The lack of an effect of budesonide on airway inflammation may be considered to be due to the small number of patients who had successful sputum inductions at each visit. However, this does not seem likely for two reasons. First, we have shown in previous studies significant effects of corticosteroid treatment on sputum indexes within and between small groups of subjects , which were paralleled by improvement in clinical parameters. Second, this lack of an effect of budesonide in this study was accompanied by an absence of any clinically important effect on cough discomfort, irrespective of whether the analysis was on the entire sample or the subgroup of patients with successful sputum production. We also considered whether the lack of effect of budesonide was due to the inclusion of patients with a longstanding cough or to the short duration of budesonide treatment. These seems unlikely because prolonged cough, after only one week of treatment with an inhaled steroid, has been highly responsive to treatment in subjects with a nonasthmatic cough associated with sputum eosinophilia .