Airway hyperresponsiveness (AHR) constitutes a risk for development of respiratory symptoms. We assessed whether blood eosinophilia (>/= 275 eosinophils/microliters), skin test positivity (sum score >/= 3) and cigarette smoking (never, ex-smoker, 1-14 cig/d, 15-24 cig/d, >/= 25 cig/d) at the first of two successive surveys are related to the development of respiratory symptoms (chronic cough or phlegm, bronchitis, persistent wheeze, dyspnea, and asthma) at the second survey, and whether these relations are the same in subjects with (PC10 </= 8 mg/ml histamine) and without AHR. We analyzed data of the longitudinal Vlagtwedde-Vlaardingen Study (1965 to 1990) using logistic regression analyses with paired observations, taking multiple measurements within a person into account. In total, 995 men and 792 women contributed 4,403 paired observations. Eosinophilia in hyperresponsive subjects significantly increased the risk to develop one or more respiratory symptoms (odds ratio [OR] = 3.67, 95% confidence interval [CI] = 1.79 to 7.52), wheeze (OR = 5. 06, 95% CI = 2.11 to 12.13), and dyspnea (OR = 2.73, 95% CI = 1.13 to 6.60), independent of smoking, age, sex, area of residence, and time between two successive surveys. Smoking at the first of two successive surveys increased the risk to develop symptoms in a dose-dependent relation. Subjects with positive skin tests in the past were less likely to develop one or more respiratory symptoms (OR = 0.64, 95% CI = 0.46 to 0.88) and chronic phlegm (OR = 0.65, 95% CI = 0.42 to 1.00), independent of AHR. This longitudinal study in the general adult population shows that cigarette smoking and hyperresponsive subjects are at increased risk to develop respiratory symptoms, and especially so when eosinophilia is present in hyperresponsive persons.