We conducted a case-control study to measure the risk of pulmonary side effects following the use of topical glaucoma medications, particularly the beta blockers, and to identify patients at highest risk for such side effects. Patients over age 55 who were active users of the New Jersey Medicaid system during the years 1981-1989 composed the study population. Two case groups were identified: (a) 21,096 patients who began new use of bronchodilator medications and (b) 3,382 patients who were users of a xanthine bronchodilator or inhaled steroid whose regimen was intensified through addition of another medication (generally a sympathomimetic). Controls were Medicaid enrollees with documented use of the Medicaid system. Use of glaucoma medications was measured in the 45 days preceding occurrence of the outcome for cases, and during a comparable period for controls. We found no consistent association between ongoing glaucoma therapy and new use of a bronchodilator; a definitive null result was obtained for timolol (odds ratio = 0.95; 95% CI = 0.84-1.09). However, nonselective topical beta blockers remained commonly used among patients already on a bronchodilator regimen. Timolol users were 47% more likely to require addition of another class of bronchodilator than were patients using no glaucoma therapy (odds ratio = 1.47; 95% CI 1.04-2.09; p = 0.03). No increase in risk was found for other glaucoma drugs. These findings suggest that patients with bronchospasm requiring xanthines or inhaled steroids who are prescribed topical timolol for glaucoma may be at significantly increased risk of pulmonary symptom exacerbation requiring additional bronchodilator therapy. No risk was found for the initiation of bronchodilators in previously untreated patients.