To test the hypothesis that a greater proportion of women than men react to methacholine challenge and investigate the possible reasons for any differences observed, we recruited 495 subjects 20 to 44 yr of age (50.9% male) in Paris and 304 subjects (51.3% male) in Montpellier (France), as part of the European Community Respiratory Health Survey. The proportion of responders (PD20 < or = 4 mg methacholine) was 33.7% in women and 11.9% in men (odds ratio = 3.8; 95% confidence interval = 2.4-6.0) in Paris and 43.2% in women and 29.5% in men (odds ratio = 1.8; 95% confidence interval = 1.1-2.9) in Montpellier. These differences could not be explained by asthma, respiratory symptoms, atopy, or lung function parameters. In stepwise logistic regressions including sex, asthma, and asthma-like symptoms, nasal allergies, atopy, baseline FEV1, FEV1%pred, FVC, and FEV1%FVC, the odds-ratios for the effect of female sex on PD20 < or = 4 mg methacholine were 5.2 (3.0-9.0) in Paris and 2.2 (1.2-3.8) in Montpellier. Reacting to low doses of methacholine (PD20 < or = 0.5 mg) was associated with asthma and atopy in both men and women. In contrast, reacting to doses between 0.5 and 4 mg was associated with asthma and atopy only in men and with heavy tobacco consumption only in women. We conclude that the excess of hyperresponsiveness in women is not due to their having smaller lung size or airway caliber than men and may be related to a greater susceptibility to smoking.