, 2024)
STAR-MH	Nine items, about six minutes	High sensitivity for depression and post-traumatic stress (Hocking and colleagues, 2018)	Lower specificity generates false positives

The table helps recall a basic rule of sensitivity, specificity and predictive values. A very sensitive instrument is great for leaving no one out, but it produces false positives. Therefore, every positive result needs a clinical conversation before it becomes a diagnosis.

The instrument in the real world

Having a good instrument does not guarantee that it will be used. Hagström and colleagues evaluated the implementation of the 13-item version in the health assessment of forced migrants at eight primary care centers in the Stockholm region (Hagström and colleagues, 2024). Use varied widely: three centers used the instrument with high frequency and five with low frequency (Hagström and colleagues, 2024).

The main barriers were language, lack of time and the nurses' distrust of the instrument's validity and usefulness (Hagström and colleagues, 2024). The facilitators were its availability in many languages and the perception that it importantly complemented the health assessment (Hagström and colleagues, 2024). One nurse described the questionnaire as an icebreaker, which appears in the study's own title (Hagström and colleagues, 2024).

Language deserves special attention. A translated questionnaire helps, but the conversation that follows needs a professional interpreter. The text on professional interpreters and ad hoc interpreters explains why using your child or a neighbor as an interpreter is risky, and the errors that happen in interpretation shows how meaning gets lost. In a subject such as trauma, violence or suicidal ideation, an ad hoc interpreter can prevent a truthful answer.

What helps in care

Screening only makes sense if there is something to offer afterwards. The good news is that there is evidence that psychosocial interventions work in this population.

Turrini and colleagues gathered 26 studies with 1959 participants in a systematic review with meta-analysis (Turrini and colleagues, 2019). Psychosocial interventions had a clinically significant beneficial effect on post-traumatic stress, depression and anxiety, with moderate-quality evidence (Turrini and colleagues, 2019). The benefit held at one month or more of follow-up (Turrini and colleagues, 2019). The earlier umbrella review already pointed to cognitive-behavioral interventions, especially narrative exposure therapy, as the most studied, with a positive result against inactive controls, but not against active controls (Turrini and colleagues, 2017).

One of the most useful findings for those working with few resources comes from a trial with Syrian refugees in the Netherlands. De Graaff and colleagues tested a brief psychological intervention delivered by peers, that is, by people from the same community trained for it (de Graaff and colleagues, 2023). There were 206 participants with psychological distress and functional impairment, divided between the intervention plus usual care and usual care alone (de Graaff and colleagues, 2023). At three months, the intervention group had a greater reduction in depression, anxiety, post-traumatic stress symptoms and problems identified by the participants themselves, but not in functional impairment (de Graaff and colleagues, 2023).

The lack of effect on daily functioning is a useful reminder. Symptoms can improve while life remains difficult for reasons that no psychological intervention solves alone, such as precarious housing, lack of work and uncertainty about the asylum claim. Mental health care for refugees usually needs coordination with social assistance, legal support and school, and the clinician helps a lot when they know where to refer each need.

This result expands what the clinician can offer. Not every refugee with distress needs a psychiatrist; many benefit from structured, brief interventions, as long as there is a clear path for those who worsen. What cannot be missing is risk assessment, discussed in what the conversation captures that the scale does not.

Children call for their own approach. Bürgin and colleagues describe that war produces acute stress responses, increased risk of specific disorders, distress from separation from parents and fear for one's own safety and the family's (Bürgin and colleagues, 2022). They advocate interventions at several levels, aimed at the child's needs, trauma-informed and oriented toward strengths and resilience (Bürgin and colleagues, 2022). Immediate interventions should guarantee basic physical and emotional resources so that the child regains external and internal safety, and screening and assessment should guide targeted interventions (Bürgin and colleagues, 2022).

For the general practitioner, this translates into a few simple questions to parents or caregivers: how the child is sleeping, whether they have started wetting the bed again, whether they are playing, whether they go to school, whether they are afraid of separating from adults. Marked changes in these points justify a more detailed assessment. And the parents' mental health counts: a depressed and lonely mother will have more difficulty offering the safety the child needs, which is one more reason to care for both at the same time.

Biases that get in the way of the consultation

The first bias is diagnostic overshadowing in reverse. Instead of attributing everything to mental illness, the clinician attributes everything to the body and treats only the headache. In populations where depression and anxiety are as frequent as post-traumatic stress (Turrini and colleagues, 2017), a repeated physical complaint with a normal exam should lead to questions about sleep, mood and worries.

The second is the opposite: calling any suffering a disorder. Refugees face real losses, legal uncertainty and family separation, and sadness in the face of that is not necessarily an illness. The text on diagnostic overshadowing shows how a label can hide another illness, including an organic cause.

The third is to assume that all refugees have severe trauma. The wide variation between studies (Morina and colleagues, 2018) and the fact that better studies find lower rates (Bogic and colleagues, 2015) show that belonging to the group does not define the diagnosis. The right question is individual, and it starts with who is in front of you: where they came from, where they passed through, how long ago they arrived.

The fourth is haste. Hearing the story in pieces, over several consultations, usually yields more than an interrogation on the first one. Narrative medicine helps remind us that the way a person tells what they have lived through is also clinical data.

The fifth is forgetting that the violence may be happening right now. Survivors of sexual violence in a crisis context need their own script, described in follow-up and mental health.

The sixth bias is forgetting the context in which the person lives today. Even far from war, many displaced people live in conditions that worsen suffering, such as crowded shelters, heat, little privacy and a lack of routine. The text about cholera, measles and malnutrition in conflicts shows how mental health comes along with the other demands of a crisis, not after them. Those who care for displaced populations also deal with heavy stories every day, and protecting the caregivers is part of keeping the service running.

Finally, it is worth recalling the sequence that usually works in the office. First, ensure communication, with a professional interpreter and enough time. Second, listen to the complaint that brought the person in and examine them with the same care given to any patient. Third, ask about sleep, mood, concerns and safety, with or without an instrument. Fourth, assess the risk. Fifth, agree on the next step clearly, with a follow-up visit scheduled. A well-conducted interpreted consultation makes each of these steps possible.

Frequently asked questions

How common is mental disorder among refugees?

In a review of 26 studies, about 31 in every 100 had post-traumatic stress and another 31 in every 100 had depression, with wide variation between studies (Blackmore and colleagues, 2020).

Does suffering decrease over time?

Not necessarily. Five years or more after displacement, rates still often stayed in the range of 20 per 100 or more (Bogic et al., 2015).

Which screening instrument should be used?

The Refugee Health Screener and the STAR-MH have been tested in this population and can be administered by non-specialists (Kaltenbach et al., 2017).

Is a positive result a diagnosis?

No. Sensitive instruments produce false positives, and every positive result needs clinical evaluation (Hocking et al., 2018).

What should be done after a positive screen?

Assess the risk, offer psychosocial intervention, which has demonstrated benefit, and refer when necessary (Turrini and colleagues, 2019).

Why does the interpreter matter so much?

Because language is one of the main barriers to using triage in practice (Hagström et al., 2024).

Take-home messages

• Post-traumatic stress and depression appear in about one third of the adult refugees studied, with wide variation (Blackmore et al., 2020).
• Depression and anxiety are at least as frequent as post-traumatic stress (Turrini et al., 2017).
• Suffering can persist for many years after arrival (Bogic et al., 2015).
• Triage requires explaining the use of the information and having somewhere to refer the patient (Magwood et al., 2022).
• Short instruments work, but a positive result calls for a clinical conversation (Hocking et al., 2018).
• Brief psychosocial interventions, including peer-led ones, reduce symptoms (de Graaff et al., 2023).

References

1. Bjärtå A, Leiler A, Ekdahl J, et al. Assessing Severity of Psychological Distress Among Refugees With the Refugee Health Screener, 13-Item Version. J Nerv Ment Dis. 2018. PMID 30256332.
2. Blackmore R, Boyle JA, Fazel M, et al. The prevalence of mental illness in refugees and asylum seekers: A systematic review and meta-analysis. PLoS Med. 2020. PMID 32956381.
3. Bogic M, Njoku A, Priebe S. Long-term mental health of war-refugees: a systematic literature review. BMC Int Health Hum Rights. 2015. PMID 26510473.
4. Bürgin D, Anagnostopoulos D, Board and Policy Division of ESCAP, et al. Impact of war and forced displacement on children's mental health-multilevel, needs-oriented, and trauma-informed approaches. Eur Child Adolesc Psychiatry. 2022. PMID 35286450.
5. de Graaff AM, Cuijpers P, Twisk JWR, et al. Peer-provided psychological intervention for Syrian refugees: results of a randomised controlled trial on the effectiveness of Problem Management Plus. BMJ Ment Health. 2023. PMID 36789918.
6. Hagström A, Hasson H, Hollander AC, et al. "Sometimes it can be like an icebreaker": A mixed method evaluation of the implementation of the Refugee Health Screener-13 (RHS-13). J Migr Health. 2024. PMID 39220097.
7. Hocking DC, Mancuso SG, Sundram S. Development and validation of a mental health screening tool for asylum-seekers and refugees: the STAR-MH. BMC Psychiatry. 2018. PMID 29548315.
8. Kaltenbach E, Härdtner E, Hermenau K, et al. Efficient identification of mental health problems in refugees in Germany: the Refugee Health Screener. Eur J Psychotraumatol. 2017. PMID 29163869.
9. Magwood O, Kassam A, Mavedatnia D, et al. Mental Health Screening Approaches for Resettling Refugees and Asylum Seekers: A Scoping Review. Int J Environ Res Public Health. 2022. PMID 35329237.
10. Morina N, Akhtar A, Barth J, et al. Psychiatric Disorders in Refugees and Internally Displaced Persons After Forced Displacement: A Systematic Review. Front Psychiatry. 2018. PMID 30298022.
11. Turrini G, Purgato M, Acarturk C, et al. Efficacy and acceptability of psychosocial interventions in asylum seekers and refugees: systematic review and meta-analysis. Epidemiol Psychiatr Sci. 2019. PMID 30739625.
12. Turrini G, Purgato M, Ballette F, et al. Common mental disorders in asylum seekers and refugees: umbrella review of prevalence and intervention studies. Int J Ment Health Syst. 2017. PMID 28855963.