This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.
—In their letter, Hartz et al note that some data elements in our risk-adjustment model for coronary bypass graft mortality may actually be consequences of care rather than risk factors. Other variables in our model, such as the use of internal mammary artery grafts, may reflect physician practice patterns. Our model might have had less predictive power if such variables had been excluded.We fully agree with these comments. Designing a risk-adjustment model is not a simple task. The investigator must make reasoned judgments about which variables are likely to represent underlying patient characteristics and which variables are more likely to represent the quality of care. The latter variables should not be used as predictors in a risk-adjustment model. In this case, we felt that the duration of time between a cardiac catheterization and a subsequent coronary artery bypass graft was more likely to reflect the findings of
Luft HS, Romano PS. Coronary Artery Bypass Graft Mortality: Patient Risk or Physician Practice?-Reply. JAMA. 1993;270(24):2928. doi:10.1001/jama.1993.03510240040024
Customize your JAMA Network experience by selecting one or more topics from the list below.