Hospital productivity and the Norwegian ownership reform – A Nordic comparative study
2008/10: Authors: Kittelsen, S.A.C., Magnussen, J., Anthun, K.S., Häkkinen, U., Linna, M., Medin, E., Olsen K.R., Rehnberg, C., Ragnar Frisch Centre, NTNU, SINTEF Health Research, STAKES, Karolinska Institutet & DSI (PDF 124 kb)
In a period where decentralisation seemed to be the prominent trend, Norway in 2002 chose to re-centralise the hospital sector. The reform had three main aims; cost control, efficiency and reduced waiting times. This study investigates whether the hospital reform has improved hospital productivity using the other four major Nordic countries as controls. Hospital productivity measures are obtained using data envelopment analysis (DEA) on a comparable dataset of 728 Nordic hospitals in the period 1999 to 2004. First a common reference frontier is established for the four countries, enveloping the technologies of each of the countries and years.
Bootstrapping techniques are applied to the obtained productivity estimates to assess uncertainty and correct for bias. Second, these are regressed on a set of explanatory variables in order to separate the effect of the hospital reform from the effects of other structural, financial and organizational variables. A fixed hospital effect model is used, as random effects and OLS specifications are rejected. Robustness is examined through alternate model specifications, including stochastic frontier analysis (SFA).
The SFA approach in performed using the Battese & Coelli (1995) one stage procedure where the inefficiency term is estimated as a function of the set of explanatory variables used in the second stage in the DEA approach. Results indicate that the hospital reform in Norway seems to have improved the level of productivity in the magnitude of approximately 4 % or more. While there are small or contradictory estimates of the effects of case mix and activity based financing, the length of stay is clearly negatively associated with estimated productivity. Results are robust to choice of efficiency estimation technique and various definition of when the reform effect takes place.
Also as Discussion Papers 3/2008 from STAKES, Helsinki, Finland 2008.