analysis. When conversion rates were examined at 6-month intervals, the rate fell from 20.7 percent during the first 6 months of the 1991 observation period to 5.8 percent during the last 6 months of 1993 (31). During that time, negative-pressure isolation rooms, ultraviolet lights, and personal respirators were all introduced at St. Clare’s Hospital.

Maloney and coworkers studied the impact of enhanced infection control measures on nosocomial transmission of tuberculosis infection at the Cabrini Medical Center in New York City following an outbreak of multidrug-resistant tuberculosis at that facility in 1991 (32). Employee tuberculin rates were determined for an 18-month period prior to the institution of enhanced infection control measures and a 12-month period subsequent to the changes in infection control. The findings of their study are summarized in Table C-9, along with annual risks of infection calculated from their data. Overall, the annual risks of infection were higher in personnel working in contact with patients and, considering the small number of conversions documented in the noncontact group, were not changed by the implementation of infection control measures. However, conversion rates decreased following the infection control intervention on medical and HIV wards admitting patients with tuberculosis. This did not happen elsewhere in the hospital. In this study, there was no evidence of residence postal code clustering of employee conversions, nor were demographic or racial characteristics identified that contributed to the infection risk.