Performance of hospital services in Ontario: DEA with truncated regression approach

• Analyzed productive performance of Hospital Services in Ontario.
• Applied non-parametric kernel density estimation and tests to efficiency distributions.
• Investigated the determinants of efficiency using truncated regression with double-bootstrap.
• Identified key factors that act as significant determinants of hospital efficiency.

In this work, we analyze production performance of hospital services in Ontario (Canada), by investigating its key determinants. Using data for the years 2003 and 2006, we follow the two-stage approach of Simar and Wilson (2007) [76]. Specifically, we use Data Envelopment Analysis (DEA) at the first stage to estimate efficiency scores and then use truncated regression estimation with double-bootstrap to test the significance of explanatory variables. We also examine distributions of efficiency across geographic locations, size and teaching status. We find that several organizational factors such as occupancy rate, rate of unit-producing personnel, outpatient–inpatient ratio, case-mix index, geographic locations, size and teaching status are significant determinants of efficiency.