BACKGROUND: Evidence on the longitudinal association of airway responsiveness with respiratory diseases is scarce. The best indicator of responsiveness is still undetermined. OBJECTIVE: We investigated the association of airway responsiveness with the incidence of asthma, chronic obstructive pulmonary disease (COPD), and allergic rhinitis. METHODS: We studied 3851 subjects who underwent spirometry and methacholine challenge tests both at baseline (1991-1993), when they were 20 to 44 years old, and at follow-up (1999-2002) in the European Community Respiratory Health Survey. Airway responsiveness was defined based on the methacholine dose-response slope on both occasions. Incidence rate ratios for the association of airway responsiveness with disease occurrence were computed by using Poisson regression. RESULTS: With respect to reference (slope of the fourth quintile or greater), subjects with the greatest degree of airway responsiveness (slope less than the first quintile) showed the greatest risk of developing asthma, COPD, and allergic rhinitis (incidence rate ratios of 10.82, 5.53, and 4.84, respectively; all P < .01). A low slope predicted disease occurrence, even in subjects who did not reach a 20% decrease in FEV1 at the cumulative dose of 1 mg of methacholine (PD20 >1 mg). A decrease in slope over time was an independent predictor of disease risk. CONCLUSION: Airway responsiveness predicted new-onset asthma, COPD, and allergic rhinitis. Our study supports the use of a continuous noncensored indicator of airway responsiveness, such as the slope of the methacholine dose-response curve, in clinical practice and research because it showed clear advantages over PD20.