OBJECTIVE: The aim of this study was to assess mortality amongst participants in long-distance ski races during the Vasaloppet week. We considered the 90 km races for men and 90 or 30 km for women. The vast majority of the participants in these races are not competing on the elite level. It is assumed, however, that they have to undergo regular physical training during a long period of time in order to successfully finish the race. DESIGN: The cohort study consisted of 49 219 men and 24 403 women, who participated in any of the races during 1989-1998. All subjects were followed up in the National-Cause-of-Death-Register until 31 December 1999. We computed the standardized mortality ratios (SMRs) adjusting for age and calendar year. RESULTS: Overall, 410 deaths occurred, compared with 850.6 expected, yielding an SMR of 0.48 [95% confidence interval (CI) 0.44-0.53]. Low SMRs were found in all age groups in both men and women and in all groups after categorization by finishing time and number of races. The lowest SMRs were found amongst older participants and in those who participated in several races. A decreased mortality was observed in all major diagnostic groups, namely cancers (SMR = 0.61; 95% CI 0.52-0.71), diseases of the circulatory system (SMR = 0.43; 95% CI 0.35-0.51), and injuries and poisoning (SMR = 0.73; 95% CI 0.60-0.89). For lung cancer the SMR was 0.22, but even after exclusion of lung cancer the all-cancer mortality was low (SMR = 0.72; 95% CI 0.59-0.86). CONCLUSIONS: We conclude that participants in long-distance skiing races, which demand prolonged regular physical training, have low mortality. The extent to which this is due to physical activity, related lifestyle factors, genetics or selection bias is yet to be assessed.