Abstract
Objective
To compare real-world treatment persistence, dose escalation, rates of opportunistic or serious infections, and healthcare costs in patients with Crohn’s disease (CD) receiving vedolizumab (VDZ) vs ustekinumab (UST) in the United States.
Methods
A retrospective observational study in adults with CD initiated on VDZ or UST on/after 26 September 2016, was performed using the IBM Truven Health MarketScan databases (1 January 2009–30 September 2018). Rates of treatment persistence, dose escalation, opportunistic or serious infection–related encounters, and healthcare costs per patient per month (PPPM) were evaluated. Entropy balancing was used to balance patient characteristics between cohorts. Event rates were assessed using weighted Kaplan-Meier analyses and compared between cohorts using log-rank tests. Healthcare costs were compared between cohorts using weighted 2-part models.
Results
589 VDZ and 599 UST patients were included (172 [29.2%] and 117 [19.5%] were bio-naïve, respectively). After weighting, baseline characteristics were comparable between cohorts. No significant difference in rates of treatment persistence (12-month: VDZ, 76.5%; UST, 82.1%; p = .17), dose escalation (12-month: VDZ, 29.3%; UST, 32.7%; p = .97), or opportunistic or serious infection–related encounters were observed between VDZ and UST. Total mean healthcare costs were significantly lower for patients treated with VDZ vs UST (mean cost difference = –$5051 PPPM; p < .01). Findings were consistent in bio-naïve patients.
Conclusions
In this real-world study, similar treatment persistence, dose escalation, and rates of opportunistic or serious infections were observed with VDZ- and UST-treated patients with CD. However, VDZ was associated with a significantly lower cost outlay for healthcare systems.
PLAIN LANGUAGE SUMMARY
Crohn’s disease (CD) causes inflammation in the digestive system. Vedolizumab (VDZ) and ustekinumab (UST) are therapies for patients with CD. Little is known about the clinical outcomes and healthcare costs of VDZ versus UST in the real world in the United States. We used health claims data and found that VDZ and UST had comparable real-world clinical outcomes. After 12 months of treatment, the proportions of patients with CD who stayed on treatment and those who needed to increase therapy dose were similar with VDZ and UST. The rate of infection was also similar between the two groups of patients. However, the monthly healthcare costs were $5051 less for patients treated with VDZ than with UST. This was mainly due to the lower cost of VDZ, which was almost half of that of UST. The lower treatment costs with VDZ may provide substantial savings for the healthcare system and patients specifically. Future cost-effectiveness studies on VDZ and UST are needed to aid treatment selection for patients with CD.
Introduction
Crohn’s disease (CD), one of the major forms of inflammatory bowel disease (IBD), is a chronic immune-mediated condition with increasing prevalence worldwideCitation1. Approximately 1.6 million individuals in the United States (US) were affected by IBD in 2014, and up to 785,000 individuals were estimated to have CDCitation2–6.
CD has been associated with a substantial economic burden. Patients with CD have been shown to have increased use of healthcare resources and an estimated $17,463 in excess annual direct healthcare costs per patient compared with patients without IBDCitation7, with the lifetime cost of CD estimated to be >$600,000 per patientCitation8. Thus, optimizing the management of patients with CD is imperative to reduce the burden of disease on individuals and the healthcare system.
Several biologic therapies have been approved for patients with moderate-to-severe CD, including anti–tumor necrosis factor (TNF) agents (e.g. infliximab, adalimumab, certolizumab pegol), anti-integrin monoclonal antibodies (e.g. vedolizumab [VDZ]), the anti–interleukin-12/23 monoclonal antibody ustekinumab (UST), and the anti–interleukin-23 monoclonal antibody risankizumabCitation9,Citation10. The American College of Gastroenterology and the European Crohn’s and Colitis Organisation have equally endorsed these biologic agents for managing patients with moderate-to-severe CDCitation9,Citation11. However, no preferred choice of first-line biologic has been recommended given the lack of head-to-head clinical trials among anti-TNF agents, VDZ, and UST in this settingCitation9,Citation11.
In an indirect comparison of randomized clinical trial data of VDZ and UST in patients with CD refractory to anti-TNF agents, no difference in clinical response or remission rates was observed between the two drugsCitation12. Real-world evidence on clinical response to VDZ versus UST in CD has also largely been limited to anti-TNF–refractory patients, and conclusions have varied across studiesCitation13–15. Meanwhile, studies evaluating suboptimal clinical response and intolerability to CD treatment have shown that these factors often lead to treatment discontinuation or dose titrationCitation16–18. Additionally, cost has been shown to be an important factor affecting treatment decisions made by patients with IBD in the real worldCitation19. However, limited data exist comparing the clinical and economic outcomes of VDZ versus UST in CD. Therefore, this study aimed to compare indicators of real-world treatment effectiveness, including treatment persistence, dose-escalation rates, and rates of opportunistic and serious infection–related encounters, as well as healthcare costs among patients with CD treated with VDZ versus UST in US clinical practice.
Methods
Data source and study design
A retrospective observational cohort study was performed using claims data from IBM Truven Health MarketScan databases (Commercial and Medicare supplemental) from January 1, 2009, to September 30, 2018. These databases comprise a collection of paid claims for inpatient (IP), outpatient (OP), and pharmaceutical benefits generated annually from nearly 51 million US employees and their dependents. All data from individual patients were deidentified and comply with the Health Insurance Portability and Accountability Act; therefore, no review by an institutional review board was required per Title 45 of CFR, Part 46.101(b)(4).
The index date was defined as the date of the first claim for VDZ or UST therapy following a diagnosis of CD (see Supplementary Methods for the codes considered for CD diagnoses and VDZ or UST therapy). The index treatment was defined as the biologic therapy (VDZ or UST) initiated on the index date. The baseline period was the 6 months preceding the index date, and the follow-up period started on the index date and ended at the first event of either the end of data availability or the end of continuous health plan enrollment.
Patient eligibility criteria
Adult patients (aged 18–89 years) were included in the study if they had a diagnosis code for CD recorded on ≥2 distinct calendar dates, of which the first observed diagnosis for CD was required to have been on or prior to the index date. To identify patients with confirmed use of VDZ (VDZ cohort) or UST (UST cohort), patients were required to have completed the induction and initiated the maintenance phase of their index treatment (defined based on the recommended dosing scheduleCitation20,Citation21; Supplementary Methods). To evaluate patients using a similar timeframe in terms of CD management options in clinical practice, the index date for both cohorts must have been on or after September 26, 2016, the date UST received approval for the treatment of patients with CDCitation20,Citation21. Patients were also required to have had ≥6 months of continuous health plan coverage before and after the index date. Because each biologic therapy has multiple indicationsCitation20,Citation21, the most recent relevant immune-mediated disease observed on or before the index date was required to have been for CD, excluding patients with a diagnosis of ulcerative colitis (UC) from the VDZ cohort and those with a diagnosis of psoriasis or psoriatic arthritis from the UST cohort, if these were observed as the most recent immune-mediated condition before the index date.
Study subgroup
For each of the study cohorts, the subgroup of biologic-naïve (bio-naïve) patients was identified and defined as those with no evidence of exposure to any biologic therapy for CD prior to the index date (using available data from 2009 to have a more comprehensive assessment of pre-index medications). Bio-naïve patients were of interest given that the index treatment (VDZ or UST) was used as a first-line biologic therapy for CD in both subgroups.
Endpoints
Treatment persistence was defined as the time from the index date to treatment discontinuation (event) or end of follow-up (censor). Treatment discontinuation was defined as (1) a treatment gap of ≥146 d (i.e. 90 d following the recommended 56-day dosing intervalCitation20,Citation21) or (2) a switch to or add-on of another biologic or biosimilar therapy indicated for CD. Date of discontinuation was defined as the earliest of (1) 56 d after the last dose of index treatment (as this was considered to be the last day on treatment), or (2) the day before receiving the new biologic or biosimilar therapy. A sensitivity analysis was performed to measure treatment persistence beginning from maintenance phase initiation.
Because dosing information was not completely available in the data, dose escalation was measured based on intensification of dosing frequency, defined as either two consecutive shortened intervals between doses of the index treatment (i.e. a dosing interval of ≤49 d, i.e. ≥7 d shorter than the recommended 56-day interval) or one shortened dosing interval followed by index treatment discontinuation. For the UST cohort, an intravenous (IV) dose of UST during the maintenance phase also indicated a dose intensificationCitation21. By design, dose escalation analysis was conducted among the subgroups of patients with ≥3 doses of the index treatment during the maintenance phase. Time from maintenance phase initiation to first observed dose intensification (event), treatment discontinuation (as defined above; censor), or end of the follow-up period (censor) was assessed.
Opportunistic infection–related encounters were defined as medical visits (OP or IP) with an International Statistical Classification of Diseases and Related Health Problems, 10th edition (ICD-10) diagnosis code for viral, mycobacterial, bacterial, fungal, and parasitic infections (Supplementary Table 1). Time from the index date to the first opportunistic infection–related encounters (events) or end of the follow-up period (censor) was assessed.
Serious infection–related encounters were defined as IP visits with ICD-10 diagnosis codes for pulmonary, gastrointestinal, skin and subcutaneous tissue, urinary tract, ear, nose and throat, musculoskeletal, and other infections (Supplementary Table 2). Time from index date to the first IP encounter with a diagnosis for a serious infection (events) or end of the follow-up period (censor) was assessed.
Healthcare costs were assessed from a payer’s perspective during the time on index treatment including the induction and maintenance phases (i.e. from the index date until treatment discontinuation or end of follow-up). Costs were adjusted to 2018 US dollars using the US Medical Care Consumer Price Index and reported per patient per month (PPPM). All-cause total healthcare costs included medical and pharmacy costs. Costs were stratified into (1) index treatment costs (i.e. pharmacy and medical costs related to the index treatment) and (2) costs excluding index treatment costs, which included all other medical costs (hospitalization, emergency department [ED], and OP) and all other pharmacy costs. In addition, CD-related medical costs, defined as costs of claims with diagnosis codes for CD or procedure codes for CD-related surgeries or for IV corticosteroids, were reported.
All endpoints were assessed in the VDZ and UST cohorts overall and in the subgroups of VDZ and UST bio-naïve patients.
Statistical analyses
Patient characteristics were balanced between the VDZ and UST cohorts in the overall sample and in the bio-naïve subgroups separately using entropy balancingCitation22. Specifically, patients in the UST cohort were reweighted so that the overall distribution of patient characteristics was similar to that in the VDZ cohort. Continuous variables were summarized using means, SDs, and medians, while categorical variables were summarized using frequency counts and percentages. In addition, standardized differences between the VDZ and UST cohorts were calculated before and after weighting, with standardized differences of absolute value >0.2 considered to be suggestive of an imbalanceCitation23.
Rates of treatment persistence, time-to-dose escalation, time-to-opportunistic infection–related encounters, and time-to-serious infection–related encounters were estimated using weighted Kaplan-Meier (WKM) analyses and compared between both cohorts using log-rank tests. Healthcare costs were compared between cohorts using weighted mean cost differences (MCDs) obtained from weighted 2-part models. P values and 95% CIs were estimated using a nonparametric bootstrap.
Results
In total, 4300 and 2403 patients had ≥1 claim for VDZ and UST, respectively, and had been diagnosed with CD as the latest relevant immune-mediated disease before initiating treatment (). Of these, 589 and 599 patients met all the selection criteria and were classified into the VDZ and UST cohorts, respectively, with 172 (29.2%) patients from the VDZ cohort and 117 (19.5%) of the UST cohort being bio-naïve.
Figure 1. Patient disposition. aThe first claim for VDZ or UST was defined as the index date. The first treatment initiated (VDZ or UST) was defined as the index treatment. bFor VDZ, no UC diagnosis and for UST, no diagnosis for psoriatic arthritis or psoriasis, between the latest CD diagnosis pre-index and the index date, and within the 30 days before the index date, inclusively. Abbreviations: CD, Crohn’s disease; UC, ulcerative colitis; UST, ustekinumab; VDZ, vedolizumab.
Patient characteristics
After weighting, baseline demographic and clinical characteristics (including prior therapy use) were balanced between the VDZ and UST cohorts overall () and between the subgroups of VDZ and UST bio-naïve patients (data not shown). Overall, the weighted VDZ and UST cohort patients had a mean age of 43.9 years (SD = 14.3), and 57.2% were female. The mean time from the first observed CD diagnosis to initiation of index treatment was 46.0 months (SD = 35.1); 65.9% of patients had received corticosteroids during the 6-month pre-index period. In the bio-naïve subgroups, the mean age of the weighted cohort patients was 44.6 years, and 58.1% were female; the mean time from first observed CD diagnosis to initiation of index treatment was 32.4 months (SD = 33.4); 57.7% of patients treated with VDZ and 59.9% of patients treated with UST had received corticosteroids during the 6-month pre-index period (Supplementary Table 3).
Table 1. Patient characteristics.
In the VDZ cohort, the mean duration of follow-up was 14.3 months (SD = 5.3; median = 13.6) overall and 13.4 months (SD = 5.0; median = 12.25) in the bio-naïve subgroup. In the UST cohort, the mean duration of follow-up was 13.2 months (SD = 4.6; median = 12.3) overall and 12.6 months (SD = 4.3; median = 12.1) in the bio-naïve subgroup.
Treatment persistence
Treatment persistence rates assessed from the index date were similar between the VDZ and UST cohorts (6-month WKM rates: 91.5% vs 91.6%; p = .97; 12-month WKM rates: 76.5% vs 82.1%; p = .50; ). Similarly, in the bio-naïve subgroups, treatment persistence rates were not statistically significantly different in patients in the VDZ and UST cohorts when assessed from the index date (6-month WKM rates: 92.4% vs 92.0%; p = .86; 12-month WKM rates: 78.6% vs 87.0%; p = .44; Supplementary Figure 1).
Figure 2. Post-index treatment persistence in VDZ and UST cohorts. Treatment persistence was defined as time from index date to treatment discontinuation date (event) or end of follow-up (censor). Treatment discontinuation was defined as a treatment gap of ≥146 d (90 d following the recommended 56-day dosing interval), or a switch to or add-on of another biologic or biosimilar therapy indicated for CD. Date of discontinuation was set at 56 d after the patient received the last dose of index treatment, or the date before receiving the switch/add-on. Abbreviations: UST, ustekinumab; VDZ, vedolizumab; WKM, weighted Kaplan–Meier.
In a sensitivity analysis assessing treatment persistence from the start of the maintenance phase, treatment persistence rates were clinically similar for VDZ and UST overall (Supplementary Figure 2) and in the bio-naïve subgroups (Supplementary Figure 3).
Dose escalation
Rates of dose escalation post-initiation of the maintenance phase were similar in the VDZ and UST cohorts (6-month WKM rates: 18.3% vs 15.3%; p = .27; 12-month WKM rates: 29.3% vs 32.7%; p = .97; ). Consistently, rates of dose escalation were not statistically significantly different between the bio-naïve subgroups of patients treated with VDZ versus UST (6-month WKM rates: 13.0% vs 9.6%; p = .66; 12-month WKM rates: 21.5% vs 32.3%; p = .43; Supplementary Figure 4).
Figure 3. Dose escalation in the VDZ and UST cohorts. Dose escalation was measured based on intensification of dosing frequency defined as two consecutive intervals between dosing of ≤49 d (i.e. at least 7 d shorter than the recommended 56-day interval); one dosing interval of ≤49 d followed by treatment discontinuation; or IV administration of UST during the maintenance phase. Time from maintenance phase initiation to first observed dose intensification (event), treatment discontinuation (censor), or end of follow-up period (censor) was assessed. Abbreviations: IV, intravenous; UST, ustekinumab; VDZ, vedolizumab; WKM, weighted Kaplan–Meier.
Opportunistic and serious infection–related encounters
Rates of opportunistic infection–related encounters post-index were similar in the VDZ and UST cohorts (6-month WKM rates: 5.8% vs 5.0%; p = .56; 12-month WKM rates: 8.7% vs 7.5%; p = .57; ). Likewise, rates of serious infection–related encounters post-index were not statistically significantly different in the VDZ and UST cohorts (6-month WKM rates: 3.1% vs 3.2%, p = .91; 12-month WKM rates: 7.7% vs 6.3%; p = .56; ).
Figure 4. Time to (a) opportunistic and (b) serious infection–related encounters post index date in VDZ and UST cohorts. Time to opportunistic infection–related encounters was measured as the time from index date to the first medical claim with a diagnosis code for an opportunistic infection (event) or end of the patient’s continuous health plan enrollment (censored). Time to serious infection–related encounters was measured as the time from index date to the first IP medical claim with a diagnosis code for a serious infection (event) or end of the patient’s continuous health plan enrollment (censored). Abbreviations: IP, inpatient; UST, ustekinumab; VDZ, vedolizumab; WKM, weighted Kaplan-Meier.
Similar results for both opportunistic and serious infection–related encounters were obtained in the bio-naïve subgroups (Supplementary Figure 5). Specifically, the 6- and 12-month WKM rates of opportunistic infection–related encounters post-index among bio-naïve patients treated with VDZ vs UST were 3.5% vs 3.3% (p = .95) and 6.3% versus 4.2% (p = .65), respectively; for serious infection–related encounters, the rates were 3.5% vs 2.8% (p = .79) and 7.1% vs 5.7% (p = .72), respectively.
Healthcare costs
Over a mean time on index treatment of 12.1 months (SD = 5.3) in the VDZ cohort and 11.5 (SD = 4.7) months in the UST cohort, the total mean healthcare costs were statistically significantly lower in the VDZ versus UST cohort ($7460 vs $12,510 PPPM; weighted MCD = –$5051 [95% CI, –$5783, –$4365] PPPM; p < .01; ; Supplementary Table 4), driven by the lower index treatment costs ($5356 vs $10,302 PPPM, respectively; weighted MCDs = –$4946 [95% CI, –$5406, –$4532]; p <.01). Excluding index treatment costs, all-cause medical costs (weighted MCDs = –$123 PPPM; p = .69) and pharmacy costs (weighted MCDs = $19 PPPM; p = .64) were similar in both cohorts (), as were CD-related medical costs (weighted MCDs = –$103 PPPM; p = .70; Supplementary Table 4).
Figure 5. Mean healthcare costs PPPM in VDZ and UST cohorts. Note: Full details, including 95% CI, provided in Supplementary Table 3. *P value for mean cost difference < 0.01. Abbreviations: PPPM, per patient per month; UST, ustekinumab; VDZ, vedolizumab.
For the bio-naïve subgroups, the mean time on index treatment was 11.6 months (SD = 4.9) among those treated with VDZ and 11.2 (SD = 4.7) months for those treated with UST. Similar to the overall cohorts, the total mean healthcare costs were statistically significantly lower in VDZ than UST bio-naïve patients ($7335 vs $11,801 PPPM, respectively; weighted MCDs = –$4466 [95% CI, –$6145, –$2661] PPPM; p < .01; Supplementary Table 5), driven by the significantly lower index treatment costs ($5238 vs $10,129 PPPM, respectively; weighted MCDs = –$4891 [95% CI, –$6154, –$3526]; p < .01).
Discussion
This US real-world study found that patients with CD initiating VDZ or UST had comparable treatment persistence, need for dose escalation, and opportunistic and serious infections suggestive of adverse events. However, healthcare costs were significantly lower with VDZ than with UST, predominantly driven by the lower index treatment costs of VDZ that were almost half those of UST.
Treatment persistence associated with VDZ and UST among patients with CD has previously been reported from other US administrative databasesCitation24,Citation25. In the Demuth et al. study conducted in patients treated with VDZ, the rate of treatment persistence (defined as no treatment gap >90 d) was reported to be 75%–80% at 6 monthsCitation24, lower than the 91.5% observed in the current VDZ cohort, which may be due to the requirement of maintenance initiation in the current study. Meanwhile, the Obando et al. study that evaluated treatment persistence with UST (defined as no treatment gap >90 d) reported a 12-month persistence rate for UST (83.6%) that was comparable to the rate found in this study (82.1%)Citation25.
In a recent large claims-based study by Teeple et al. comparing the persistence of several biologics among patients with CD pooled from three large claims databasesCitation26, the proportion of patients persistent on treatment at 12 months post-initiation was reported to be numerically higher with UST than with VDZ. However, no adjustments for differences in patient characteristics were performed in that study. As biases and unmeasured confounders are inherent in real-world studies (in which treatment choice can be dictated by insurance coverage, provider, or patient preferences), adjustment techniques to balance patient characteristics at baseline are crucial to reduce bias between cohorts.
Regarding other clinical outcomes assessed, the previously noted Obando study found that 17.9% of patients experienced dose escalation at 6 months, which is similar to the 15.3% found in the current UST cohortCitation25. Meanwhile, previous real-world studies assessing rates of infection associated with VDZ and UST have largely been conducted among bio-experienced patients, and the results have been inconsistentCitation13,Citation14,Citation27. For instance, a prospective registry study in the Netherlands found similar odds of infection (p = .758) between matched cohorts treated with VDZ versus USTCitation14 among patients with CD in whom anti-TNF therapy had failed, whereas a multicenter retrospective French study described numerically higher rates of infections with VDZ (28.7%) than with UST (11.2%)Citation13. The heterogeneity of treatment history among bio-experienced patients may partially explain the discrepancies observed.
The current study applied entropy-balancing techniques to minimize differences in patient characteristics between cohorts. Nonetheless, it is notable that in the absence of stringently controlled head-to-head clinical trials, bio-naïve patients could represent a more homogeneous and clinically meaningful population for real-world studies. Specifically, the effectiveness of second-line treatment for CD may depend on the reasons for discontinuing prior anti-TNF therapy, which may include nonresponse, loss of response, or intolerability; and these factors exert an impact on clinical decision-making during second-line treatmentCitation28. Therefore, comparing treatment effectiveness, including treatment persistence, among bio-naïve patients could be more clinically representative. In this regard, the current study found that among bio-naïve patients, VDZ and UST are associated with comparable outcomes for all indicators of treatment effectiveness measured. Of note, analyses were also performed for bio-experienced patients, and the results were very similar to those in the overall population (data not shown).
In the current study, total healthcare costs were significantly higher in patients treated with UST compared with VDZ, predominantly attributable to the substantially higher UST pharmacy costs in the US. Otherwise, costs associated with hospitalization, ED visits, and OP visits were not significantly different for VDZ and UST. These findings suggest that the higher cost of treatment with UST does not translate to substantially different outcomes compared with outcomes in patients treated with VDZ. It is also worth noting that given that the difference in healthcare costs was mainly due to the difference in treatment costs, patients with a higher risk of dose escalation may incur even higher costs owing to the additional doses required.
Limitations
The findings of the current study should be considered in light of some limitations inherent in claims studies, including limited information regarding disease extent, behavior (inflammatory, stricturing, penetrating), severity, as well as clinical or objective response to therapy. Pharmacy claims for a filled prescription do not guarantee treatment compliance, particularly with injectable drugs such as UST. The time period of this study (2016–2018) could represent a potential limitation for this study.
Conclusions
We have shown that VDZ and UST offer similar real-world treatment persistence, dose escalation, and rates of opportunistic or serious infections in patients with CD. However, lower treatment costs with VDZ compared with UST may provide substantial savings for patients and the healthcare system. Treatment decisions should take into account multiple factors, but when several treatment options are available, choosing the most cost-effective therapy can help address rising healthcare costs in the US. Future comparative cost-effectiveness studies with these biologics in the treatment of CD are warranted.
Transparency
Declaration of funding
This study was supported by Takeda Pharmaceuticals U.S.A., Inc. The study sponsor was involved in several aspects of the research, including the study design, the interpretation of data, the writing of the manuscript, and the decision to submit the manuscript for publication.
Declaration of financial/other relationships
M. Chiorean has received consulting and/or speaking fees from AbbVie, Arena Pharmaceuticals, Celgene, Janssen, Medtronic, Pfizer, Shire, Takeda, Lilly, BMS, and UCB, and has received educational/research grants from AbbVie, Takeda, Novartis, and Pfizer. J. Jiang, N. Candela, G. Chen, and T. Fan are employees of Takeda Pharmaceuticals U.S.A., Inc., and have stock or stock options. H. Romdhani, D. Latremouille-Viau, S. Shi, R. Bungay, and A. Guerin are employees of Analysis Group, Inc., a consulting company that received funding from Takeda Pharmaceuticals U.S.A., Inc.
Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.
Author contributions
All authors were involved in the conception, design, data acquisition, analyses, and interpretation of the study results; contributed to the drafting of the manuscript; and revised it critically for intellectual content. All authors approved the final version to be published and agree to be accountable for all aspects of the work.
CMRO-2023-ST-0473 Supplement clean_30Nov23.pdf
Download PDF (567.6 KB)Acknowledgements
Medical writing assistance was provided by Blair Hesp, PhD, CMPP, and Edwin A. Obana, PhD, on behalf of Cadent, a Syneos Health group company, as well as Flora Chik, PhD, of Analysis Group, Inc., and supported by Takeda Pharmaceuticals U.S.A., Inc. Certain data used in this study were supplied by International Business Machines (IBM) Corporation as part of one or more IBM MarketScan Research Databases. Any analysis, interpretation, or conclusion based on these data is solely that of the authors and not IBM Corporation.
Data availability statement
Due to the nature of the research in this study, the supporting data are not available.
References
- The global, regional, and national burden of inflammatory bowel disease in 195 countries and territories, 1990-2017: a systematic analysis for the global burden of disease study 2017. Lancet Gastroenterol Hepatol. 2020;5:17–30.
- Crohn’s & Colitis Foundation of America [Internet]. The facts about inflammatory bowel diseases. 2014 [cited Sep 29, 2022]. https://www.crohnscolitisfoundation.org/sites/default/files/legacy/assets/pdfs/updatedibdfactbook.pdf.
- Loftus EV Jr. Update on the incidence and prevalence of inflammatory bowel disease in the United States. Gastroenterol Hepatol. 2016;12(11):704–707.
- Shivashankar R, Tremaine WJ, Harmsen WS, et al. Incidence and prevalence of Crohn’s disease and ulcerative colitis in Olmsted county, Minnesota from 1970 through 2010. Clin Gastroenterol Hepatol. 2017;15(6):857–863. doi: 10.1016/j.cgh.2016.10.039.
- Ye Y, Manne S, Treem WR, et al. Prevalence of inflammatory bowel disease in pediatric and adult populations: recent estimates from large national databases in the United States, 2007-2016. Inflamm Bowel Dis. 2020;26:619–625.
- Gajendran M, Loganathan P, Catinella AP, et al. A comprehensive review and update on Crohn’s disease. Dis Mon. 2018;64(2):20–57. doi: 10.1016/j.disamonth.2017.07.001.
- Manceur AM, Ding Z, Muser E, et al. Burden of Crohn’s disease in the United States: long-term healthcare and work-loss related costs. J Med Econ. 2020;23(10):1092–1101. doi: 10.1080/13696998.2020.1789649.
- Lichtenstein GR, Shahabi A, Seabury SA, et al. Lifetime economic burden of Crohn’s disease and ulcerative colitis by age at diagnosis. Clin Gastroenterol Hepatol. 2020;18(4):889–897.e10. doi: 10.1016/j.cgh.2019.07.022.
- Lichtenstein GR, Loftus EV, Isaacs KL, et al. ACG clinical guideline: management of Crohn’s disease in adults. Am J Gastroenterol. 2018;113(4):481–517. doi: 10.1038/ajg.2018.27.
- Skyrizi. Prescribing information. AbbVie Inc; 2022. https://www.rxabbvie.com/pdf/skyrizi_pi.pdf.
- Torres J, Bonovas S, Doherty G, et al. ECCO guidelines on therapeutics in Crohn’s disease: medical treatment. J Crohns Colitis. 2020;14(1):4–22. doi: 10.1093/ecco-jcc/jjz180.
- Kawalec P, Moćko P. An indirect comparison of ustekinumab and vedolizumab in the therapy of TNF-failure Crohn’s disease patients. J Comp Eff Res. 2018;7(2):101–111. doi: 10.2217/cer-2017-0041.
- Alric H, Amiot A, Kirchgesner J, et al. The effectiveness of either ustekinumab or vedolizumab in 239 patients with Crohn’s disease refractory to anti-tumour necrosis factor. Aliment Pharmacol Ther. 2020;51(10):948–957. doi: 10.1111/apt.15706.
- Biemans VBC, van der Woude CJ, Dijkstra G, et al. Ustekinumab is associated with superior effectiveness outcomes compared to vedolizumab in Crohn’s disease patients with prior failure to anti-TNF treatment. Aliment Pharmacol Ther. 2020;52(1):123–134. doi: 10.1111/apt.15745.
- Manlay L, Boschetti G, Pereira B, et al. Comparison of short- and long-term effectiveness between ustekinumab and vedolizumab in patients with Crohn’s disease refractory to anti-tumour necrosis factor therapy. Aliment Pharmacol Ther. 2021;53(12):1289–1299. doi: 10.1111/apt.16377.
- Billioud V, Sandborn WJ, Peyrin-Biroulet L. Loss of response and need for adalimumab dose intensification in Crohn’s disease: a systematic review. Am J Gastroenterol. 2011;106(4):674–684. doi: 10.1038/ajg.2011.60.
- Gisbert JP, Panés J. Loss of response and requirement of infliximab dose intensification in Crohn’s disease: a review. Am J Gastroenterol. 2009;104(3):760–767. doi: 10.1038/ajg.2008.88.
- Patel H, Lissoos T, Rubin DT. Indicators of suboptimal biologic therapy over time in patients with ulcerative colitis and Crohn’s disease in the United States. PLOS One. 2017;12(4):e0175099. doi: 10.1371/journal.pone.0175099.
- Park KT, Ehrlich OG, Allen JI, et al. The cost of inflammatory bowel disease: an initiative from the Crohn’s & colitis foundation. Inflamm Bowel Dis. 2020;26(1):1–10. doi: 10.1093/ibd/izz104.
- Entyvio. Prescribing information. Takeda Pharmaceuticals U.S.A., Inc; 2022. https://www.general.takedapharm.com/ENTYVIOPI.
- Stelara. Prescribing information. Janssen Biotech Inc; 2019. https://www.janssenlabels.com/package-insert/product-monograph/prescribing-information/STELARA-pi.pdf.
- Hainmueller J. Entropy balancing for causal effects: a multivariate reweighting method to produce balanced samples in observational studies. Polit Anal. 2012;20(1):25–46. doi: 10.1093/pan/mpr025.
- Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale, NJ: Lawrence Erlbaum Associates; 1988.
- Demuth D, Patel H, Chastek B, et al. Real-world treatment persistence with vedolizumab in Crohn’s disease and ulcerative colitis patients in the United States (US): a retrospective analysis using the Optum research database: 721. Am J Gastroenterol. 2016;111: s 324. doi: 10.14309/00000434-201610001-00721.
- Obando C, Ding Z, Muser E, et al. Persistence, dose titration, and health care resource utilization among Crohn’s disease patients treated with ustekinumab: a real-world analysis in the United States. Adv Ther. 2020;37(5):2127–2143. doi: 10.1007/s12325-020-01276-3.
- Teeple A, Sah J, Mallampati R, et al. Persistence, dosing, and other treatment patterns among Crohn’s disease patients initiating biologics in United States. Crohn’s Colitis. 2021;3:otab076.
- Dulai PS, Singh S, Jiang X, et al. The real-world effectiveness and safety of vedolizumab for moderate-severe Crohn’s disease: results from the US VICTORY consortium. Am J Gastroenterol. 2016;111(8):1147–1155. doi: 10.1038/ajg.2016.236.
- Singh S. Network meta-analysis to inform positioning of biologics in patients with Crohn’s disease: promise and perils. Best Pract Res Clin Gastroenterol. 2019;38-39:101614. doi: 10.1016/j.bpg.2019.05.001.