Objective: This research examined resident and facility-specific factors associated with a diagnosis of a urinary tract infection (UTI) in the nursing home setting. Method: Minimum Data Set and Online Survey, Certification and Reporting system data were used to identify all nursing home residents in the United States on April 1, 2006, who did not have a UTI (n = 1,138,418). Residents were followed until they contracted a UTI (9.5%), died (8.3%), left the nursing home (33.2%), or the year ended (49.0%). A Cox proportional hazards model was estimated, controlling for resident and facility characteristics and for the state of residence. Result: The presence of an indwelling catheter was the primary predictor of whether a resident contracted a UTI (adjusted incidence ratio = 3.35, p < .001), but only 6.1% of the residents in the sample had such a catheter. Therefore, only one eighth of the UTIs were contracted by residents with a catheter. Thus, subsequent analysis examined the populations with and without catheters separately. Demographic characteristics (such as age) have a much greater association with incidence among residents without catheters. The association with facility factors such as percentage of Medicaid residents, for-profit, and chain status was less significant. Estimates regarding staffing levels indicate that increased contact hours with more highly educated nursing staff are associated with less catheter use. Discussion: Several facility-specific risk factors are of significance. Of significance, UTIs may be reduced by modifying factors such as staffing levels.