Hester van Eeren

| Addendum | 156 was externalizing problem behavior. The secondary outcomes were the proportion of adolescents who were living at home, who were engaged in school or work after treatment, and who lacked police contact during treatment. No difference was found between MST and FFT regarding externalizing problem behavior, but the adolescents who received MST were more likely to be engaged in school or work after treatment compared with FFT. Because the risk-need-responsivity (RNR) model guided treatment assignment, effectiveness was also estimated in youth with and without a court order, as an indicator of their risk level. For adolescents without a court order, MST yielded a larger effect on externalizing problems. The propensity score could not balance the treatment groups in the subsample of adolescents with a court order, and, therefore, MST and FFT could not be compared on their effectiveness in this subsample. Though treatment assignment was based on the RNR model, results in the group without a court order were not in accordance with this model, while higher-risk adolescents with a court order were indeed more often assigned to the more intensive treatment, namely MST. Although MST is expected to be the more expensive treatment because it is more intensive, estimating the cost-effectiveness of these interventions seems only relevant in the subgroup of adolescents without a court order. In the general discussion in Chapter 6 we summarized the findings of this thesis and we conclude that cost-effectiveness analyses provide valuable information to allocate public budgets to available youth care interventions wisely. In addition, we conclude that clinical practice data in youth care, that is routinely gathered, is needed to evaluate interventions on their effectiveness, and ultimately on their cost-effectiveness. Furthermore, the implications for youth care and for future research and policy are discussed. From a youth care perspective, this thesis implicates that research findings from evaluation studies in every day practice should be brought to clinical practice by effectively communicating these findings to clinicians, patients, and policymakers to enable them to assign youth to evidence-based interventions. Data on health care costs besides the intervention itself should be routinely gathered to evaluate interventions on their cost-effectiveness. Cost-effectiveness research is especially important since a large number of municipalities have to decide on reimbursing interventions and current standards may differ between municipalities. From a research and policy perspective, it is recommended, whenever possible, to follow guidelines in reporting findings from studies using clinical practice data, since in such studies the context of the interventions is even more important than in randomized controlled trials. Furthermore, the criteria to decide upon the level of effectiveness and cost-effectiveness evidence should be made more explicit and transparent, especially if this evidence is used to decide upon the reimbursement of youth care interventions. To improve the implementation of health economic evaluation studies in youth care, guidelines on cost-effectiveness research in youth care should be developed. When developing these guidelines, one should be prone to learning from other related research fields and combine relevant perspectives and methods.