A meta-analysis of randomized controlled asthma drug therapy trials published in the English literature from January 1991 to June 1995 was performed to estimate the magnitude and direction of the placebo effect in stable ambulatory asthmatic patients. Among placebo groups, the mean absolute increase in forced expiratory volume in 1 sec (FEV₁), weighted for-sample size and variance, was 0.11 L/min, and the mean percent increase in FEV₁ was 4.81%. The corresponding placebo group changes in peak expiratory flow (PEF) were in an opposite direction to those of FEV₁; there was a mean absolute decrease of 2.24 L/min, and a mean percent decrease of 4.21%. Changes for active treatment groups were greater in magnitude. However, there were no statistically significant differences in mean changes comparing the placebo groups to the treatment groups, for any of the outcome measures. Mean increases in PEF and FEV, exceeded 10% in 5 of 33 placebo groups, as compared to 13 of 33 active treatment groups. In conclusion, in well-designed long-term drug therapy studies in stable asthmatics the pooled placebo effect is small but measurable, with FEV, and PEF showing different directions of response. Moreover, a modest number of patients receiving placebo have changes in pulmonary function that might be interpreted as clinically significant.