We thank Beyersmann and Schrade¹ for their comments, and we agree that the use of an estimator that accounts for death without prior infection is appropriate for our data. As they point out, the raw incidence proportions of methicillin-resistant Staphylococcus aureus (MRSA) we reported (16.33% for persistently colonized, 11.18% for intermittently colonized, and 0.51% for non-colonized) reflected observed infections, and did not account for infections that may have occurred after administrative censoring at the study end date. Thus, these data represent an underestimation of risk.

For this reason, we reported proportions based on Kaplan-Meier analysis: 21.26% for persistently colonized, 12.83% for intermittently colonized, and 0.55% for non-colonized. Beyersmann and Schrade aptly pointed out that this analysis does not account for death without prior infection and thus overestimates risk. They propose that the Aalen-Johansen method is a better estimator in this case because it accounts for competing causes. We agree, and we conducted an additional analysis. These new calculations produced results that fall between the other 2 estimates: 20.61% for persistently colonized, 12.16% for intermittently colonized, and 0.54% for non-colonized. The use of the Aalen-Johansen estimator increases the precision of the estimates without changing our overall conclusions.