The association between obesity and bronchial hyperresponsiveness (BHR) is incompletely characterised. Using the 2006 follow-up of the Tasmanian Longitudinal Health Study, we measured the association between obesity and BHR and whether it was mediated by small airway closure or modified by asthma and sex of the patient. A methacholine challenge measured BHR. Multivariable logistic regression measured associations between body mass index (BMI) and BHR, adjusting for sex, asthma, smoking, corticosteroid use, family history and lung function. Mediation by airway closure was also measured. Each increase in BMI of 1 kg m-2 was associated with a 5% increase in the odds of BHR (OR 1.05, 95% CI 1.01-1.09) and 43% of this association was mediated by airway closure. In a multivariable model, BMI (OR 1.06, 95% CI 1.00-1.16) was associated with BHR independent of female sex (OR 3.26, 95% CI 1.95-5.45), atopy (OR 2.30, 95% CI 1.34-3.94), current asthma (OR 5.74, 95% CI 2.79-11.82), remitted asthma (OR 2.35, 95% CI 1.27-4.35), low socioeconomic status (OR 2.11, 95% CI 1.03-4.31) and forced expiratory volume in 1 s/forced vital capacity (OR 0.86, 95% CI 0.82-0.91). Asthma modified the association with an increasing probability of BHR as BMI increased, only in those with no or remitted asthma. An important fraction of the BMI/BHR association was mediated via airway closure. Conflicting findings in previous studies could be explained by failure to consider this intermediate step.