Background
Post-traumatic stress disorder (PTSD) is a major health problem among refugees worldwide [
1]. A review by Fazel and colleagues [
2] showed that the overall prevalence of PTSD among refugees resettled in Western countries was about 9%, with substantial heterogeneity between studies. Although this overall percentage was not as high as that suggested by others [
3‐
5], Fazel and colleagues suggested that refugees resettled in Western countries are about ten times more likely to have PTSD than age-matched general populations in those countries. For example, in the Netherlands, the lifetime prevalence of PTSD is 7.4% for the general population [
6], whereas this is around 20% for refugees and asylum seekers from Iran, Afghanistan, and Somalia [
5].
While the proportion of refugees with PTSD is smaller after resettlement in the host countries, it remains relatively high compared with the general population [
5,
7]. Although this might indicate the chronicity of a severe mental illness, this persistently high prevalence seems to be at odds with the availability of effective forms of treatment for PTSD [
8,
9]. Since several PTSD treatments are known to be effective among diverse groups, the question arises as to why there is little change in the proportion of resettled refugees with PTSD, even several years after resettlement [
5].
A first possibility, or hypothesis, is that although the prevalence of PTSD remains high, it represents PTSD in different subjects. An almost unchanged proportion of refugees with PTSD over time does not necessarily mean that the positive cases are in fact the same persons. Late onset of PTSD several years after the traumatic events took place is highly probable [
10], and this must be taken into account when examining changes over time. Late-onset PTSD has been shown up to 14 years after the traumatic events [
11].
A second hypothesis is that the available treatment services are inadequately used by refugees with PTSD, implying underutilisation. Timely use of appropriate mental health care is considered necessary for recovery from PTSD. A prompt intervention based on cognitive behavioural treatment can relieve the complaints and prevent the development of PTSD [
12‐
14]. Dutch guidelines for general practitioners (GPs) recommend direct referral to mental health care (Dutch College of General Practitioners’ Standard M62) [
15]. The regulations for access to Dutch health care provide refugees, asylum seekers, and Dutch citizens with similar rights to primary health care and specialised mental health care through GP referrals. However, a distrust of mental health care, a lack of knowledge about mental health treatment possibilities, and language barriers might limit access to mental health care among this group of newcomers, which is unfamiliar with the new country’s health system. A study conducted in the United States showed that, in the years following resettlement, many refugees do not receive adequate care [
16]. Language barriers, acculturation issues, and cultural beliefs about several forms of health care contribute jointly to an access problem for this group.
Finally, a third hypothesis is that mental health care may not have had the expected positive effect on the course of mental health for refugees who used mental health care services. For refugees, the ineffectiveness of mental health care treatments for PTSD might account for long-lasting symptoms. Although current guidelines recommend trauma-focused psychotherapy for patients with PTSD [
17,
18] and effective forms of treatment for PTSD are available [
8,
9], very limited or no effectiveness has been reported for PTSD treatment specifically for asylum seekers and refugees [
19,
20]. Although two pilot studies demonstrated the feasibility and effectiveness of some trauma-focused approaches for treating PTSD in refugees [
21,
22], more evidence of treatment effectiveness is needed for this group.
The second and third hypotheses concern the use and effectiveness of mental health care. The role of mental health care utilisation in the course of mental health is embedded in the resettlement situation of refugees, and is also related to pre-resettlement events. Traumatic events preceding the flight, whether experienced or witnessed, are direct risk factors for the onset of mental illness [
23], as well as for its persistence [
7,
24]. Post-migration factors can also jeopardise mental health [
25]. Among post-migration factors, living difficulties related to employment, social and family networks, dealing with a new culture, and social position show a direct relationship to mental illness among resettled refugees [Lamkaddem et al., submitted]. Therefore, these pre- and post-migration factors must be taken into account when examining the role of mental health care utilisation in the course of PTSD.
A longitudinal study design was required to answer our research question on how to explain the persistently high prevalence of PTSD among resettled refugees. Therefore, we examined the course of PTSD and the related role of mental health care utilisation among a cohort of refugees from Iran, Afghanistan, and Somalia shortly after resettlement in the Netherlands.
Discussion
In this paper, we examined the course of PTSD among refugees after resettlement. The results show that there are two main explanations for the seemingly unchanged high rate of refugees with PTSD at a 7-year interval.
The first explanation concerns the onset and persistence of PTSD symptoms. Only half of the respondents who had PTSD during the second measurement also had PTSD during the first measurement. The other half of the respondents were new cases who developed PTSD later (i.e. between the two measurements).
The second explanation concerns the use of mental health care at T1. We saw that at T1, relatively few respondents with PTSD had had contact with a mental health provider (21%). We were more likely to see an improvement in PTSD symptoms during the second wave for those who had made use of mental health care during the first wave, which suggests that the low use of health care services also contributes to the remaining high prevalence of PTSD.
Our results on the course (onset and persistence) of PTSD are confirmed by other longitudinal studies among different ethnic groups (whether refugees or not). PTSD symptoms can persist several years after exposure to trauma [
32,
33], and can also have a delayed onset, whereby symptoms only become apparent several years after the traumatic events have taken place [
34,
10]. These results stress the importance of a longitudinal cohort design when examining the course of PTSD. Our study shows that late onset of PTSD is an important reason for the ongoing high prevalence of PTSD after resettlement.
Our results on the effectiveness of mental health care provide evidence for the effect of mental health care utilisation for this group. In refugees who had used mental health care, PTSD symptoms generally improved. These findings are confirmed by the few studies that specifically addressed this topic among refugees. One Dutch clinical study reported that 73% of patients diagnosed with PTSD no longer met the criteria for the diagnosis 6 months after treatment, whereas 90% of those who had refused treatment were later diagnosed with PTSD [
35]. Therefore, for refugees, accessing and using mental health care is an important precondition for improving PTSD status.
While mental health care seems to be beneficial to PTSD recovery, our results show that the percentage of refugees who actually use mental health care services is relatively low (i.e. one fifth of those reporting PTSD symptoms). Although these results are similar to studies in other countries of resettlement, few have specifically examined Dutch mental health care services. The last study on this topic was conducted among Iraqi asylum seekers [
36], and indicated a large unmet need for mental health care: over 90% of asylum seekers with a psychiatric disorder did not visit a Dutch mental health care service. In the present study among refugees with PTSD, this percentage was almost 80%. The reasons for the low use of mental health care services among refugees need to be investigated further. Barriers to accessing mental health care can be due to low language fluency [
37], lack of knowledge on existing treatments [
38], and culturally and/or linguistically unsuitable forms of information about the mental health care supply. Just as for other types of migrants [
39], these factors might in turn affect the perceived mental health care needs of refugees, and impede access.
The present study has several limitations. First, only 172 respondents took part in both waves of the study. This might affect analyses on the extent to which pre-migration factors (i.e. traumatic events before the flight) impact the course of PTSD after resettlement. The same analyses among a larger sample might show a significant (negative) association between the number/severity of traumatic events and PTSD recovery. Also, it was not possible to accurately assess the extent to which those who made use of mental health care during the first wave were more likely to see an improvement in their PTSD symptoms (the large 95% CI reflects the uncertainty of the estimation).
Second, the relatively high attrition rate between the first and second wave might have induced some bias. However, non-inclusion was mainly attributable to remigration (30% of the initial sample), which was unavoidable in these types of settings. The non-response analysis presented in Table
1 shows that the socio-demographic characteristics of our study sample do not differ from the initial sample. Moreover, given the objective of our research and the focus on the explanatory factors rather than on the prevalence of PTSD, we do not expect the attrition to have had a considerable impact on our results.
Third, our study included three different ethnic groups, which represent only part of the refugee population in the Netherlands. The extent to which these results can be generalised to all refugee groups in the Netherlands and elsewhere remains to be confirmed in larger research samples.
Fourth, the measurement of PTSD is not a clinical one, and the questionnaire used for this purpose must be seen as only an approximation of PTSD. However, studies have successfully used the HTQ (or parts of it), yielding results comparable to clinical diagnoses [
40,
41].
Finally, the information on the use of mental health care services was not directly linked to the diagnosis. We have no evidence that the reported contacts with mental health care providers reported by refugees with PTSD symptoms were in fact specifically for PTSD treatment. Therefore, we can draw no conclusions on the effectiveness of a specific PTSD treatment, but merely on the association between the use of mental health care in general and the course of PTSD symptoms.
Conclusions
Despite these limitations, the explanations presented here for the seemingly unchanged high rates of PTSD offer interesting insights for further research on the mental health of refugees. The findings emphasise the need for primary care providers to follow existing guidelines on quick referral to mental health care for patients presenting with PTSD symptoms, and underline the possibility of late-onset PTSD. Finally, in addition to the use of mental health care, the results show the importance of improvements in contextual factors (e.g. in employment, social/family networks, becoming familiar with the new culture, and social position) on the course of PTSD.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
ML carried out the study, performed the statistical analyses, and drafted the manuscript. MLE and KS participated in the design and coordination of the study, and helped draft the manuscript. AG, WD, and MO contributed to drafting the manuscript. All authors read and approved the final manuscript.