123 The effects of supported housing for individuals with mental disorders Subgroup analysis We study subgroup effects by age, personal income quintile and mental disorder diagnosis, which are relevant characteristic for potential heterogeneous effects. First, different age groups (18-30, 31-50 and 51-79 years old) correspond to different profiles in terms of living situation and social network. While the youngest group often applies for supported housing leaving behind their family house, the older age groups tend to have a much more reduced social network (living on their own, in other institutions or be homeless), with different potential effects on social functioning and parental outcomes. Second, social disadvantage (proxied by the lower personal income quintiles) likely increases the effect of eligibility on admission to supported housing, while individuals in higher income quintiles are better surrounded by (in)formal caregiving and support at home. Last, the type of disorder (substance use disorder, psychotic disorder, disorder diagnosed in the childhood, mood disorder, personality disorder, other disorder and no treatment/diagnosis) may influence the extent to which individuals may recover and live an independent life and work. Our findings for subgroup effects show limited heterogeneity for most outcomes, even though mostly imprecise (Figure 2 and Figure A1 in the appendix). Point estimates are suggestive of potential heterogeneous effects in admission to supported housing in the next calendar year (for older individuals and those with a personality disorder) and in the likelihood of working in the short-run (for those with mood disorders). The spillover effects among parents living with the child ahead of her application confirm the patterns found in the main analyses for the parental outcomes, but the effects are much less precisely estimated (Table A19 of the Appendix). 4
RkJQdWJsaXNoZXIy MjY0ODMw