127 The effects of supported housing for individuals with mental disorders provided is limited15, and should not represent a large difference in terms of support for these subgroups. The remaining 6% of non-eligible individuals was granted different forms of institutional care such as nursing homes or institutions for disabled: our checks indicate that this small group (<1% study population) should have no influence on our results (Table A14 and Figures S2 to S4 of the appendix). Last, it is important to acknowledge the limitations of our data in capturing some relevant dimensions. First, administrative data does not fully capture individual level benefits of supported housing in terms of well-being and health-related quality of life and housing retention. Second, paid employment might also be considered as limited to capture social participation: it does not capture voluntary work, and does not allow distinguishing those employed in competitive (regular) jobs from those in sheltered employment. It is not clear if being in sheltered employment would affect the probability of work, or the amount of earnings. It might be worth noticing that both voluntary arrangements and sheltered employment were more prominent at the time of our study than in the present days, given the current focus of programs such as Individual Placement and Support to place individuals in supported housing in competitive employment [28]. Overall, our results seem to support some of the criticisms towards supported housing: that it is costly for of care and might undermine patient’s social functioning [9, 10], which we document through labour market outcomes. On the other hand, the results suggest positive spillovers to parental employment. While we are only able to measure a subset of relevant outcomes, our study suggests that there is a trade-off to be considered between different effects. Further policy decisions on extending supervised environments should deal with such trade-offs by weighting the effects in the various domains and perspectives – the system, the individuals and their families. 15 Considering the average hourly cost of 58 euros for home care (weighted average between individual guidance 80%, group guidance 15% and nursing 5%), the average consumption of home care in the next calendar year is 26 minutes/week (1513 euros/58 euros/52 weeks). 4
RkJQdWJsaXNoZXIy MjY0ODMw