Effects on Resource Utilization of Adding Salmeterol in Combination or Separately to Inhaled Corticosteroids

BACKGROUND: The addition of a long-acting beta-agonist (LABA) to an inhaled corticosteroid (ICS) for patients with moderate or severe persistent asthma improves outcomes such as pulmonary function, reduces exacerbations requiring oral steroids, and reduces use of rescue beta-agonists. OBJECTIVES: To assess the key resource utilization outcomes of adding salmeterol, a LABA, to fluticasone, an ICS, either as a fixed-combination inhaler (fluticasone-salmeterol [FSA] or as a separate inhaler used concomitantly with the ICS beclomethasone (BSA). METHODS: This is a retrospective, observational database study that extracted data from electronic medical and prescription records in which the prescription written was identical to the prescription dispensed. The sample included asthmatic patients aged 12 to 55 years who received a medium dose of an ICS (240-480 mcg of beclomethasone, 264-660 mcg of fluticasone, 600-1,200 mcg of budesonide, or 1,000-2,000 mcg of triamcinolone acetonide) between July 2001 and December 2002 and had salmeterol added to the regimen (index date). From this population of patients, the analytical cohort was derived to include 1,213 patients who received FSA and a matched cohort of 1,213 patients who received BSA. The primary endpoint was an asthma-related event (ARE), which was defined as (1) an emergency department (ED) visit or (2) hospital admission with a primary asthma diagnosis code (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] code 493.xx). The secondary endpoints were the (1) use of short-acting beta-agonist (SABA) equivalents, (2) percentage of patients who received 1 or more oral steroid prescriptions, (3) patterns of ICS use, and (4) refill rates of salmeterol. All data were collected for 6 months before and 6 months after the index date, defined as the first prescription dispensed to the patient that included salmeterol as an ingredient. RESULTS: Outcomes were improved in both cohorts with no significant difference in the likelihood of an ARE, 60 patients (4.9%) for FSA and 90 patients (8.1%) for BSA (odds ratio [OR], 0.668; 95% confidence interval [CI], 0.443-1.008); P=0.055). FSA was associated with a reduction in AREs of 55% (10.9%-4.9%; P less than0.001), and BSA with a reduction in AREs of 39% (13.3%-8.1%; P less than0.001). FSA compared with BSA was associated with a greater reduction in SABA use (-0.66 canister equivalents over 6 months, P less than0.001) and a lower likelihood of filling an oral steroid prescription, 35.8% of FSA patients compared with 38.0% of BSA patients (OR, 0.801; 95% CI, 0.662-0.970; P=0.023). For the 132 FSA patients (10.9%) and 162 BSA patients (13.4%) who had an ARE in the preperiod, those who received FSA in the postperiod had a 47% lower likelihood of a subsequent ARE, 17.4% of 132 patients compared with 27.8% of 162 BSA patients (OR, 0.527; 95% CI, 0.291-0.954; P=0.034). No ARE differences in subgroup analyses were noted for patients without an ARE in the preperiod or for patients using more than 6 canisters of SABA. More patients in the FSA group took daily doses of 400 mcg or more of ICS than those in the BSA group (32.0% compared with 10.0%, P less than0.001). The average refill rate for salmeterol was 2.71 prescriptions (SD=1.42) over 6 months for FSA compared with 2.38 (SD=1.49) for BSA (P less than 0.001). CONCLUSIONS: Overall, the addition of salmeterol as a fixed combination with fluticasone or with beclomethasone as separate inhalers was associated with a reduction in the ARE rate. Patients who received FSA were more likely to be exposed to a higher dose of ICS compared with those who received BSA. Differences in resource utilization may be attributed to how these drugs are prescribed and taken by patients in a real-practice (naturalistic) setting rather than to any inherent difference between the drugs (i.e., higher ICS dose rather than greater efficacy).