FEV(1) is an objective measure of airflow obstruction used in clinical practice and in therapeutic trials. The precise relationship of FEV(1) to clinical outcomes is generally uncertain. As part of a randomized trial to assess systemic corticosteroid efficacy, we obtained serial FEV(1) measurements in patients hospitalized for exacerbations of chronic obstructive pulmonary disease (COPD). Over the first 14 Study Days at least one FEV(1) value was obtained in 261 subjects. Sixty-four of these subjects experienced treatment failure, defined as death, intubation, readmission for COPD, or intensification of drug therapy, by Study Day 30. After adjustment, both FEV(1) at entry into the study (odds ratio [OR] for a 100-ml increase, 0.87; 95% confidence interval [CI], 0.79 to 0.96) and change in FEV(1) over the first two Study Days (OR for a 100 ml increase, 0.80; 95% CI, 0.69 to 0.92) predicted treatment failure. We identified no baseline characteristic that was significantly related to FEV(1) at entry into the study. Assignment to the systemic corticosteroid treatment arm was associated with a significantly larger FEV(1) at Study Day two (p = 0.01). We conclude that FEV(1) measurements at admission and over the first several days of hospitalization are highly predictive of clinical outcomes during exacerbations of COPD.