Mental health service use among refugee children, other migrant children, and children born in Sweden : what are the differences and why does it differ?

Abstract

Background: Research shows that migrant children, and especially refugee migrant children, are at increased risk of poor mental health. Given that mental disorders are the leading cause of disability among children and youth globally, and that many children and adolescents with mental disorders do not reach services for their mental health conditions, research into utilization among potentially vulnerable groups is much needed. Emerging evidence suggests discrepancies between need and service receipt among migrant children and youth in high-income countries, but more research is required to understand which factors are most important to drive differences, and how utilization differs among subgroups of migrants.

Aim: To explore differences in utilization of mental health services among migrant children and Swedish-born children, and to understand possible explanations for these differences.

Methods: Using a mixed-methods approach, utilization was measured through register-based epidemiological investigations, and barriers to care were further explored using individual in-depth interviews with migrant parents whose children were in contact with mental health services. For the epidemiological register-based studies (study 1-3), we used a database of linked registers to longitudinally study regional cohorts of all children and youth in the Swedish capital Stockholm, in terms of their mental health service use during the follow-up period between 2006 and 2015. We included primary and specialist psychiatric outpatient and inpatient care in our outcome measures. Our main exposure groups were non-refugee migrant children, refugee migrant children, descendants of migrants, and Swedish majority children. Quantitative studies 1 and 3 used cox regression models to study mental health service use, comparing migrant with Swedish majority children and the effects of age, reason for migration, parental presence at migration, and neighborhood vulnerability on service utilization. Neighborhood vulnerability was defined according to the Swedish police authority’s list of socio-economically deprived areas. Quantitative study 2 used logistic regression models to examine differences in received psychiatric diagnoses and treatments as well as level of care for first contact with services, comparing migrant sub-groups and majority children. We identified treatment recommendations from official clinical guidelines to examine potential differences in treatment receipt between migrant and majority children with the same diagnoses. The qualitative study (study 4) used in-depth interviews and thematic analysis to explore perceptions and experiences among migrant parents who had gained access to services for their children’s mental health needs. We used purposeful sampling to recruit parents from different outpatient clinics in the Stockholm region and included non-native speakers in our data gathering by conducting interviews with the help of professional interpreters.

Results: We found several indications of underutilization among migrant children as compared to their majority peers, in addition to recognizing the importance of sub-group analyses. Thus, study 1 showed that refugee adolescents and young adults used statistically significantly less mental health care compared to their majority peers, whereas refugee migrants below the age of 10 did not differ in terms of utilization from the reference group. When controlling for age and looking at effects of time in Sweden, we found that refugee migrant children and youth, and particularly unaccompanied refugee minors, who had been resident in Sweden for less than four years, had statistically significantly higher service use compared to majority children. In contrast, non-refugee migrants had consistently lower use across all ages and irrespective of time in Sweden, as compared to their majority peers (study 1). We also observed that migrant children had significantly lower odds of receiving several psychiatric diagnoses, such as neurodevelopmental, anxiety and mood disorders, compared to their majority counterparts (study 2). Additionally, migrant children had lower odds of receiving several recommended treatments as stated in official clinical guidelines, compared to their majority counterparts who were diagnosed with the same conditions (study 2). Moreover, refugee migrant children and youth, as well as descendants of migrants, who had lived for more than two years in a vulnerable neighborhood (characterized by the Swedish police authority as areas with socio-economic deprivation and high rates of criminality) had significantly lower mental health service use compared to their peers who had not lived in such neighborhoods (study 3). By contrast, majority children and youth who had lived in these neighborhoods for more than two years used more mental health services than their majority counterparts who had never lived in such areas. Lastly, the qualitative study (study 4) revealed that among those migrant parents who were interviewed, even those who were familiar with mental health symptoms and the existence of services did not know how to access services on their own. Rather, parents described a reliance on referring agents and that they had turned to trusted professionals with whom they were already in contact, to secure help for their children.

Conclusion: We found indications of underutilization of mental health services, especially among refugee migrant adolescents and young adults and among those with refugee or descendant backgrounds in disadvantaged neighborhoods. Discrepancies between migrant and Swedish children in terms of treatment receipt call for further investigation. Adolescent and young adult refugees, at increased risk of poor mental health, who were observed to use less mental health care than their younger counterparts, may benefit from strengthening the capacities of those in their surroundings to detect needs and increase referrals to mental health services. In line with such implications, the reliance on referring agents that emerged in the qualitative study suggests that trust and collaboration across sectors may enhance access to care. Overall, the findings of this thesis imply that mental health needs are not necessarily translated into service utilization and that social, cultural, and structural barriers to care may impede those most in need from accessing help. Policies should be directed against inequalities in access, as well as towards the deeper determinants of mental health discrepancies across different population groups.