Health and health-related behaviours in refugees and migrants who self-identify as sexual or gender minority – A National population-based study in Sweden
Introduction
- Aldridge RW
- Nellums LB
- Bartlett S
- et al.
However, the reverse is observed in refugees. Being a refugee is associated with increased short- and long-term risk of poorer mental and physical health, even several years after settlement in the host country.
- Blackmore R
- Boyle JA
- Fazel M
- et al.
,
,
- Bränström R
- Hatzenbuehler ML
- Pachankis JE.
are more likely to engage in adverse health-related behaviors (for example, substance use and risky sex),
- Marshal MP
- Friedman MS
- Stall R
- et al.
and experience the co-occurrence of both.
- Bränström R
- Pachankis JE.
They are also at increased risk of exposure to violence.
- Blondeel K
- de Vasconcelos S
- García-Moreno C
- Stephenson R
- Temmerman M
- Toskin I.
However, this evidence is largely based on studies including White individuals and mostly originate from the USA. Health in individuals who identify as being both ethnic minority and sexual/gender minority (EM-SGM) i.e., dual minority identities are limited. These studies on health in EM-SGM individuals are often smaller in size, focusing on select ethnic minorities (EM) and rarely use national population-based data, especially in the European context.
- Toomey RB
- Huynh VW
- Jones SK
- Lee S
- Revels-Macalinao M.
,
- Abu-Ras W
- Suárez ZE
- RRi Breiwish
These studies also mostly originate from the USA, with a different healthcare system and distribution of ethnic minority groups making it incomparable to the European context.
- Fox KR
- Choukas-Bradley S
- Salk RH
- Marshal MP
- Thoma BC.
,
- Lett E
- Dowshen NL
- Baker KE.
,
- Schuler MS
- Prince DM
- Breslau J
- Collins RL.
Further, these studies show contradictory results. While most show evidence for worse mental health in EM-SGM individuals compared to their White-SM peers, others have found the opposite.
- Lett E
- Dowshen NL
- Baker KE.
,
- Burns MN
- Ryan DT
- Garofalo R
- Newcomb ME
- Mustanski B.
Additionally, many of the studies have focused on younger EM-SGM individuals.
- Mongelli F
- Perrone D
- Balducci J
- et al.
Refugees who identify as SGM could be at an even higher risk for adverse health because of past experiences of violence and mental ill-health associated with migration combined with racism and stigma in their home and host countries. There is ample evidence for discrimination including racism towards refugee and migrant individuals, including those of Middle-Eastern background in Sweden (for example, in accessing the labour market and housing).
,
- Agerström J
- Björklund F
- Carlsson R
- Rooth D-O.
- White LC
- Cooper M
- Lawrence D.
,
- Fox SD
- Griffin RH
- Pachankis JE.
To our knowledge, no study has examined health in SGM refugees in a nationally representative sample including comparisons with both heterosexual and SGM White and migrant populations. Including these different groups enables comparing risks between these groups of migrants who have migrated for potentially different reasons. Limited research shows that transgender individuals have increased risks for discrimination and mental ill-health compared to cisgender peers. However, the evidence on health in EM and refugee transgender individuals is limited due to very small numbers and lack of data.
- White LC
- Cooper M
- Lawrence D.
,
- Su D
- Irwin JA
- Fisher C
- et al.
These multiple minority identities need to be investigated jointly to fully understand their impact on health inequalities, and it is crucial to investigate them including all identity variables in the same model and testing for interactions between them and the outcomes of interest.
- Bauer GR
- Churchill SM
- Mahendran M
- Walwyn C
- Lizotte D
- Villa-Rueda AA.
Identifying and examining health in marginalized groups is essential as it can aid in designing interventions that can be implemented earlier when refugees and migrants arrive in a host country. Mental health can also affect integration
- Bakker L
- Dagevos J
- Engbersen G.
resulting in long-term consequences (e.g., future financial situation and educational attainment).
The main aim of this study was to use an intersectional framework to examine whether individuals with dual ethnic (refugee or migrant) and sexual minority identities were more likely to have poorer mental and general health, worse health-related behaviours and exposure to violence in comparison, to their heterosexual Swedish- and western-born cisgender peers, in a national population-based sample. Additionally, we also examined whether these associations differed between transgender and cisgender individuals with refugee or migrant background compared to Swedish- and western-born peers.
Results
Health and health-related behaviours
Table 2Associations between sexual/gender and ethnic (including migrant and refugee) identities and binary health and well-being in 157,414 individuals aged 16-84 years from the Swedish National Public Health Survey (2018 and 2020). Estimates are from multiple logistic and linear regression models (models adjusted for sex, age, and educational level).
Text in bold indicates estimates with 95% CI that do not include 1 2 Short version of the Warwick Edinburgh Mental Well-Being Scale (WEMWBS) 3Assessed using the Kessler-6 and GHQ-5 instruments.
Table 3Associations between sexual/gender and ethnic (including migrant and refugee) identities and experiences of violence in 157,414 individuals aged 16-84 from the Swedish National Public Health Survey. Estimates are from multivariable logistics regression models (models adjusted for sex, age, and educational level).
Text in bold indicates estimates with 95% CI that do not include 1.
Figure 1Odds ratios for health and health-related behaviours based on dual sexual/gender and ethnic (migrant and refugee) identities in 168 952 individuals aged 16-84 years who answered the Swedish National Public Health Survey in 2018-20.
Note: Estimates are from multivariable logistic regression models with adjustment for sex, age and educational level.
White-SGM (OR3·39, 3·02-3·82) and refugee-SGM (OR2·42, 1·44-4·08) groups had higher odds ratios for suicidal thoughts compared to White-heterosexuals. All three SGM groups [White, (OR3·84, 3·23–4·55), migrant, (OR2·82, 1·30–6·15), and refugee, (OR2·07, 0·95–4·55)] had substantially higher odds for suicide attempts compared to White-heterosexuals.
In general, ethnic minority (regardless of sexual identity) groups were less likely to report risk alcohol and drug use compared to their White peers. For example, risk alcohol consumption was significantly lower in migrant-heterosexuals (OR0·19,0·15-0·24), migrant-sexual minority (OR0·21,0·09-0·51), and refugee-heterosexuals (OR0·40,0·33-0·49). However, migrants and refugees (heterosexual and SGM) were more likely to report risk gambling.
Experience of different types of violence and discrimination
Sensitivity analysis
Supplementary Tables 4 and 5 display the distributions of all outcomes according to gender and the more detailed sexual identities (heterosexual, homosexual and bisexual identities), respectively. Compared to White cisgender individuals, all other transgender groups had significantly higher percentages of individuals reporting worse general health, mental ill-health (including suicidal ideation and attempts) and exposure to violence and discrimination. In general, greater percentages of bisexual individuals reported mental ill-health compared to their heterosexual and homosexual peers (with the highest percentages observed in the White bisexual group). While all transgender groups had increased odds for risk gambling, the highest odds were observed for refugee transgender individuals (White: OR1·48, 0·66–3·31, migrant: OR1·42, 0·72-2·82 and refugee OR8·62, 1·94-38·40). Transgender individuals also had higher odds for exposure to physical violence compared to cisgender peers (White transgender: OR3·16 (1·86–5·36), transgender migrants: OR6·31, 2·75-14·52 and transgender refugees OR7·46, 2·97-18·70) (Supplemental Table 7).
When examining differences in associations between SGM subgroups, bisexual individuals had the highest odds for mental ill-health and poorer general health [for example OR2·32 and OR1·86 for mental ill-health and OR2·49 and OR1·89 for general health in White-bisexuals and White-homosexuals, respectively, and compared to White-heterosexuals individuals (Supplemental Table 8)]. For migrant and refugee bisexuals vs. homosexuals, the results were mixed. For health and health-related behaviours, migrant homosexuals had higher odds ratios than bisexuals. For example, migrant homosexuals (OR3·25, 0·89-11·83) had higher odds ratios than migrant bisexuals (OR0·36, 0·11-1·22) for drug use. For refugees, we found the opposite. Refugee homosexuals (OR1·39, 0·60-3·23) had lower odds ratios than refugee bisexuals (OR3·82, 1·90-7·68) for suicidal ideation.
Discussion
This study found: 1) White-SGM individuals had consistently higher odds for mental ill-health, poorer wellbeing and general health compared to White heterosexual peers. However, findings were not consistent for EM-SGM individuals. 2) There was a marked difference in the observed pattern for health-related behaviours between White- and EM-SGM groups. 3), Both White and EM transgender individuals had higher odds for exposure to physical violence. White Transgender individuals had higher odds for all adverse health outcomes and all types of violence but not for adverse health-related behaviours. Refugee transgender individuals had higher odds for risk gambling. 4) White bisexuals had higher odds for poorer general health, mental ill-health, and drug use compared to White homosexuals. 5). In general, there were no consistent differences in health and health-related behaviour between migrant and refugee bisexuals and homosexuals
To our knowledge, this is the first study to separately examine health and health-related behaviours in SGM migrant and refugee individuals including transgender individuals in a national population-based sample.
- Bränström R
- Pachankis JE.
,
It also confirms results from the only study on SM-migrants to Sweden which found that being Nordic-born and SM is a stronger risk factor for mental ill-health including suicidal ideation and attempts than being a migrant and SM. However, this study did not include refugees and asylum seekers, only examined mental health, with a smaller sample that was drawn from a regional cohort.
- Clark K
- Björkenstam C
- Kosidou K
- Björkenstam E.
The only other Swedish study to examine mental health in foreign-born individuals residing in Sweden included a sample of 247 SM men, found that country-of-origin stigma was associated with poorer mental health.
- van der Star A
- Bränström R
- Pachankis JE.
This study did not have a heterosexual comparator group, nor did it distinguish between migrants and refugees.
- Aldridge RW
- Nellums LB
- Bartlett S
- et al.
and worse health among EM individuals.
Our findings indicate the complexity of these associations; health, health-related behaviours, and exposure to violence in EM individuals potentially depend on their status as migrants or refugees further compounded by their sexual and gender identities. For example, migrants and refugees were less likely to report risk behaviours (alcohol and drug use) but mental and general health differed by their sexual identity. Refugees as a group have higher risks for poorer health compared to the host population.
- Blackmore R
- Boyle JA
- Fazel M
- et al.
However, our results show that while heterosexual refugees reported worse general health and risk gambling, refugee SGM individuals reported higher odds for suicidality. Importantly, regardless of sexuality, being a refugee is significantly associated with exposure to physical violence and discrimination (also seen in all SGM groups in this study).
- White LC
- Cooper M
- Lawrence D.
Our study partially confirms these results (higher odds for suicidal ideation and all types of violence). Unexpectedly we did not find evidence for worse general and mental health. We also corroborate earlier findings that White transgender individuals have poorer health compared to their cisgender peers.
- Dhejne C
- Van Vlerken R
- Heylens G
- Arcelus J.
Few studies investigated refugee and migrant transgender health, but the existing research reports high rates of mental ill-health and experiences of violence but findings are based on smaller non-probability samples.
- Gowin M
- Taylor EL
- Dunnington J
- Alshuwaiyer G
- Cheney MK.
In this study, we found that transgender refugees have significantly poorer general health, not reported before. A previous study found White transgender youth have a higher risk for problem gambling compared to cisgender youth.
- Rider GN
- McMorris BJ
- Gower AL
- Coleman E
- Eisenberg ME.
Our study is the first to report higher odds of problem gambling in refugee transgender individuals. Lastly, our results are in line with previous Swedish studies that showed higher odds for risk alcohol-use and substance use in sexual minority individuals.
- Bränström R
- Pachankis JE.
Nonetheless, country of birth is considered a reasonable and often used substitute in the absence of self-identified ethnicity and is widely used in epidemiological studies in Europe.
- Khanolkar AR
- Sovio U
- Bartlett JW
- Wallby T
- Koupil I.
,
- Kist JM
- Smit GWG
- Mairuhu ATA
- et al.
,
- Hussen HI
- Persson M
- Moradi T.
However, we recognize that using country of birth as a proxy has certain limitations; individuals born in Sweden are categorised as Swedish, even though they may identify as EM. Further, distinct EM groups such as the Kurdish cannot be identified by country of birth. We acknowledge the Swedish- and Western-born group includes individuals who do not identify as White but we nonetheless expect the overwhelming majority in this category to be of White/ European-origin. We had to combine asylum seekers, persons in need of subsidiary protection and refugees in the same group as it is impossible to distinguish these categories of individuals in the dataset. Nonetheless, the distinction between these groups is not always clear and they are often used interchangeably.
- Bradby H
- Humphris R
- Newall D
- Phillimore J
,
- Derluyn I
- Mels C
- Broekaert E.
Lastly, the framework (developed in collaboration with the Swedish Migration Agency) to ascertain and distinguish migrant and refugee individuals was robust and used key variables like birth country, migration year, refugee status in high quality national registers alongside historical trends in national migration policy.
A low response rate, common with national health surveys of this kind is a limitation, but the use of population weights helps compensate for non-response, and the sample remains nationally representative. Lastly, the 2020 survey did not include any questions related to the covid-19 pandemic. There is a possibility that some participants might have been affected by covid-related issues but unfortunately, we cannot ascertain this impact in our study, which needs to be investigated in future studies.
- Raybould JN
- Larkin M
- Tunney RJ.
This is relevant to public health policy as gambling is rarely examined and discussed in studies on refugee/migrant health.
We hypothesized that EM-SGM individuals would have worse health due to the additive effects of two or more minority identities but found limited evidence for this. The few studies that examined health in refugee SGM individuals suggest a cumulative effect of being both a refugee and SGM, but evidence is scarce. Our findings highlight that the problem is likely more complex than a gradient effect and factors other than minority status (for e.g., expectations on equal treatment in society) might affect mental health.
This model hypothesizes that poorer health outcomes in individuals with one or more minority identities results from multiple and repeated exposures to stigma, discrimination and microaggressions. This may explain the observed higher odds for violence and discrimination found in all minority groups including the White-SM, EM-heterosexual and dual minority groups in this study.
- Matlin S.A.
- Depoux A.
- Schütte S.
- et al.
This could potentially explain some of our results as the EM-SGM group are potentially more exposed to a greater burden of discrimination associated with their ethnicity and sexuality leading to poorer health, compared to heterosexual peers. These mechanisms can be appropriately investigated in longitudinal data which is currently unavailable in Sweden.
,
- Bauer GR
- Churchill SM
- Mahendran M
- Walwyn C
- Lizotte D
- Villa-Rueda AA.
This framework theory helps understand how unique and multiple social identities (like ethnicity, gender, sexuality, faith belief etc.) intersect at the individual-level, reflecting various multiple and reciprocal systems of discrimination that impact health in individuals with ≥2 minority identities.
,
- Bauer GR
- Churchill SM
- Mahendran M
- Walwyn C
- Lizotte D
- Villa-Rueda AA.
Thus, each individual experiences their own unique forms of discrimination and oppression associated with their identities. These social identities are not isolated from one another or simply additive but are interdependent and mutually constitutive, i.e., they should be examined jointly to better understand the interdependence of minority identities. In Sweden, research has primarily examined health associated with one minority identity (most often ethnicity) and disregarded the potential impact of intersecting multiple minority identities. The intersectionality theory can explain some of our results, including contradictory findings. Minority groups may be differently impacted based on their unique intersecting minority identities and the environments they live in, including the host country’s policies. For example, while White-SGM individuals had increased risk for all adverse health-related behaviours except risk gambling, the opposite was observed in EM-SGM individuals. This could reflect differences in coping mechanisms and/or recreational activities intrinsically associated with unique identities. White-SGM individuals are more likely to report excessive alcohol and drug use, often a coping mechanism to deal with discrimination and stress associated with being SGM,
- Marshal MP
- Friedman MS
- Stall R
- et al.
,
- Kalb N
- Gillis R
- Goldstein A.
and non-White individuals potentially use other forms of more culturally accepted coping mechanisms like gambling. Lived experiences and their intrinsic associations with unique identities and consequences on health can be further impacted by wider society and the host country’s health policies. Currently, Swedish policies on gambling are more general without any specific guidelines related to ethnic and/or sexual identities and our results suggest that this might be relevant in future policies.
Migrant and refugee individuals who may have experienced terrible events in their home countries (for example, discrimination associated with SGM identities or experiences related to war/conflict), might feel relieved and safe after arrival in a host country. This differs from individuals who grow-up and live in a country like Sweden with liberal SGM rights but still experience discrimination, which can lead to disappointment and depression.
We hypothesize that the exposure of interest i.e., ethnic and sexual identities is conceptualized within individuals before the occurrence of outcomes (i.e., temporality). Nonetheless, as this is a cross-sectional study, we do not interpret findings as an indication of causality. This highlights the need for longitudinal data on individuals with self-identified ethnicity and sexuality which currently does not exist in Sweden.
- Satinsky E
- Fuhr DC
- Woodward A
- Sondorp E
- Roberts B.
,
- Alencar Albuquerque G
- de Lima Garcia C
- da Silva Quirino G
- et al.
Barriers include language (inability to speak Swedish), stigma, low awareness and different help-seeking behaviours in refugees
- Satinsky E
- Fuhr DC
- Woodward A
- Sondorp E
- Roberts B.
and heteronormative attitudes towards sexual minorities.
- Alencar Albuquerque G
- de Lima Garcia C
- da Silva Quirino G
- et al.
There is thus good reason to assume that this access and use of care might be even lower in SGM migrants and refugees. Hence, there is need for more research and practices into increasing access to care for this population as well as developing interventions specifically tailored to meet their needs. To be able to fully investigate these differences in the future more detailed data sets including ethnicity and sexual and gender minorities are warranted. Future public health policy needs to address existing barriers such as improving access to healthcare services and training of professionals including being made aware of difficulties faced by SGM individuals. A specific focus should be on risk gambling. Further, preventive measures to reduce exposure to violence and discrimination in both SGM- and EM-groups. Finally, there is a need to involve SGM groups in informing public policy since their views on these health disparities are often not considered.
