We studied 68 chronic asthmatic patients, 18 to 76 yr of age, with a percent predicted FEV1 between 33 and 81, comparing cromolyn sodium with placebo. We used a double-blind, comparative group trial design. A 4-wk baseline period was followed by 3 months of active treatment or placebo. Patients recorded symptom severity and frequency of study drug and concomitant medication usage on daily diary cards. At each clinic visit, patients independently assessed the effectiveness of the test medication in controlling their asthma. Physicians also assessed the severity of the patients' symptoms, pulmonary function, and effectiveness of test medication at monthly intervals. Methacholine challenges were done pre- and post-treatment. Use of concomitant therapy was reduced according to a specified schedule. There was significant improvement in the severity of daytime asthma, nighttime asthma, and cough as assessed by the patients in the cromolyn sodium group. Mean use of concomitant medications decreased significantly in cromolyn sodium patients. Despite the reductions in the use of bronchodilators, pulmonary function (FEV1, FVC, FEF25-75) improved significantly in the cromolyn sodium group. Similar improvements did not occur in the control group. The physicians' assessments of symptoms showed significant improvement in favor of the cromolyn sodium group. Both physicians and patients judged cromolyn sodium to be moderately or very effective for 61% of the patients as compared to 27% (by physicians) and 24% (by patients) in the placebo group. There was no significant difference in methacholine response between the two groups, although the mean value for methacholine sensitivity in the cromolyn sodium group was significantly less at the end of the study than at baseline.(ABSTRACT TRUNCATED AT 250 WORDS)