RESUMO
A prevalence survey was performed at a Veterans Administration nursing home care unit to detect the frequency, demographics, features, and potential risk factors associated with infections, compare different methods for calculating infection rates, and compare prevalence data with routine physician reporting of infections during a one-month period. In 176 evaluable residents, 22 nursing home-acquired infections were detected, with rates of 12.5% by infections per residents at risk, and 4.6% by infections per 1000 resident-days. Fifty percent of these infections (11 of 22) involved the urinary tract, seven of which were associated with indwelling catheters. Factors potentially associated with increased overall infection rates included immobility (P less than .02), acute-care hospitalization in the 28 days preceding the study (P less than .01) and antibiotics given preceding the infection (P less than .001). An indwelling urinary catheter (P less than .01) potentially was associated with an increased urinary tract infection rate. During the concurrent period, routine physician reporting of infection detected four of the 22 infections (18%) that were identified by the survey. These findings suggest that physician reporting of infection underestimates infection risk as compared with conventional surveillance practices, and that hospital-based care units may have infection problems more similar to acute-care hospitals than to community nursing homes. Identification of potential risk factors is critical in planning preventive practices based on institution-specific needs.