Male circumcision may reduce the risk of HIV acquisition in heterosexual men. In observational studies, confounding by demographic factors could produce spurious associations between circumcision and HIV risk. To assess this possibility, cross-sectional data on self-reported circumcision status and demographic factors were collected at baseline as part of the Health in Men study. Two-thirds of 1426 subjects were circumcised, mostly in the neonatal period. In multivariate analyses, age, ethnicity and country of birth were independently associated with circumcision status. Confounding by these demographic factors must be considered in future epidemiological studies addressing circumcision and HIV acquisition.