Abstract
To assess the association of smoking with the risk of glucose intolerance (diabetes plus impaired glucose tolerance). A cohort consisting of 1,711 Finnish men born in 1900-1919 were followed up from 1959 to 1994. Smoking status was assessed in a similar way at each of the six surveys from 1959 to 1989, and subjects were classified as never, former, or current smokers. Diagnosis of diabetes and impaired glucose tolerance was made according to the oral glucose tolerance tests made in 1984 and 1989, and the 1985 World Health Organization criteria was applied. Association between smoking and glucose intolerance was estimated separately for 420 participants and 243 nonparticipants in 1989. Multiple logistic regression analyses show that odds ratios of glucose intolerance in 1984 for current smokers in 1984 were 0.36 (0.19-0.70) and 1.20 (0.52-2.78), respectively, in the participants and the nonparticipants in 1989. Among the nonparticipants in 1989, the odds ratio for current smokers in 1969 was 2.23 (1.00-4.96). A reduced risk of glucose intolerance in 1989 associated with smoking in the participants in 1989 was found to be significant from the beginning of the follow-up. The participants in 1989 were generally healthier and had a longer life expectancy than the nonparticipants in 1989. In a retrospective study of men, an increased risk of diabetes and impaired glucose tolerance in smokers was found among the nonparticipants, but a reduced risk was found among the participants in 1989. The difference observed might be attributed to the fact that the participants were constitutionally different from the nonparticipants.