Abstract
OBJECTIVE: To examine dosing patterns and costs among rheumatoid arthritis (RA) patients newly treated with infliximab in a large national health care claims database. METHODS: Using data from a proprietary database of pharmacy and medical claims for 75 U.S. health plans, RA patients newly treated with infliximab between June 2000 and June 2002 were selected and assigned an .index date. based on the first infusion. A pretreatment period of 6 months was created; patients were also followed for a minimum of 6 months after the initial infusion. Follow-up was allowed to vary beyond this minimum 6 months in order to preserve all available patient data. A maintenance number of infliximab vials was determined as of the second infusion; patients with 1 subsequent increase in vials used or 2 intervals between infusions of <49 days were considered to have had an upward dose adjustment. Differences (i.e., between those with upward dose adjustment and those with no upward dose) in patient characteristics were examined using descriptive statistics. In addition, time to upward dose adjustment and factors influencing its likelihood were analyzed using Kaplan-Meier and Cox proportional hazards techniques. Finally, differences in RA-related and unrelated costs (medication, outpatient, inpatient, and total, expressed in 2003 dollars) were examined using Wilcoxon rank-sum tests and were also stratified by a number of patient characteristics found to differ between the 2 groups. RESULTS: A total of 1,236 patients met all study entry criteria and were included in these analyses. One or more upward dose adjustments were experienced by 61.7% (N=762) of patients during an average of 15 months of follow-up (median =13 months, range=6 to 31 months). The majority (63.3%) of upward dose adjustments were due to increases in the number of billed vials. Median time to upward dose adjustment was 254 days and declined steadily based on year of initiation (from 330 days in 2000 to 224 days in 2002). Factors significantly influencing upward dose adjustment included pretreatment use of leflunomide, comorbid Crohn.s disease, and pretreatment liver function testing. During followup, patients in the upward dose adjustment group used a mean (SD) of 30.28 (20.90) vials of infliximab, compared with 15.90 (14.28) among those not adjusting dose (P<0.001). Annualized (i.e., standardized to a 365-day rate) RA-related costs were higher by more than 50% among patients with upward dose adjustment (SD $22,283 [$20,517] versus $14,425 [$10,828] for those without upward dose adjustment; P<0.001); differences were driven almost entirely by the costs of infliximab itself ($16,336 [$9,490] versus $9,573 [$6,790], P<0.001). CONCLUSIONS: In a cohort of managed care members with RA, upward dose adjustment with infliximab was frequent and appeared to occur earlier in the drug therapy in 2002 compared with 2000. Upward dose adjustment was associated with significant increases in drug treatment costs; therefore, payers and providers should consider the impact of current dosing trends when monitoring the use of biologics for autoimmune diseases.