It has been assumed that a younger age at initiation of cigarette smoking is associated with an increased risk of lung cancer, but previous studies have not adjusted for two strong risk factors, the amount smoked and duration smoked. We used data from a population-based case-control study with 282 histologically confirmed lung cancer cases matched to 3,282 random controls to determine whether age at initiation of smoking plays an independent role in the occurrence of lung cancer. After controlling for age, sex, and amount of tobacco exposure, men who began to smoke before age 20 had a substantially higher risk of developing lung cancer [odds ratio (OR) = 12.7; 95% confidence interval (CI) = 6.39-25.2] compared with men who began smoking at age 20 or older (OR = 6.03; 95% CI = 2.82-12.9). For women, the heavy increase in risk continued until age 25 (OR = 9.97; 95% CI = 4.68-21.2) compared with women who began smoking at age 26 or older (OR = 2.58; 95% CI = 0.53-12.4). There was no predisposition toward a specific histologic type of lung cancer. In this study, up to 52.4% of lung cancer cases in men and up to 73.0% of lung cancer cases in women could be attributed to this effect of early age of first smoking.