To evaluate outcomes associated with six treatment regimens with theophylline versus each regimen without theophylline in patients with chronic obstructive pulmonary disease (COPD).
Retrospective cohort study.
Veterans Affairs health care system.
A total of 183,573 patients aged 45 years or older who had a diagnosis of COPD and were receiving respiratory drug therapy.
Patients' treatment regimens were identified by using data from October 1, 2002-March 31, 2003, and patients were followed for events by using data from April 1, 2003-September 30, 2005. Data from October 1, 2001-September 30, 2002, were used to define the patients' baseline characteristics. Primary outcome measures were all-cause mortality, COPD exacerbations, and COPD-related hospitalizations. Two approaches were used: first, treatment assignment was based on drug therapy at baseline, and second, exposure was measured as a time-varying covariate. Treatment groups were stratified based on propensity to receive theophylline. Mortality was compared by using Cox proportional hazards models, and other outcomes were compared with use of negative binomial models. Comparisons were conducted within individual treatment regimens that were the same with the exception of theophylline. Patients treated with ipratropium plus theophylline (largest group) compared with those treated with ipratropium alone had a 1.11-fold increase in the risk of death (95% confidence interval [CI] 1.04-1.18). For each of the other regimens, the risk of mortality associated with theophylline was greater than that in the regimens without theophylline (hazard ratios [HRs] 1.17-1.31). In the time-varying exposure analysis, theophylline (HR 1.23, 95% CI 1.09-1.39) was associated with an increased mortality risk.
Patients receiving regimens that included theophylline had slightly increased risks of mortality, COPD exacerbations, and COPD hospitalizations compared with patients receiving the same regimens without theophylline. However, the benefits of theophylline on other factors, including symptoms, quality of life, and activities of daily living, were not measured. Clinicians should consider all of the potential benefits and harms associated with theophylline when making treatment recommendations.