OBJECTIVE: To estimate the fraction of carpal tunnel syndrome (CTS) that is attributable to work in the total adult population of the island of Montreal. METHODS: The population consisted of 1.1 million people 20-64 years of age, with 73.2% of men and 60.6% of women employed. The rates of first surgery for CTS were compared between occupational groups and the total adult population with the standardised incidence ratio (SIR) method. Rates of surgery for the island of Montreal were obtained from the provincial data base of payments. The occupational history was obtained from telephone interviews of a sample of surgical cases. The attributable fractions in exposed people were calculated with odds ratios (ORs) obtained from logistic regressions with non-manual workers as the control group. RESULTS: The surgical incidence of CTS was 0.9/1000 adults. SIRs for all manual workers were 1.9 (95% confidence interval (95% CI) 1.4-2.5) in men and 1.8 (95% CI 1.4-2.2) in women, and the fractions attributable to work were 76% (95% CI 47-88) and 55% (95% CI 33-69), respectively. Seven occupational groups were identified as having excess risk of surgical CTS, with fractions attributable to occupation ranging from 75% to 99%. CONCLUSION: Among manual workers on the island of Montreal, 55% of surgical CTS in women and 76% in men was attributable to work. Increased risk of surgical CTS was found in seven occupational groups.