Sharon-Lise Normand, PhD, and colleagues recently published results of a statistical study investigating the accuracy of different data sources in comparing mortality following coronary artery bypass grafting (CABG) across hospitals. Dr. Normand is a professor of health care policy (biostatistics) at HCP and director of the Mass-DAC cardiac surgery data analysis center.

The researchers, who had access to both audited clinical data and administrative data from 2003 in Massachusetts, carried out a comprehensive comparison of these two data sources and the hospital-quality profiles derived from them. The team concluded that the cardiac surgery report cards using the administrative data were less reliable than those using the validated clinical data, mainly because the administrative data were more likely to include cases in the public report cards that should in fact not have been included. Moreover, higher-volume hospitals were more at risk for this type of error than smaller volume hospitals.

The abstract of this article is available on the Circulation website. The study was also discussed in an editorial in the same issue (Circulation 2007:115).