Cases of C. difficile averted by public reporting based on differences in observed and expected monthly case counts.

aRelative risks for post-intervention months are observed over expected C. difficile counts, where expected count is based on the Poisson model predictions for that month.bExpected counts are based on Poisson model predictions for that month, adjusting for hospital, age strata, calendar month, and population antibiotic consumption with lags of 0–12 mo.