A randomised, double-blind, placebo-controlled study examined the effects of alternate-day prednisone therapy on morbidity and progression of lung disease in cystic fibrosis (CF). At baseline the patients (aged 1-12 years) had mild to moderate lung disease, and the prednisone group did not differ significantly from the placebo group for any values measured. After 4 years, the prednisone-treated group had significant advantages over the placebo group for height, weight, vital capacity, forced expiratory volume in 1 s, peak flow rate, erythrocyte sedimentation rate, and serum IgG. The prednisone-treated group required 9 admissions to hospital for CF-related pulmonary disease compared with 35 for the placebo group. There were no steroid-induced side-effects. To rule out bias in case selection, 69 CF clinic patients comparable in age and clinical status but not included in the study were compared with the placebo group at 4 years; no significant differences between the groups were found.