Asthma-related health care resource use among patients starting fluticasone or montelukast therapy

Study Objectives. To compare patterns of asthma-related health care resource use among patients prescribed fluticasone or montelukast as single-controller therapy for asthma, and to confirm patterns previously observed in retrospective analyses examining outcomes among patients receiving fluticasone or montelukast for asthma. Design. Retrospective cohort study. Data Source. Administrative claims data drawn from United States health insurers in 35 states, covering 17 million privately insured patients. Patients. A total of 4758 patients aged 2-55 years with asthma who were prescribed either fluticasone or montelukast from July 1, 1998-June 30, 1999, were continuously enrolled for at least 24 months, had no evidence of controller therapy for 6 months before the start of drug therapy, and had no evidence of chronic obstructive pulmonary disease or other respiratory illness. Measurements and Main Results. Patients were identified using an algorithm based on medical and pharmacy insurance claims. Patients were matched between groups based on a propensity score of clinical characteristics and age; this resulted in 1512 patients/treatment group. Asthma-related health care claims incurred for 12 months before and after the start of controller therapy were analyzed. After adjustment, the fluticasone-treated group had greater risk than the montelukast-treated group of requiring therapy with a short-acting beta-agonist in the follow-up period (relative risk 1.12, 95% confidence interval [CI] 1.03-1.20). Odds were similar across treatment groups for needing an emergency department visit and/or hospitalization (odds ratio 1.08, 95% CI 0.74-1.58) and for needing therapy with an oral corticosteroid (odds ratio 1.02, 95% CI 0.84-1.26). Conclusion. The start of therapy with either fluticasone or montelukast as a single-con troller for asthma was associated with similar asthma-related health care resource use in this matched population.