Lung cancer incidence rate after the cessation of smoking was assessed for male ex-smokers according to the age at cessation, using the results from a case-control study for ex-smoker versus continuing smoker, and the lung cancer incidence rate function for continuing smoker estimated from Japan Vital Statistics and the "Six-prefectural Cohort Study" in Japan. This hospital-based case-control study consisted of 776 lung cancer cases (553 current smokers and 223 ex-smokers) and 772 controls (490 current smokers and 282 ex-smokers) who started smoking at ages 18-22. The odds ratio of developing lung cancer for ex-smokers compared to continuing smokers according to years since the cessation of smoking was estimated for four age groups (55-64, 60-69, 65-74 and 70-79). Given that the number of years since cessation of smoking is the same, reduction of the odds ratio appeared to be greater for the younger age group than for the older age group, reflecting the shorter period of exposure for the younger age group. Lung cancer incidence rate (per 100,000) was assumed to be expressed by the following function; 1.7 x 10(-5) x (age -24.3) for continuing smokers and 0.15 x 10(-5) x (age) for nonsmokers. Lung cancer incidence rate among ex-smokers according to years since cessation was then estimated to be the above function multiplied by the odds ratio from the case-control study for each age group. In contrast to the greater reduction of the odds ratio among younger ex-smokers, reduction of the incidence rate, in terms of rate difference, was considerably greater for older ex-smokers due to a high incidence rate of lung cancer for older continuing smokers. This indicates that the absolute magnitude of reduction of the lung cancer incidence rate after cessation of smoking is greater for older ex-smokers, although the relative magnitude of reduction is greater for younger ex-smokers.