15581-m-nanninga

GENERAL DISCUSSION 139 This shows that children and adolescents with psychosocial problems and their families are a more hard-to-reach target group, potentially causing selection bias due to systematic non-response [45]. However, when comparing the respondents and non- respondents only small differences were found in age, gender, rural/urban area and psychosocial difficulties; this decreases the likelihood of selection bias. Finally, the TakeCare study enabled us to longitudinally compare children and adolescents enroled in care, and those not enroled, with high retention (i.e. loss to follow-up at the second and third waves was 6.9% and 8.8%, respectively, for the care sample, and 2.0% and 2.2%, respectively, for the community sample). Quality of the information obtained This thesis focused on the client perspective, relying on self-report information from parents and adolescents. In this way we obtained a broad range of information, mainly using standardized measurement instruments regarding both determinants and outcomes. Inclusion of other perspectives, like those of care professionals or teachers, might have yielded additional insights, since perspectives on psychosocial problems differ [56, 57]. This evidently deserves further study. Causality and confounding Our study on outcomes, Chapter 6, was set up longitudinally, enabling us to assess time sequences in care. Our study aimed to assess the associations between characteristics of children, adolescents and their families on the one hand and outcomes regarding enrolment on the other hand (Chapters 2-5); many factors may confound these associations. We therefore took into account potentially confounding variables such as socio-demographic characteristics and past use of psychosocial care. It should be noted that we assessed outcomes, and not effects, of care. IMPLICATIONS Implications for practice and policy Our findings have several implications for practice and policies on psychosocial care for children and adolescents. First, we found that the social environment plays an important role in enrolment in care: little family social support and poor parenting, i.e. poor supervision and inconsistent parenting, increased the likelihood of enrolment of children and adolescents in psychosocial care. This suggests that better parenting skills and stronger support from the family’s social network may to some extent prevent children’s psychosocial problems, or help them to cope with or solve problems, either