The Centers for Medicare & Medicaid Services' Overall Hospital Quality Star Rating program has raised concerns since its introduction in 2016. Using both national data and data from a large urban teaching hospital, the authors examined a few methodological issues of one heavily weighted measure group, the Safety of Care group. The authors investigated the validity of the assumption that a single underlying quality trait exists among the 8 Safety measures, and the sensitivity of the Safety group score in response to a range of measure improvement scenarios. Also explored were the effects of an alternative weighting method and an alternative measure score calculation method on the results of a single hospital's Safety group score. Evidence was found for 4 (rather than 1) underlying quality dimensions among the 8 Safety measures, and the Safety group score calculated using the current method was notably different from that calculated using the alternative methods.
Keywords: Star Rating; hospital; quality of care; safety measures.