The androgen receptor has been implicated in the development and progression of bladder cancer (BCa), largely based on studies of animal models. We investigated whether finasteride was associated with a reduced incidence of BCa as observed by self-report in the Prostate, Lung, Colorectal, and Ovarian cancer screening trial. Cox proportional hazard regression analysis was performed to determine the association of finasteride use with time to diagnosis of BCa, controlling for age and tobacco use. Of the 72 370 male participants who met inclusion criteria, 6069 (8.4%) had reported the use of finasteride. BCa was diagnosed in 1.07% (65 of 6069) of those who reported finasteride compared with 1.46% (966 of 66 301) of those who reported no use during the trial. In a multiple Cox regression analysis, self-reported use of finasteride was associated with a decreased risk of development of BCa (hazard ratio: 0.634; 95% confidence interval, 0.493-0.816; p=0.0004), controlling for age and smoking. Limitations of this study include that it is observational and not randomized, that many of the confounding variables for BCa, such as alcohol use, were not available for use in the analysis, and that finasteride use was by annual self-report, which is subject to missing values and error.