Productivity growth in Norwegian psychiatric outpatient clinics - A panel data analysis of the period 1996-2001

Fulltext 498 kb

Norwegian government policy is to increase the supply of psychiatric services to children and young persons, both by increasing the number of personnel and by increasing productivity in the psychiatric outpatient clinics (BUP). Increased accessibility to services is observed for the last years, measured as the number of children receiving services every year. The question is to what extent this is related to increased productivity. The paper aims to estimate change in productivity among outpatient clinics. Questions whether change in productivity is related to the personnel mix of the clinics, growth in treatment capacity or change in financial incentives are analysed. We utilise a non-parametric method called Data Envelopment Analysis (DEA) to estimate a best-practise production frontier. The potential for efficiency improvement are estimated as the difference between actual and best-practice performance, while allowing for trade-offs between different staff groups and different mixes of service production. A Malmquist output-based productivity index is calculated, decomposed in technical efficiency change, scale efficiency change and frontier shifts. The paper analyses panel data on the psychiatric outpatient clinics of Norway for the period of 1996-2001. Output is measured as number of direct and indirect patient-related interventions (visits and consultations) while input is measured by usage of different types of personnel. The results indicate increased overall productivity, with important contribution from increased technical efficiency. Personell growth has a negative influence on productivity growth, while a growth in the share of university educated personell improves productivity. The financial reform of 1997 that gave greater weight for interventions per patient lead to lower productivity growth in the subsequent period for those that had an inital budgetary gain from the reform.