Hester van Eeren

| Chapter 6 6 | 140 This dissertation addresses two issues of contemporary interest in the field of youth care: 1) the feasibility of cost-effectiveness research in this field and 2) the use of available, non-randomized, data in investigating the effectiveness of interventions in youth care practice. These issues will be discussed in light of the results in this dissertation and, subsequently, several recommendations will be provided for clinical practice and further research. Cost-effectiveness research in youth care We investigated whether state of the art methods commonly applied in health care evaluation studies can be applied to systemic interventions in youth care. In Chapter 2, Functional Family Therapy (FFT) was compared with Treatment as usual (TAU) in an illustrative probabilistic Markov model, in which parameter uncertainty and long-term cost-effectiveness were taken into account. Estimating long-term cost-effectiveness is essential when interventions are applied to youth aged 12 to 18 years and when treatment effects are expected to last into adulthood. By expressing the cost-effectiveness ratio in costs per Criminal Activity Free Year (CAFY), we used an outcome measure that addressed not only the clinical, but also the societal effect of the interventions. By doing so, clinically relevant issues and health economic questions were brought together. Chapter 2 thus showed that commonly applied economic evaluation methods are applicable in evaluating youth care. However, when modelling the cost-effectiveness of interventions, the results in both cost and effect estimates can be subject to uncertainty. This parameter uncertainty can be reduced by information obtained from further research. A value of information analysis can reveal the value and justify the direction of further research. In Chapter 3, an illustrative value of information analysis was applied using the cost-effectiveness model developed in Chapter 2. Thereby, it was needed to assume a ‘willingness-to- pay (WTP) value’ for one CAFY (i.e., what is society willing to pay for one year without criminal activity of one adolescent?) since the expected value of further information depends on this WTP value. Findings from a value of information analysis may lead to reimbursing the intervention studied or not under certain conditions. Chapter 3 showed that a value of information analysis in the field of youth care can also be interpreted as similar to cost/ Quality Adjusted Life Year (QALY) studies in health care evaluations and is particularly meaningful in this field, because the interest in cost-effectiveness research is increasing, and its analyses could use a wider range of input parameters than cost- effectiveness research in health care, for instance. Both the cost-effectiveness analysis in Chapter 2 and the value of information analysis in Chapter 3 revealed important issues that should be considered in future cost- effectiveness research in youth care. First, in contrast to health care evaluation studies, in youth care not only the referred client (i.e., the adolescent) should be the focus of research, but also the systems surrounding him. For example, the societal perspective chosen in the analysis should include the effect of the intervention in terms of reduced costs and increased well-being of family members, reduced victim costs, and reduced costs because of avoided crimes, all taken into account over the defined time-horizon of