In view of the ubiquitous practice of using bronchodilator responsiveness to determine suitable patients for clinical trials, we wanted to know whether changes in FEV1 or forced vital capacity (FVC) really were useful in differentiating COPD from asthma. Pulmonary function test results from 450 patients were documented by two technicians who had been asked to select consecutive studies in which flow-volume loops showed an obstructive pattern. The respirologist responsible for the care of each patient was asked to record the clinical diagnosis from the existing outpatient chart using clinical judgment based on American Thoracic Society criteria. In 395 cases, a single, unambiguous diagnosis of asthma or COPD was recorded; this diagnosis then formed the database for subsequent analysis. While the mean change in FEV1 in patients judged to have asthma was different from that found in COPD patients (16.4 vs 10.6 percent, p < 0.01), the change in FVC was similar (9.8 vs 10.3 percent, p > 0.06). However the sensitivities and specificities of postbronchodilator changes in FEV1 (dFEV1) for the diagnosis of asthma were not generally sufficient to diagnose or exclude asthma reliably. The FEV1 correlated better with residual volume (RV) in COPD (r = -0.55 vs r = -0.31), but with total lung capacity (TLC) in asthma (r = 0.51 vs r = -0.09). However, FEV1 correlated better with the RV-TLC ratio than RV or TLC alone in both groups, the correlation in each being similar (asthma, r = -0.72; COPD, r = -0.78). We conclude that acute responses of FEV1 and FVC following a standard dose of inhaled bronchodilator are neither sufficiently sensitive nor sufficiently specific to differentiate asthma from COPD purely on spirometric grounds. Furthermore, neither RV nor TLC reflected degrees of airflow limitation as well as did the RV-TLC ratio.