Pathways to labour market marginalisation : aspects of common mental disorders, migration status and age

Background: Labour market marginalisation (LMM), measured as sickness absence (SA)/disability pension (DP) or unemployment, differs between young immigrants and natives in several countries. The aim of this project was to investigate the risk of LMM among immigrants, second generation immigrants, and refugees and if the risk differed with region/country of birth, age, generation, refugee status and before and after a diagnosis of common mental disorder (CMD). A second aim was to assess psychiatric healthcare utilization before and after DP due to CMD among immigrants and natives, and if it differed before and after changes in social insurance regulations in Sweden in 2008. Methods: Four studies were conducted, all based on Swedish nationwide register data. Study I: a cohort study of all 3 507 055 individuals living in Sweden in 2004 (19-50 years), with a 6- year follow-up. Hazard ratios (HR) and 95% confidence intervals (CI) for DP were estimated by Cox regression for first and second generation immigrants compared to natives, across regions of birth and stratified by age. Study II: individuals with incident DP due to CMD (n=28 354), before or after the introduction of stricter social insurance rules, were included. Patterns of psychiatric in- and specialised outpatient healthcare utilization during a 7-year window before and after being granted DP were assessed by Generalized Estimating Equations. Study III: A total of 28,971 individuals (19-30 years) with an incident CMD in 2007 were included. Groupbased trajectory models were utilised to identify trajectories of annual months of LMM three years before and six years after the diagnosis. Study IV: A prospective cohort study including individuals (19-30 years) with CMDs during 2009 (N=69,515). Cox regression was used to calculate HR for future LMM (2010-2013). Results: In the first generation immigrants compared to natives, HRs for DP were higher (range: 1.17 to 1.74). Also in the second generation immigrants, HRs were higher (range: 1.10 to 1.30) (study I). Prevalence rates of psychiatric inpatient healthcare were comparable among immigrants and natives, lower though in non-Western immigrants. Three years after granted DP, non-Western immigrants in comparison to natives and Western immigrants had a stronger decrease in inpatient psychiatric healthcare. After 2008, a strong reduction in outpatient psychiatric healthcare after granted DP was observed, similarly in immigrants and natives (study II). Young natives and immigrants showed similar trajectories of SA/DP in individuals with CMDs. A higher proportion of non-Western immigrants (21%) followed trajectories of high levels of unemployment compared to Western immigrants (15%) and natives (17%). Educational level and duration of residence in Sweden discriminated trajectories of both SA/DP and unemployment (study III). Regarding subsequent unemployment, both refugees and nonrefugee immigrants with CMDs had a higher risk compared to natives. Regarding subsequent SA, refugees from Africa and Asia showed a lower risk compared to natives (study IV). Conclusions: Immigrants differed in comparison to natives in LMM in terms of a higher level of DP in both generations, in lower levels of inpatient psychiatric healthcare after DP granting, particularly among non-Western immigrants, in higher unemployment trajectories after a CMD and in higher unemployment risk and lower SA risk, particularly among refugees. Educational level, psychiatric comorbidity and duration of residence are relevant factors in discriminating differences in LMM trajectories among immigrants and natives