BACKGROUND: The institution of inhaled corticosteroids is generally advocated for effective treatment of patients with asthma. It is yet unknown what is the best time to start inhaled corticosteroid therapy and especially whether delayed introduction is harmful. PHASE 1: In a previous study in patients with asthma and COPD, we found that 2.5 years of treatment with a beta 2-agonist plus inhaled corticosteroid (BA + CS) was more effective in improving the FEV1 and the provocative concentration of histamine causing a 20% reduction in FEV1 (PC20) than treatment with a beta 2-agonist plus anticholinergic (BA + AC) or placebo (BA + PL). PHASE 2: We extended this study with 6 months to investigate whether delayed introduction of inhaled CS therapy (800 micrograms beclomethasone dipropionate) in the groups previously not treated with inhaled CS (BA +/- AC) could also improve FEV1 and PC20 to the same degree. A distinction was made between patients with predominantly asthma (high baseline reversibility, delta FEV1 > or = 9% of predicted), and predominantly COPD (low baseline reversibility, delta FEV1 < 9% of predicted). RESULTS: Improvement of FEV1 percent predicted by inhaled CS was comparable both in the asthmatics between phase 1 (13.8% predicted) and phase 2 (8.5% predicted; p = 0.31) as well as in the patients with COPD (2.5% and 1.5% predicted, respectively). PC20, however, increased significantly more in the asthmatics in phase 1 (1.77 doubling concentration [DC]) than in phase 2 (0.79 DC; p = 0.03). Improvement of PC20 in the patients with COPD was not significantly higher in phase 1 (0.74 DC) than in phase 2 (0.00 DC; p = 0.19). CONCLUSIONS: Our study indicates that although delayed introduction of inhaled CS in asthmatics leads to similar improvements in FEV1, improvements in PC20 are significantly less. These findings in patients with longer-existing asthma concur with other findings in newly detected asthma. We suggest that institution of inhaled CS therapy should not be postponed in asthmatics with ...