In Reply
—Our study found that severity-adjusted odds of death was 19% lower in major teaching hospitals than in nonteaching hospitals. As noted by Dr Blumberg, patients in major teaching hospitals were less likely to be admitted from a SNF. This factor was included as a variable in our severityadjustment models and may have contributed to the higher predicted probability of death in minor teaching and nonteaching hospitals. However, our primary analyses did not exclude patients discharged to SNFs and were, in fact, based on the entire sample of 89 851 patients. Moreover, in analyses limited to patients admitted from home, the adjusted odds of death was 21% lower (95% confidence interval, 2%-35%) in major teaching hospitals. We conducted several secondary analyses to further control for possible differences in discharge practices, such as including patient length of stay as a covariate, examining deaths that occurred 3, 5, 7, 10, or 14 days