Systemic corticosteroids are almost universally used in the treatment of severe acute asthma but the optimum length of treatment with corticosteroids following recovery from an acute attack of asthma is not established. Thirty-five patients admitted with acute asthma and treated with oral prednisolone 40 mg daily in addition to bronchodilator therapy until full recovery, with stable peak expiratory flow recordings (PEF) within 15% of their previous best PEF or predicted PEF were studied. They were all discharged home on regular inhaled corticosteroids and regular or as required use of bronchodilators and randomized to receive either prednisolone 40 mg daily or placebo for the first 14 days. Median PEF values increased from 31% predicted on admission to hospital to 71% predicted on discharge from hospital in the active treatment group (19 patients) and from 32-73% in the placebo group (16 patients). There was no difference between the two groups in the median values of the forced expiratory volume in one second, forced vital capacity, total lung capacity or diurnal variation in PEF either at the time of discharge from hospital or at 14 and 28 days after discharge from hospital. This study suggests that there is no need to reduce prednisolone gradually following recovery from an exacerbation of asthma, provided systemic corticosteroid treatment is continued until a satisfactory and stable PEF is achieved.