The trade-off between efficiency and quality in hospital departments

Hospital managers are confronted with decisions that have to account for multiple objectives, which may be in conflict with regard to efficiency and quality of care. In empirical studies occupancy and staffing ratios as well as in-hospital mortality are frequently used measures for efficiency and quality-of-care, respectively. Efficiency and quality measures vary on a daily basis. However, most empirical studies fail to take this variation into account, especially because data of daily staffing levels are lacking. The paper seeks to exploit the notion that staffing levels are planned according to expected occupancy levels, i.e. estimated daily occupancy levels account for unobserved daily staffing levels.

Using administrative data from 2004 for a sample of 62 departments in 33 German hospitals, the relation between daily occupancy levels and in-hospital mortality count on the department level is analyzed. In an OLS-framework the paper estimates daily occupancy level for all departments and then uses the predicted occupancy levels in a zero-inflated Poisson (ZIP) regression framework to explain in-hospital mortality count.

The results show a potential trade-off relation between predicted occupancy rates and mortality. More specifically, the paper finds that the trade-off relation is less pronounced in hospitals with a higher number of available staff per bed.

First, the paper shows evidence for a negative trade-off between measures of managerial and medical performance on a day-to-day basis. Second, interactions between single measures of efficiency are modeled, namely predicted occupancy rate and staff per bed ratios, and policy implications are developed. Third, first empirical results in this respect using German data are presented.