Several studies in subjects occupationally exposed to arylamine carcinogens have shown increased risks for bladder cancer associated with the slow acetylator phenotype. To follow up these reports, a case-control study of N-acetylation and bladder cancer risk was carried out among subjects occupationally exposed to benzidine, in benzidine dye production and use facilities in China. Thirty-eight bladder cancer cases and 43 controls from these factories were included for study of acetylation phenotype, by dapsone administration, and for polymorphisms in the NAT2 gene, by a polymerase chain reaction (PCR)-based test. In contrast to previous studies, no increase in bladder cancer risk was found for the slow N-acetylation phenotype (OR = 0.3; 95% CI = 0.1-1.3) or for slow N-acetylation-associated double mutations in NAT2 (OR = 0.5; 95% CI = 0.1-1.8). Examination of specific mutations and adjustment for age, weight, city and tobacco use did not alter the results. When examined by level of benzidine exposure in the cases, the bladder cancer risks associated with low (OR = 0.3, 95% CI = 0.0-2.2), medium (OR = 0.7, 95% CI = 0.1-4.5) and high (OR = 0.6, 95% CI = 0.1-3.5) exposure showed no interaction between genotype and benzidine exposure, within the range of exposures experienced by subjects in this study. This study, which is the first to incorporate phenotypic and genotypic analyses, provides evidence that the NAT2-related slow N-acetylation polymorphism is not associated with an increased risk of bladder cancer in workers exposed to benzidine, and may have a protective effect.