With strong US licensing body mandates for quality improvement (QI) education, it is more important than ever to ensure that common QI indicators really measure quality. Hospital length of stay (LOS) is an extremely common quality measure because it is easy to quantify and has a large influence on hospital economics. Thirty-day mortality has more face validity as a quality measure, but is a fairly uncommon outcome and requires tracking after hospital discharge. Because shorter LOS is likely correlated with lower severity of illness, a patient-based comparison of LOS vs. mortality would be hopelessly confounded. In this single-site observational study, researchers tracked the 30-day mortality of inpatients who were arbitrarily (based on routine procedure) assigned to the teaching service of one of 79 internal medicine attendings. Attendings were categorized into two equal-sized groups by their average LOS, lower or higher than 4.77 days. Patients discharged by the short-LOS physicians had a higher 30-day mortality than those discharged by the long-LOS physicians (5.5% vs. 4.3%, p=0.007). These results were maintained after either logistic regression or propensity score matching. Pending further exploration, these results suggest that LOS, and perhaps other quality measures, may need re-thinking. — Laura Willett, MD
Southern WN, Arnsten JH. Increased risk of mortality among patients cared for by physicians with short length-of-stay tendencies. JGIM 2015;30:712-8.