OBJECTIVE: To determine the extent to which a history of Bacille Calmette-Guerin (BCG) vaccination influences the likelihood of positive tuberculin skin test (TST) results.DESIGN: Cross-sectional survey using a hospital-based tuberculosis surveillance program.SETTING: Health Sciences Centre, a tertiary care hospital in Winnipeg, Manitoba.METHODS: The 476 health care workers (HCWs) who had TST as part of the surveillance program between 1993 and 1997 constituted the study population. The two-step test was done in 91% of the participants who did not have a positive initial test, defined as 10 mm or greater of induration. Data were gathered through chart review supplemented by a short questionnaire administered to the HCWs.MAIN RESULTS: One hundred and thirty-eight HCWs (29%) had a positive TST. In a stepwise, multiple logistic model controlling for age, sex, job title, work area, age of receiving BCG, time since BCG and duration of employment, only a history of BCG vaccination (odds ratio [OR] 22; 95% CI 12 to 41) and birth outside of Canada (OR 2.6; 95% CI 1.4 to 5.8) were significantly associated with a positive TST. When the definition of a positive TST was modified by increments of 1 mm, from 10 mm up to 20 mm of induration in BCG recipients, BCG was associated with positive reactions with indurations up to 19 mm but not 20 mm or greater. The OR declined with each increment. Of the 84 HCWs who were documented to have at least 20 years between BCG vaccination and testing, 41 (49%) had positive reactions.CONCLUSIONS: BCG vaccination can produce lasting tuberculin reactivity, and indurations of 19 mm or less may be due to the effects of the vaccine.