When the effectiveness of asthma interventions are evaluated in the research setting, the physiologic manifestation of asthma-variable airways obstruction-is always objectively measured by some of the following pulmonary function tests: (1) Baseline spirometry gives a highly accurate "snapshot" of asthma severity and the degree of airways obstruction. The FEV1, derived from spirometry, is the most reproducible pulmonary function parameter and is linearly related to the severity of airways obstruction. There are no contraindications for the test, spirometers are widely available at reasonable cost, and methods and result interpretation are comprehensively standardized. (2) The post-bronchodilator FEV1 measures the best lung function that can be achieved by bronchodilator therapy on the day of the visit and therefore is a more stable measure in asthmatics than comparing visit-to-visit baseline FEV1. Although a positive acute response to bronchodilator helps to confirm the diagnosis of asthma, the degree of bronchodilator reversibility from visit-to-visit (change in reversibility) is not a useful index of asthma outcome. (3) Airway responsiveness (bronchial challenge) measures the degree to which an individual withstands nonspecific stimuli that trigger asthmatic attacks. The methacholine challenge test is safe and requires less than an hour, but it requires more technical skill than baseline spirometry and is contraindicated in some situations. (4) Ambulatory monitoring, using peak flow meters or hand-held spirometers, provides multiple measurements of the degree of obstruction for days to weeks in the patient's natural setting. PEF meters are very inexpensive and almost all asthmatics can use them, but PEF results are less reliable than the FEV1. The often asymptomatic obstruction of an asthmatic has both short-term (within a day and day-to-day) and longer-term variations that are triggered by naturally occurring stimuli. These changes are measured by PEF lability but not by spirometry during clinic visits. (5) Other pulmonary function tests, such as absolute lung volumes and airways resistance, may provide confirmatory data, but the instruments are large, expensive, and technically demanding. The results of all the above pulmonary function tests are significantly correlated with each other and with symptom scores and medication use in large groups of patients with widely varying degrees of asthma severity. Since a "gold standard" with which to measure asthma severity does not currently exist, all of these tests contribute an additional amount of unique information when measuring asthma outcome in a clinical trial.