High dose inhaled corticosteroids may cause suppression of the hypothalamo-pituitary-adrenal (HPA) axis. Several tests are available to screen for this suppression but it is not clear which is the most useful. HPA function was assessed in 78 adult asthmatics inhaling long-term, high dose (median 1600 micrograms; range 1200-2650 micrograms) beclomethasone dipropionate (n = 69) or budesonide (n = 9). Screening tests performed in all patients were 9 am serum cortisol, short tetracosactrin test and 24-h urine free cortisol excretion. Eleven patients also underwent insulin stress tests. Subnormal results were: 9 am cortisol less than 190 nmol l-1; urine free cortisol less than 80 nmol 24 h-1; rise in cortisol in response to tetracosactrin or hypoglycaemia less than 200 nmol l-1 and/or achieved cortisol less than 500 nmol l-1. HPA suppression (defined as subnormal results of at least two of the three initial tests and/or subnormal response to hypoglycaemia), was found in 16 patients. In the 11 patients who underwent insulin stress tests, results of all initial tests were normal in three, one test was abnormal in three and two tests were abnormal in four patients. All three tests were abnormal in the remaining patient. The response to hypoglycaemia was normal in the three patients whose screening tests were all normal; HPA suppression was present in seven patients and one patient had a borderline result. Close correlation was observed between the maximum cortisol during hypoglycaemia and both urine free cortisol (rs = 0.84; P = 0.001) and post-tetracosactrin cortisol (r = 0.75; P less than 0.01).(ABSTRACT TRUNCATED AT 250 WORDS)