Venue: The Fuqua School of Business, Duke University, 1 Towerview Drive, Durham, NC 27708-0120
Presentation
Do Higher Quality Hospitals Practice Differently than Lower Quality Hospitals
Researchers have responded to the movement toward pay-for-performance (P4P) initiatives, as well as the heightened interest in the quality of care provided by hospitals, with studies focusing on these issues. In this paper, we aim to add to the literature by determining whether or not hospitals providing high quality of care practice differently than hospitals providing low quality of care. We opt for this approach in order to measure and develop benchmarks that can be applied to hospitals nationwide. The methodology we use here is an extension of the data envelopment analysis (DEA) approach. Because of our rich data set - urban hospitals operating in 34 states during 2004 complete with several measures of inputs, outputs, case-mix, and patient safety indicators - we can derive two separate production frontiers that account for both efficiency and quality. By comparing the frontier of the high quality hospitals with the low quality hospital frontier, we can ascertain how these frontiers differ, as well as if the cause of the difference is relative measures of inefficiency or quality of care.
The data we use for this study comes from the American Hospital Association (AHA) Annual Survey of Hospitals and the Medicare Cost Reports. Nurse-sensitive measures of quality come from the application of the Patient Safety Indicator (PSI) module of the Agency for Healthcare Research and Quality (AHRQ) Quality Indicator (QI) software to State Inpatient Databases (SIDs) made available by the Healthcare Cost and Utilization Project (HCUP). Our total sample size is 1377 hospitals.
Because of these data, we can side-step earlier criticisms that using a deterministic model does not account for differences in case mix and quality of care. In our approach, we can directly use these measures in the input-output correspondence yielding information on best practice hospitals in terms of both efficiency and quality.
To preview our results, we found that lower quality hospitals, once correcting for inefficiency, were still inferior to high quality hospitals in terms of their productive practices (i.e., converting inputs into outputs). Therefore, we conclude that high quality hospitals are indeed systematically different. This finding suggests that their practices could be studied by lower quality hospitals to perhaps improve on their production of inpatient care.