Economic assessment of early initiation of inhaled corticosteroids in chronic obstructive pulmonary disease using propensity score matching

Abstract

Background: Chronic obstructive pulmonary disease (COPD), characterized by airway obstruction and inflammation leading to chronic bronchitis and emphysema, is an important cause of morbidity, mortality, and increased health care utilization and expenditures. Objective: Observational data were used to examine potential benefits of inhaled corticosteroid (ICS) treatment initiated earlier than the current guideline-recommended stepwise approach among patients with COPD diagnosed between 1998 and 2004. Methods: Data for this retrospective cohort study of COPD patients aged ≥40 years were drawn from a large managed care claims database. All patients received ICS therapy in addition to regular inhaled bronchodilator treatment. Annual exacerbation events, defined by COPD-related resource utilization (hospitalizations, emergency department visits, and/or unplanned office visits) and expenditures, were compared between patients with early initiation (ie, within 3 months of starting regular bronchodilators) and those who began ICS treatment later. Propensity score matching was used to identify treatment and control groups, controlling for baseline characteristics that could influence treatment choices and outcomes. Logistic and negative binomial regression models were used to estimate differences in utilization. Differences in medical, pharmacy, and total expenditures were estimated using a generalized linear model. Results: The study included 7712 patients (4146 females, 3566 males; mean age, 61.6 years). The mean duration of follow-up was 1040 days (2.8 years). Early ICS initiation was associated with a lower likelihood of mild (49.6% vs 56.3%; P < 0.001) and severe (20.9% vs 27.9%; P < 0.001) exacerbation events compared with later ICS initiation. Patients who began ICSs within 3 months had higher annual pharmacy expenditures ($1283 vs $1139; mean difference [95% CI], $145 [$106 to $187]) to treat COPD conditions. However, because of lower medical services use, early ICS initiation resulted in lower COPD-related medical expenditures ($1926 vs $2498; mean difference [95% CI], -$572 [-$835 to -$319]) and total expenditures ($3208 vs $3635; mean difference [95% CI], -$427 [-$704 to -$159]) per person annually. Early ICS initiation also was associated with lower all-cause medical expenditures but not lower total all-cause expenditures. Conclusions: ICS treatment initiated along with bronchodilators reduced COPD exacerbation events and expenditures in these managed care plans. Further research is needed to address potential selection bias due to unobserved factors.