To the Editor.
—In discussing estimates of quality of hospital care, Dr Miller and colleagues1 wrote, "Standardized hospital mortality ratios based on analyses that do not account for variation in admission practice among hospitals are biased.1(p1493)... This problem is related to unmeasured severity, but it is more than that.1(p1495)... It is important to understand that the bias introduced by differences in admission practices will affect any quality measure that is estimated only on hospitalized samples."1(p1497)Miller et al believe comparing care inside hospitals requires collecting data about the health of populations that are sending patients to these hospitals. We believe this difficult task is unnecessary for the specific goal of comparing the care provided inside hospitals. Resources are better spent on collecting complete, standardized, high-quality data that describe the initial health and subsequent outcomes of patients treated inside hospitals.Miller et al are correct on several
Silber JH, Rosenbaum PR. Measuring Quality of Hospital Care. JAMA. 1995;273(1):21. doi:10.1001/jama.1995.03520250035014