Hospital quality-review spending and patient safety: a longitudinal analysis using instrumental variables

Abstract

Since the landmark Institute of Medicine’s (IOM’s) 2000 report first focused attention to the problem of the safety of inpatient care, it has been a priority of hospital staffs, administrators, and policymakers. Despite remarkable progress in the 20 years since the IOM report, there is still much unknown about how these improvements in safety have been achieved. Using a 12-year (2004 – 2015) panel of Florida acute-care general hospitals, we estimate the relationship between hospital quality-review (QR) spending and patient-safety outcomes, using a composite measure of patient safety (PSI-90) from the Agency for Healthcare Research and Quality. Our identification strategy to account for endogenous QR spending relies on exogeneity from within the hospital, in which we use staffing of non-acute ancillary services as instruments for QR spending. Estimation of hospital fixed effects (FE) and fixed effects with instrumental variables (FEIV) models both yield statistically significant and beneficial effects of QR spending on patient safety, with much larger effects in the FEIV model. From this model, we find, on average, that a 10 percent increase in QR spending is associated with a 2.5 percent (0.25 elasticity) beneficial decrease in patient-safety events (i.e. PSI-90). Broadly, this study represents a unique contribution to the literature by examining the direct relationship between hospital spending and quality.