Introduction
The world is now facing the highest number of forcibly displaced people ever recorded—65 million people. In 2015, a year when a high number of migrants arrived in Europe, Sweden received 163,000 applications for asylum. The greatest share of these applicants originated from Syria [
1].
Potentially traumatic experiences, e.g., war at close quarters, are common among refugees coming from war-torn regions [
2,
3], and many refugees may suffer from mental ill health due to these experiences [
2,
4,
5]. Prevalence of mental ill health such as anxiety, depression and PTSD seems to be high among war refugees [
2,
4‐
7]. Specifically, high prevalence of mental ill health has been found in a large random sample of Syrian refugees resettled in Sweden [
2]. Given that health, as defined by WHO [
8] is an integrated concept including mental aspects as well as physical and social well-being, a too narrow focus on mental health may, however, conceal a broader understanding of refugee health and how health impacts the lives of refugees in a broader sense. Health-related quality of life (HRQoL) is a measure closely related to the WHO’s definition of health as it takes into account several dimensions of health and thus warrants to be a target of examination in this context.
Refugees’ health is a highly gendered area. Women are exposed to particular health risks in the migration process such as sexual violence and exploitation. Previous research has shown that refugee women report both more physical and more mental problems than refugee men [
4,
9]. Research focusing on refugees’ health therefore needs to include specific gender analyses [
10]. Besides gender, age is important for health among refugees. Older refugees report more mental ill health [
2,
9], as well as worse physical functioning [
11] than younger refugees.
Social support and networks are, in general, considered important protective factors relating to both mental and physical health [
12,
13]. Social support has been described as the social resources that persons perceive to be available or that are actually provided to them by nonprofessionals in the context of both formal support groups and informal helping relationships [
14]. Refugees’ social networks are often upheaved due to migration, and rebuilding social networks and ties may prove to be challenging.
There are few previous studies published regarding the HRQoL of refugees resettled in high-income countries. Results from a study in Finland among older Somali refugees showed that they had a lower HRQoL than the matched Finnish host population. Somali refugees reported more on the dimension of anxiety/depression, whereas the Finnish control group reported more on pain/discomfort [
15]. The HRQoL in elderly has also been investigated among Iranian refugees resettled in Sweden indicating that Iranian refugees in Sweden reported higher HRQoL than Iranians living in Iran, but lower than native Swedes living in Sweden [
16].
Today, there is a lack of robust knowledge about the general health status of refugees. In particular, knowledge about the health status of Syrian refugees resettled in a high-income country is scarce. In order to provide effective and useful health-related interventions adapted for men, women or for different age groups, it is important to assess the health status in this refugee population, how it is distributed within the group and other factors it could be influenced by. Therefore, the aim of this study was to assess the HRQoL among Syrian refugees resettled in Sweden, in total and stratified by sex and age group, and also to investigate whether there is an association between HRQoL and sex, age, educational level, cohabitation, place of residence, and social support in this group.
Discussion
There is currently a lack of knowledge regarding the health status among refugees resettled in high-income countries. The main aim of this study was, thus, to assess the level of HRQoL among Syrian refugees resettled in Sweden. In the same line, the distribution of problems in the five different dimensions of the EQ-5D in this population was examined. We also wanted to investigate possible differences in HRQoL between women and men and across age groups, and explore whether educational level, cohabitation, place of residence after resettlement, and social support were associated with HRQoL. This is the first study, to our knowledge, that investigates HRQoL among Syrian refugees in a high-income country, using a highly standardized measure of HRQoL, EQ-5D-5L, which provides the possibility to compare the health status of this refugee population to other refugee and non-refugee populations.
Depression/anxiety and pain/discomfort were the two particular dimensions in which most health problems were reported in this study population. A majority of the study population reported some level of problems in these two dimensions, and about 10% reported either severe or extreme problems. Women had an increased risk of experiencing problems with mobility and pain/discomfort compared to men. These results suggest that predominant health problems among the study population appear to be those which are generally related to exposure to traumatic experiences and distressful life conditions, i.e., mental health [
4] and pain-related distress [
23]. As it has been shown in previous research, these health sequelae of refugee-related exposures seem to be persistent long after resettlement [
4].
Our results show that the mean index value of the whole population was 0.754. On average, men and younger individuals had a higher level of HRQoL than women and older individuals. Interestingly, the results of the stratified analyses on both sex and age revealed that the highest mean HRQoL index value in the study population was found among the women in the youngest age group. This indicates that the overall low HRQoL among women in the study population could be ascribed to poor health status among women in the older age groups, and possibly suggests that there may be a steeper age-related decline in HRQoL among women compared to men. This, however, needs to be more closely investigated by means of longitudinal studies. Moreover, the results of the regression models corroborated the patterns of differences between men and women and across age groups, beyond which also cohabitation and social support emerged as correlates of HRQoL, indicating these factors embedded in the living conditions of resettled refugees assert an impact on the health status of refugees.
Our study population consisted of Syrian refugee women and men who had been resettled in Sweden since 3 to 5 years. In comparison with the most recent data from the Swedish general population in 2002, the mean HRQoL index value was lower for both women and men in our study population (0.797 vs. 0.735 for women and 0.841 vs. 0.765 for men, respectively) [
18]. The two EQ-5D dimensions pain/discomfort and anxiety/depression were also the two dimensions accounting for the most frequently reported problems among the Swedish general population, which indicates that pain/discomfort and anxiety/depression may act as key drivers of low HRQoL across different non-patient populations. In previous studies from Sweden and Finland, comparing HRQoL between elderly Iranian migrants/Somali refugees and the Swedish/Finnish host population, similar results were found [
15,
16]. Sex differences in HRQoL were also shown in a study with data from Sweden, where female Iranian migrants had lower scores in all dimensions compared to Swedish women, whereas male Iranian migrants had lower score in six of the eight dimensions measured by SF-36 [
16]. These findings are, furthermore, in line with previous research comparing health between immigrants born outside EU, not specifically refugees, and the Swedish host population, where differences in reported poor health were multifold [
24]. Taken together, our results indicate that worse health among refugee populations compared to host society’s general population may not merely be limited to mental health but seem to extend to health status in general and HRQoL.
Furthermore, this study provides some evidence of the importance of social support for overall health among refugees. This is in line with previous research among general population [
12,
13], immigrants in Sweden [
25] and refugee torture-survivors in Denmark [
26], as well as a recently published study among Syrian refugees from our research group [
22]. The stress buffering model [
16] which suggests that supportive social networks enhance individuals’ coping with stressful life events to buffer against the development of stress-related psychopathology, in light of our results, may be extended to be applicable to HRQoL among refugee populations that face multiple pre- and post-migratory stressful living conditions.
Sex and age are well-known determinants of health, where women tend to have worse health but live longer than men, and older tend to have worse health than younger [
27‐
29]. Similar associations have also been shown regarding HRQoL [
30]. Worth highlighting is, however, the extremely elevated risk of poor health found among the oldest age group in most of the measured dimensions of health of this study. Another important and somewhat surprising finding in this study is the lack of importance of educational level for HRQoL. Education is one of the measures used for socioeconomic position, which often is found associated with different outcomes in health [
31]. As education is expected to contain an individual’s potential for both income and employment, it has been assessed as a good marker for an individual’s living conditions, which in turn would influence health [
31]. In contrast, a study by Porter and Haslam [
9] found pre-migration high educational level associated with mental ill health, which could be explained by the higher likelihood of loss of status in the host country implied in pre-arrival higher social position. In regard to HRQoL, our results neither show patterns confirming to a positive or negative association between education level and health status. This could, perhaps, be viewed as either suggesting that high education does not function as a marker for socioeconomic position among newly resettled refugees or that other refugee-related living conditions such as pre-migratory trauma, and post-resettlement social support override the potential effect of educational level on health.
Whereas sex, age, and cohabitation are demographic and inherent subject factors, our study shows social support, a modifiable factor, to also be relevant to refugees’ overall health. Facilitating, mobilizing and enhancing refugees’ social support may therefore be considered a target of public health-level interventions. In contrary, it is worth pointing out that policies and regulations that constrain and impede access to close supportive relationships, such as policies restricting family reunification, may risk impacting negatively on refugees’ resources for HRQoL.
Our overall results concerning long-lasting health problems among the study population indicate that there is a profound need for policies and interventions promoting refugees’ health, e.g., allocation of resources to specialist care for rehabilitation of trauma-afflicted refugees. Given the large scale of the health challenges faced by refugees as indicated by our results and previous research [
2], however, there is also a need for public health-level and scalable health-promoting policies that prevent the health of refugees from deteriorating in post-resettlement. This is particularly important given the complexity and persistence of health problems among those refugees who have been through severe or multiple traumatic events [
32].
Strengths and limitations
A unique strength of this study, one of the larger studies of HRQoL within this research area, is self-reported data from a large random sample of Syrian refugees selected from a complete and known sample frame. The study population consists of refugees who are generally considered hard to reach populations [
33], and this is reflected in the response rate of 30.4%. Although within-subject associations analyses have been suggested to be less prone to non-response bias [
34,
35], a low response rate might imply a risk of selection bias leading to, i.e., an overestimation of HRQoL [
36]. Due to this important limitation, which is inherent in research with hard to reach populations, the risk of bias in estimating finite population characteristics should be acknowledged. Although our previous studies have shown that the socio-demographic characteristics of the sample corresponds closely to that of the randomly selected sample frame from the target population, the results should be viewed cautiously in regard to its generalizability to other refugee populations.
Furthermore, given that the study has a cross-sectional design, causal directions in association analyses should not be assumed, although the socio-demographic variables could be viewed as antecedents in this regard as they remain non-modifiable.
Another strength of our study is the use of validated, standardized instrument for assessment of HRQoL and also for assessment of social support. However, previous comparisons of HRQoL between countries have suggested that the value might differ depending on the norm-data used [
37]. Furthermore, information about the socio-demographic variables, except civil status, was retrieved from national, high-quality registers, reducing the risk of information bias. Finally, the assessment lacks data on EQ-5D VAS due to technical difficulties. Although this constitutes a limitation, on basis of previous research on correspondence between index and VAS values [
38] it is possible that the available data approximate the evaluation that VAS could have provided.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.