Increased duration of treatment after a decline in the number of patients: evidence of supplier-induced demand?

This paper investigates the increase in average duration of mental health care treatments in the Netherlands in 2012. In the same year, a compulsory copayment of 200 euros was intro-duced in secondary mental health care. This intended to incentivise patients with mild condi-tions to seek primary care rather than secondary, specialised care. The compulsory copayment was abolished in 2013. Exploiting the exogenous shock to the number of patients in 2012, pooled OLS and fixed effects estimations are used to identify the effect of a change in the number of patients of a provider on treatment duration. Treatment duration clearly increased in the year 2012, captured by a dummy variable for 2012 in the regressions. But the analysis fails to provide evidence for supplier-induced demand, as the change in the number of pa-tients of a provider fails to yield significant results. This investigation was based on an admin-istrative database which covers all mental health care treatments between 2008 and the end of 2013. In the period, budgeted and non-budgeted providers were reimbursed in different man-ners. However, all providers were effectively reimbursed based on a retrospective, fee-for-service system, providing a financial incentive to extend treatment duration. A robustness check reveals that increase in treatment duration in 2012, as captured by a dummy variable, is mostly driven by budgeted institutions. This does suggest that the manner of reimbursement is important for treatment decisions.