Analysis of productivity, efficiency and technological changes in hospital services in Ontario: How does case-mix matter?

In this paper we compare productivity, efficiency and technological changes with and without case-mix as output categories using panel data on Ontario hospitals for the period 2002-2006. We use the Malmquist Productivity Index (MPI), Data Envelopment Analysis (DEA), and non-parametric density estimation and related statistical tests. We also decompose MPI into efficiency change (ECH) and technological change (TCH). We found that productivity, efficiency and technological changes were significantly different between two alternative models, with case-mix (as a weighting device, Model 2) and without case-mix (Model 1). On the other hand, adding case-mix as a separate output (Model 3) did not have any significant impact on the distribution of efficiency estimates compared to the model without case-mix. We also found that in Model 2, productivity growth was driven mainly by improvements in technology rather than increases in efficiency while in Model 1 and Model 3, the two effects (ECH and TCH) almost balanced each other leading to no productivity change.