Based on occupational data for all (n = 264) prostate cancer cases diagnosed during 1980-84 in urban Shanghai and on employment information from the 1982 census, standardized incidence ratios (SIR) were calculated for occupational groups classified by job type and physical activity level. White-collar workers (professionals, government officials, clerical workers, salespersons) had an elevated incidence of prostate cancer, although the excesses were not significant. In addition, when jobs were classified by time spent sitting or energy expenditure, men employed in occupations with low physical activity levels tended to have moderately elevated risks of prostate cancer. Findings from this study in an area with one of the world's lowest incidence rates of prostate cancer add to the accumulating evidence that jobs with a low level of physical activity are associated with an increased prostate-cancer risk.