The American Journal of Managed Care June 2011
Impact of Persistence With Infliximab on Hospitalizations in Ulcerative Colitis
Objectives: To assess infliximab infusion patterns in ulcerative colitis (UC) and assess the impact of persistence with infliximab maintenance therapy on UC-related hospitalizations, lengths of stay, and inpatient costs.
Study Design: Retrospective analysis of medical claims for UC patients newly initiating infliximab treatment.
Methods: Patients were aged >18 years and had 2 UC diagnosis codes, an infliximab index date between September 1, 2005, and January 31, 2008, and continuous enrollment for >12 months before and >14 months after the index date. Infliximab induction (first 56 days postindex) and maintenance (>56 days and <12 months postinduction) patterns were evaluated. Of patients with maintenance treatment, persistence was defined as a medication possession ratio (MPR) of >80%, and this group was compared with those without persistence (<80% MPR).
Results: Overall, 420 patients were included in the analysis; 84.3% (n = 354) continued to maintenance therapy. Maintenance infusion patterns were consistent with recommended prescribing information. A smaller proportion of patients with maintenance therapy persistence required hospitalization compared with patients without persistence (3.0% vs 20.4%; P <.001). Hospitalized patients with maintenance therapy persistence had significantly lower mean inpatient costs ($14,243 vs $32,745; P = .046), with a trend toward shorter mean lengths of stay (6.67 vs 9.71 days; P = .147) than patients without persistence.
Conclusions: Infliximab maintenance therapy persistence in UC was associated with significantly fewer hospitalizations. Once hospitalized, patients with therapeutic persistence had significantly decreased inpatient costs.
(Am J Manag Care. 2011;17(6):385-392)
The findings of this observational study support approved labeling for infliximab and clinical trial data showing that induction and subsequent maintenance infliximab therapy may reduce negative outcomes in ulcerative colitis (UC).
- Infliximab induction and maintenance infusion patterns during the first year of therapy were consistent with prescribing recommendations.
- Therapeutic persistence with maintenance therapy was associated with fewer hospitalizations and significantly decreased inpatient costs.
- These data might suggest that UC patients who respond to infliximab should be monitored to ensure that appropriate infliximab maintenance regimens are followed.
The anti-tumor necrosis factor therapy infliximab has been shown to effectively induce response and remission in patients with moderately to severely active UC, as demonstrated in the Active Ulcerative Colitis Trials (ACT 1 and ACT 2).5 Infliximab was also shown to reduce hospitalizations and surgeries in the ACT studies, and there was a significant overall risk reduction of 7% for colectomy.6 Infliximab was initially approved by the US Food and Drug Administration (FDA) in September 2005 for reduction of signs and symptoms in UC, achievement of clinical remission and mucosal healing in UC, and corticosteroid elimination in patients with moderately to severely active UC who have had an inadequate response to conventional therapy.7 A subsequent expanded indication for infliximab was announced in October 2006 for the maintenance of clinical remission and mucosal healing in UC.7 The FDA dosing guidelines for infliximab in adults with moderately to severely active UC is 5 mg/kg given as an intravenous induction regimen at 0, 2, and 6 weeks, followed by a maintenance regimen of 5 mg/kg every 8 weeks thereafter.7
Unfortunately, few data studying the dosing patterns of infliximab and the effects of maintenance dosing on hospitalizations, surgeries, and other negative outcomes in UC exist outside clinical trials. In this study, we sought (1) to describe the induction and maintenance dosing patterns of infliximab in the treatment of UC and (2) to assess the impact of complete induction and persistence with maintenance therapy on hospitalization for UC.
Study Design and Sample Selection
We conducted a retrospective analysis using medical and pharmacy claims from the IMS LifeLink Health Plan Claims (US) Database received between September 1, 2004, and March 31, 2009. Infliximab claims were identified using the Healthcare Common Procedure Coding System code J1745; for each patient, the date of the first infliximab infusion claim between September 1, 2005, and January 31, 2008, was deemed the index date. Patients were included if they were aged >18 years, had newly initiated treatment with infliximab (as identified by a 6-month absence of claims for infliximab prior to the index date), and had at least 2 medical claims with an International Classification of Disease, 9th Revision (ICD-9) diagnostic code for UC (ICD-9 code 556.x) prior to the index date. Continuous enrollment for 12 months before (preindex) and 14 months after (postindex) the first infliximab infusion claim was required. Patients were excluded if they had more than 3 infliximab claims during the first 56 days postindex; 2 claims with a diagnosis of Crohn’s disease (ICD-9 code 555.x) during the preindex period; or 1 claim with a diagnosis of psoriatic arthritis (ICD-9 code 696.0), psoriasis (ICD-9 code 696.1), rheumatoid arthritis (ICD-9 codes 714.x), or ankylosing spondylitis (ICD-9 code 720) during the preindex period. To be consistent with the ACT study exclusion criteria, patients with evidence of either a partial or complete colectomy within the first 12 weeks postindex were also excluded from the analysis. Partial colectomies were identified by the following Current Procedural Terminology (CPT) codes: 44139-44141, 44143-44147, 44204-44208, 44213, and 45123. Complete colectomies were identified by CPT codes 44150, 44151, 44155-44158, 44160, 44210-44212, 45121, 44152, and 44153.
Measures and Analyses
Infliximab Induction and Maintenance Doses. Infliximab induction doses were defined as infusion claims received during the first 56 days postindex. Infliximab maintenance doses were defined as any infusion claims received after 56 days postindex and less than 12 months postinduction. Patients with no infusions after 56 days postindex were considered to have no maintenance therapy. Patients with at least 1 infusion after 56 days postindex were considered to have maintenance therapy. Results for those patients receiving maintenance therapy were stratified by the number of inductiondoses received (ie, 1, 2, or 3) during the 56-day postindex period. The number of infusions and time (days) between infusions were reported for patients receiving maintenance therapy.
Treatment Persistence. Treatment persistence was calculated for the maintenance period as a medication possession ratio (MPR), where the total days of supply of infliximab administered during the maintenance period was divided by 360. Patients were required to have at least 1 infliximab infusion claim after day 56, thus indicating maintenance treatment, for the MPR to be calculated. The MPR was calculated through the 14-month postindex period (ie, 12 months of maintenance therapy postinduction). Results for patients with therapeutic persistence (>80% MPR) were compared with results for patients without therapeutic persistence (<80% MPR) during maintenance therapy. Prior adherence research has recognized an MPR threshold of 80% as an accepted and standard definition of adherence in UC.8
Ulcerative Colitis–Related Hospitalizations, Lengths of Stay, and Inpatient Costs. The number of UC-related hospitalizations (ie, inpatient discharge claims with a UC diagnosis in any position on the claim), lengths of UC-related hospitalizations (days), and the associated UC-related inpatient costs were evaluated for the total population for the 14 months after the index infliximab infusion and for the 12 months postinduction for those patients with maintenance therapy. Payments and admissions were evaluated with health plan–paid claims. Ulcerative colitis–related hospitalizations and costs were compared for patients with and without infliximab maintenance therapy. For those patients receiving maintenance therapy, UC-related hospitalizations and costs were compared for patients with and without therapeutic persistence.
Univariate statistics were generated for the preindex and postindex infliximab periods. Categorical variables were summarized with frequency counts and percentages, with statistical differences assessed via c2 tests. Continuous variables were represented by means, medians, and standard deviations, with statistical differences assessed via nonparametric Wilcoxon tests. Statistical significance was defined at a 2-sided level of .05 or less. All statistical analyses were performed using SAS release 8.2 (SAS Institute Inc, Cary, North Carolina). Multivariate models were used to assess the relationship between continuing to maintenance or having therapeutic persistence with infliximab and UC-related hospitalizations and inpatient costs. Logistic regression was used for the risk of hospitalization, and generalized linear models with log-link function and gamma distribution were used for the inpatient costs. The multivariate models controlled for baseline demographic and utilization characteristics, including sex, age, geographic location, baseline UC-related hospitalizations, and baseline UC-related outpatient costs.
Patient Demographics and Cohort Assignments
A total of 420 UC patients receiving infliximab (maintenance, n = 354; no maintenance, n = 66; therapeutic persistence, n = 202; no therapeutic persistence, n = 152) met the selection criteria and were included in the analyses (Figure 1). Overall, the mean (SD) age was 43.9 (14.1) years, and 47.9% were female. The majority of patients had health plan coverage with a commercial payer (97.6%). Most (72.8%) received their index infliximab infusion in an outpatient setting (ie, physician office or outpatient hospital). Demographic characteristics were reported for the overall UC group and by cohort (ie, maintenance with 1, 2, or 3 induction doses; no maintenance; therapeutic persistence; and no therapeutic persistence). No statistically significant differences in demographic characteristics were observed (Tables 1A and 1B).
Infliximab Induction and Maintenance Doses
Overall, patients received a mean (SD) of 6.08 (2.49) infliximab infusions during the 12 months postindex. The median number of infliximab infusions postindex was 7. Patients receiving infliximab maintenance therapy had a mean (SD) of 6.83 (1.91) cumulative infusions during the 12-month postindex period, whereas those with only induction infusions had a mean (SD) of 2.05 (0.83) infusions. Of those continuing to maintenance therapy, a mean (SD) of 5.87 (1.93) infusions were administered during the maintenance phase (ie, 12 months postinduction). In addition, patients with therapeutic persistence during maintenance therapy had a higher mean (SD) cumulative number of infusions (7.22 [0.85]) compared with patients without therapeutic persistence (4.08 [1.44]) (Figure 2).
Of the 354 patients with UC receiving maintenance therapy, the majority (68.9%) had 3 infliximab infusions during the induction period, while 23.4% had 2 induction infusions and 7.6% had only 1 induction infusion. The overall mean (SD) number of days during the induction period was 35 (14), and days during the induction period increased with the number of induction doses received (1 induction dose = 1 day ; 2 induction doses = 24.9 days [12.9]; 3 induction doses = 42.4 days [4.5]). The median time between infusions during the first year postinduction was 56 days. Table 2 reports the times between infusions for the overall group and by number of induction doses received.
Ulcerative Colitis–Related Hospitalizations, Lengths of Stay, and Inpatient Costs
Ulcerative colitis–related hospitalizations and their associated costs for all patients are reported in Table 3. Among all patients, those continuing to maintenance therapy incurred fewer mean hospitalizations (0.12 vs 0.32) and lower mean inpatient costs ($3118 vs $8610) than those patients with induction infusions only. Furthermore, those patients with maintenance therapy and therapeutic persistence had fewer mean hospitalizations (0.03 vs 0.22) and lower mean inpatient costs ($423 vs $6678) than those without therapeutic persistence.