Resource utilization before and during infliximab therapy in patients with inflammatory bowel disease

Abstract

Objective:

Although Remicade (infliximab) is costly relative to non-biologic therapy, its impact on healthcare resource utilization and mucosal healing may make it a cost-effective option. This study aimed to compare gastrointestinal (GI)-related healthcare resource utilization and severity of mucosal damage before and during infliximab therapy in Crohn’s disease (CD) or ulcerative colitis (UC) patients.

Methods:

A retrospective chart review was conducted at 14 gastroenterology practices from across the country, which varied in practice sizes and types. Patients were aged ≥18 years, diagnosed with CD or UC, and had an infliximab index date between January 1, 2005 and September 30, 2007. GI-related utilization 12 months before and 12 months after the index date was compared. Endoscopic disease severity was categorized based on blinded review of abstracted reports.

Results:

Results from 268 patients indicated significantly lower rates of surgery (29.7% to 9.9%, p < 0.0001, CD; 24.4% to 12.8%, p = 0.042, UC) and colonoscopy (54.4% to 17.6%, p < 0.0001, CD; 50.0% to 22.1%, p = 0.0007, UC) during infliximab therapy. The rates of hospitalizations in UC (15.1% to 3.5%, p = 0.0124) and radiology assessments in CD (23.1% to 10.4%, p = 0.006) also decreased. Based on severity data from 183 procedures, greater proportions of patients had normal or mild ratings during infliximab treatment compared with pre-treatment.

Limitations:

This retrospective descriptive study is limited by the type and quantity of information available in patient charts from 14 gastroenterology clinics during the first year of infliximab treatment. In addition, the number of patients with pre-treatment and post-treatment disease severity information was too small to make comparisons among disease severity groups. Further information about the severity of disease and the extent of mucosal healing could be helpful in determining the effect of therapy on resource utilization in future research.

Conclusions:

GI-related resource utilization was significantly lower and attenuation of mucosal damage severity was observed during infliximab treatment compared with the pre-treatment period.

Key words::

• Infliximab
• Inflammatory bowel disease
• Resource utilization

Introduction

Inflammatory bowel disease (IBD) affects an estimated 1 million people in the US, with 30,000 new cases diagnosed every year1. The chronic conditions of IBD may develop at any age, although the peak age of onset is 15–30 years, with ∼10% of cases occurring in individuals aged <18 years1.

The cause of IBD is idiopathic; however, it likely involves an immune reaction of the body to its own intestinal tract. It includes 2 conditions: ulcerative colitis (UC), a mucosal inflammatory condition of the rectum and colon, and Crohn’s disease (CD), a chronic relapsing and remitting disease associated with inflammation and damage to any portion of the gastrointestinal (GI) tract and may also include extraintestinal manifestations2. Inflammatory bowel disease has been associated with increased mortality3, and untreated or inadequately treated IBD can exact a large healthcare burden due to GI-related procedures, emergency room (ER) visits, hospitalizations, and surgeries4. Patients with IBD have an increased risk of colorectal cancer5, and recent evidence suggests that patients with IBD may have an elevated risk of endothelial dysfunction and coronary artery disease6. Inflammatory bowel disease also negatively affects quality-of-life, productivity, and functionality7–11.

Therapy is frequently initiated with 2 major goals: (1) induction and maintenance of disease remission, and (2) a reduction in complications and surgery rates12,13. In order to achieve these goals (i.e., optimal long-term outcomes), targeting mucosal inflammation and associated tissue damage (achieving mucosal healing) may be as important as symptomatic improvement14,15.

Traditionally, therapy for IBD has taken a step-up approach, and flares have been managed with glucocorticosteroids. In the step-up approach to therapy, patients with milder disease are generally treated with aminosalicylates, and treatment progresses to immunomodulators, including 6-mercaptopurine, azathioprine, and methotrexate, as disease progresses to moderate severity. Anti-tumor necrosis factor (TNF) agents such as Remicade (infliximab)16, Humira (adalimumab)17, and Cimzia (certolizumab pegol)18 have become important components of treatment for moderate-to-severe IBD. (Remicade is a registered trademark of Janssen Biotech, Inc., Horsham, PA; Humira is a registered trademark of Abbott Laboratories, North Chicago, IL; and Cimzia is a registered trademark of UCB, Inc., Smyrna, GA.) Anti-TNF agents are redefining what is considered an acceptable symptomatic response to therapy and hold promise for modifying the natural history of the disease. Infliximab, the first anti-TNF agent approved for IBD in the US, is effective for the treatment of luminal and fistulizing CD, pediatric CD, and UC19–25. Adalimumab17 and certolizumab pegol18 are approved for treatment of luminal CD but not for UC.

Anti-TNF therapy apparently serves as a pivotal mediator in the cascade of mucosal inflammation. These agents are believed to reduce intestinal inflammation in patients with IBD by binding to and neutralizing TNF-alpha on the cell membrane and in the blood and by destroying TNF-producing cells26. These actions explain how anti-TNF agents are particularly effective inhibitors of TNF and why they have rapid and substantial clinical benefit.

Inflammatory bowel disease is an expensive disease, particularly because affected patients are usually in their most productive years at the onset of disease, and life expectancy is nearly normal. Longobardi et al.27 noted that within the first 5 years of diagnosis, patients with IBD had a higher likelihood than the general population to visit an ER, to be hospitalized, and to undergo surgery. In a later study28, they determined that the most costly services were employed within the first 2 years after diagnosis. According to the 2007 Healthcare Cost and Utilization Project (HCUP) statistics, the mean charge for an IBD-related hospitalization was $968729. Another study examined the direct costs of CD and UC and estimated that the mean amount paid by health plans for an inpatient hospitalization was $2593 for CD and $1906 for UC30. The authors concluded that the distribution of resource utilization may be shifting as increasing drug spending may result in decreased utilization of resources such as hospitalizations and surgeries30. Therefore, although infliximab is expensive relative to non-biologic therapy, its effectiveness in inducing and sustaining remission in patients with CD and UC and its association with mucosal healing may make it a cost-effective choice.

The objective of this retrospective chart review was to investigate the effect of infliximab therapy on IBD in terms of healthcare resource utilization and to compare the severity of mucosal damage before and during the initial year of infliximab therapy.

Materials and methods

A retrospective medical record review was conducted at 14 community gastroenterology practices in the US. The 14 sites were selected to provide national geographic representation and a variety of practice sizes and types. All available charts for IBD patients started on infliximab within the study period were identified for review; a maximum of 60 patients per clinic, who met all inclusion criteria, were included in the study.

Data collection

Data were collected through a standardized chart review form using a graphical user interface developed in Microsoft Access (Microsoft Corporation, Redmond, WA). Charts of all patients were de-identified at the study sites using a unique alpha-numeric value. A participant log sheet was developed and stored at the study site to permit the ability to match the de-identified patient marker (unique for all study participants) to the actual patient chart, allowing for future data validation if the need arose. Collected data were downloaded into a secure data warehouse, and paper records were shredded. Only investigators involved in the study had access to study data.

Patient selection

Patients who started infliximab treatment between January 1, 2005 and September 30, 2007, had a diagnosis of CD or UC, and were ≥18 years old were included in the study. Patients were excluded if they had <12 months of infliximab use, had used biologics in the 12 months before the infliximab index date, or were enrolled in an IBD-related clinical trial.

Variables and analysis

Data collected from the charts included baseline and demographic characteristics (IBD diagnosis information, age, sex, weight, IBD-related complications), medication utilization, GI-related resource utilization, and disease severity.

Medication utilization data included infliximab dosing patterns (dose, frequency, changes) and non-anti-TNF therapy use (aminosalicylates, antibiotics, corticosteroids, immunomodulators, probiotics/total parenteral nutrition).

The healthcare resource utilization assessment included GI-related hospitalizations, surgical interventions, ER visits, and radiology assessments as well as GI-related endoscopic diagnostic procedures (colonoscopy, esophagogastroduodenoscopy [EGD], sigmoidoscopy). Resource utilization rates (i.e., the proportions of patients utilizing the service) were compared for the 12 months before and 12 months after infliximab initiation using McNemar’s tests for correlated proportions. The level of significance (alpha) was set a priori to 0.05. All statistical analyses were conducted with SAS software, version 9.1 (SAS Corp., Cary, NC).

Cost reductions were estimated for hospitalizations and endoscopic diagnostic procedures (colonoscopy, EGD, sigmoidoscopy) for the 12 months following infliximab initiation. Mean costs for inpatient hospitalizations were estimated using data from HCUP29. Kappelman et al.30 extracted medical and pharmacy claims from an administrative database and estimated IBD-attributable claims paid by health plans for both CD and UC, which were used to quantify cost savings associated with surgery, ER visits, and radiology visits. Mean costs for endoscopic diagnostic procedures were calculated from the 2010 Medicare Physician Fee Schedule31. The national fees for colonoscopy (Healthcare Common Procedure Coding System [HCPCS] Codes: G0105 and 45380), EGD (HCPCS Code: 49440), and sigmoidoscopy (HCPCS Code: G0104) were identified, and an average of costs from all the records were obtained. The mean non-facility price and mean facility price were then summed to compute the average cost for each procedure. Cost of infliximab was not included in these analyses.

The endoscopic procedure notes were reviewed by an experienced gastroenterologist in blinded fashion. Mucosal damage was assessed as accurately as possible on the basis of the procedure notes. For the majority of procedures, a description of the mucosal damage could be determined. Severity of disease was categorized as normal, mild, mild-to-moderate, moderate, moderate-to-severe, or severe, based on the information contained in the abstracted reports. When the notes did not reflect a definitive description of the mucosa, it was assessed as ‘unknown’.

The study protocol was approved by the New England Institutional Review Board.

Results

Demographics

Baseline characteristics of the study population are presented in Table 1. Charts from 182 patients were included in the CD cohort and from 86 patients in the UC cohort. Mean age in the CD group was ∼6 years younger than in the UC group (42.2 vs 47.8 years). Mean patient weight was ∼80 kg, and nearly half the patients were female in both groups. Mean time to infliximab treatment after diagnosis was 7.6 years in the CD group and 9.9 years in the UC group. These characteristics are consistent with the general IBD population in the US1.

Table 1.  Baseline patient characteristics.

	Crohn’s disease (n = 182)	Ulcerative colitis (n = 86)
Mean (SD) age, years	42.2 (15.2)	47.8 (14.8)
Female, %	47.3	47.7
Weight, kg	81.0 (20)	80.2 (19.1)
Time from diagnosis to infliximab treatment, years	7.6 (8.4)	9.9 (9.4)

Data presented as mean (SD) unless otherwise noted.

Infliximab dosing

More than 90% of patients in both groups were initiated at a dose of 5 mg/kg (data not shown), as recommended by the US Food and Drug Administration16. In the 2 groups combined, nearly 95% continued on infliximab therapy throughout the study, indicating that infliximab was well tolerated. Furthermore, nearly three-fourths of patients maintained their starting dose of 5 mg/kg throughout the study, and dose frequency remained stable in over half the patients, with the majority of patients being dosed every 8 weeks.

Resource utilization

Hospitalizations, surgical interventions, emergency room visits, and radiology assessments (Figure 1)

According to the charts, the proportion of patients requiring hospitalization decreased during infliximab therapy when compared with pre-therapy. In the CD cohort, 19 (10.4%) patients required hospitalization prior to infliximab initiation, while 16 (8.8%) required hospitalization following infliximab initiation, but the difference was not statistically significant (NS). In the UC cohort, 10 fewer patients required hospitalization following infliximab initiation (13 vs 3), resulting in a statistically significant decrease (15.1% vs 3.5%; p = 0.0124). When dollarized using the average amount paid for an IBD-related hospitalization according to the HCUP statistics ($9687)29, the reduction in hospitalization in CD results in an estimated total savings of $29,061, while the reduction in UC-related hospitalization amounts to $96,870.

Figure 1.  Healthcare resource utilization frequency before versus during infliximab therapy. NS, not statistically significant; ER, emergency room.

Compared with the pre-therapy period, significantly smaller percentages of patients underwent surgical interventions during infliximab therapy in both the CD group (29.7% [n = 54] vs 9.9% [n = 18]; p < 0.0001) and the UC group (24.4% [n = 21] vs 12.8% [n = 11]; p = 0.042). Kappelman et al.30 found that the mean cost for a surgical inpatient hospitalization was $1026 for CD and $807 for UC. Therefore, the estimated cost savings due to avoided surgeries in this study would be $36,936 for the CD cohort and $8070 for the UC cohort.

Based on information available in the charts, few patients had an ER visit either before or after initiation of infliximab therapy (CD, 1.1% vs 2.2%, p = NS; UC, 5.8% vs 1.2%, p = NS), although this aspect of resource utilization may not have been well documented in the charts. The mean cost for an ER visit was $97 for CD and $44 for UC26. Therefore, the 2 additional ER visits following infliximab initiation for CD would cost an additional $194, while the 4 fewer ER visits following infliximab initiation in the UC cohort would amount to an estimated savings of $176.

A significantly lower proportion of patients in the CD group had radiology assessments after initiation of infliximab therapy (23.1% [n = 42] vs 10.4% [n = 19]; p = 0.006). At a cost of $272 per radiology visit30, this would represent an estimated cost savings of $6256 following initiation of infliximab therapy. In the UC group, 9.3% (n = 8) of patients had an assessment before infliximab therapy compared with 3.5% (n = 3) of patients during therapy (p = NS). At a cost of $165 per radiology visit30, this would represent an estimated cost savings of $825 for the UC cohort.

Endoscopic diagnostic procedures (Figure 2)

Colonoscopy (colonoscopy and/or virtual/capsule colonoscopy)

Colonoscopy was the most frequent GI procedure. The charts indicated that more than half (54.4%; n = 99) the patients in the CD group and 50.0% (n = 43) of the UC group underwent colonoscopy in the 12 months prior to infliximab initiation. Colonoscopy rates decreased significantly in both groups during infliximab therapy, to 17.6% (n = 32) of patients in the CD group (p < 0.0001) and 22.1% (n = 19) of patients in the UC group (p = 0.0007). According to the 2010 Medicare Physician Fee Schedule, the mean cost of a screening colonscopy is $380.88, and the mean cost of a colonoscopy with a biopsy is $704.9131. Therefore, the estimated cost savings following infliximab initiation associated with avoided colonoscopies amount to $25,518.96 to $47,228.97 for CD and $9141.12 to $16,917.84 for UC.

Figure 2.  Colonoscopy, esophagogastroduodenoscopy (EGD), and sigmoidoscopy frequency before versus during infliximab therapy. NS, not statistically significant.

Esophagogastroduodenoscopy

No significant changes in the proportion of patients undergoing EGD were observed in the CD or UC cohorts. In the CD group, 9.9% (n = 18) and 4.9% (n = 9) of the patients underwent EGD before vs during infliximab treatment, respectively (p = NS). In the UC group, the EGD rate increased slightly, from 4.7% (n = 4) to 7.0% (n = 6) of patients (p = NS), during infliximab therapy. The 2010 Medicare Physician Fee Schedule sets the mean cost of an EGD at $1268.13. Therefore, the reduction in EGD-related costs following infliximab initiation amounts to an estimated savings of $11,413.17 for the CD cohort, which is offset slightly by the increase of $2536.26 in the UC cohort.

Sigmoidoscopy

Few patients underwent sigmoidoscopy either before or during infliximab treatment. In the CD group, only 1 (0.5%) patient had a sigmoidoscopy before infliximab initiation and no patients had a sigmoidoscopy during infliximab therapy (p = NS). In the UC group, 4 (4.7%) patients required sigmoidoscopy before initiation of infliximab therapy, as did 4 (4.7%) patients during infliximab therapy (p = NS). The savings post-infliximab initiation amounted to $185.83 for the 1-patient difference in CD using the rate set for flexible sigmoidoscopy in the 2010 Medicare Physician Fee Schedule.

Severity of mucosal damage (Figure 3)

Charts revealed adequate information about mucosal damage severity in 120 cases in the CD group (90 pre, 30 post) and 63 cases in the UC group (49 pre, 14 post). The charts from patients with CD indicated that pre-infliximab severity was rated as normal in 9%, mild in 9%, moderate in 37%, moderate-to-severe in 1%, and severe in 26%. During infliximab therapy, the charts revealed that severity was rated normal in 40%, mild in 37%, and moderate in 23%; in no case was damage rated as severe or moderate-to-severe. Among the charts of patients with UC, before initiation of infliximab therapy, severity was labeled as normal in 8%, mild in 22%, moderate in 41%, and severe in 29%. After initiation of infliximab therapy, severity was rated as normal in 29%, mild in 29%, moderate in 36%, and severe in 7%.

Figure 3.  Mucosal severity before versus during infliximab therapy.

Discussion

This observational study showed substantially lower GI-related resource utilization during the first year of infliximab therapy compared with the year before initiation of therapy. Significantly fewer patients in both groups required surgical interventions during infliximab therapy. Likewise, significantly fewer UC patients required hospitalization for GI conditions, and significantly fewer CD patients had GI-related radiology assessments after initiation of therapy. Few patients in either group required ER visits either before or during infliximab therapy; however, these results may not reflect true utilization since ER visits may not have been well documented in the charts.

With regard to endoscopic procedures, the use reductions were most marked for colonoscopy, with the proportion of patients who needed colonoscopy reduced by two-thirds in the CD group and by more than half in the UC group. These changes were statistically significant in both groups. No significant differences were reported for the rates of EGD or sigmoidoscopy, which were <10% and <5%, respectively, before and during infliximab therapy.

When estimating the cost savings associated with reduced hospitalizations, surgeries, ER visits, radiology visits, and diagnostic procedures in the year following initiation of infliximab therapy, the savings would range between $109,177 and $130,887 for the 182 patients with CD and $112,546 to $120,323 for the 86 patients with UC whose charts were abstracted in this study. The majority of the cost savings was due to fewer hospitalizations, surgeries, and colonoscopies following infliximab initiation; however, these estimates are likely under-estimated. The Kappelman et al.30 reference cited the mean cost of services for all patients included in the analysis, rather than for only those who utilized the specific services; therefore, the mean costs included patients who had zero costs for some services. If only those patients who utilized services were included in Kappelman et al.’s analysis, the mean costs would likely be much higher. Despite this under-estimation, cost offsets should be taken into consideration when determining the cost-effectiveness of therapies to treat IBD. Of note, drug costs were not included in this study, as these costs vary widely. Including the cost of infliximab therapy may negate potential cost savings in the short-term, and longer-term data are needed to fully characterize the impact of infliximab therapy. Although this study assessed healthcare resource utilization and associated cost reduction in the year following infliximab initiation, it did not address the lifetime cost savings associated with infliximab use. Future studies may want to examine the impact of infliximab on healthcare resource utilization and costs.

In general, mucosal inflammation was less severe after infliximab treatment based on the limited available data and the subjective assessment of that data; the charts revealed that no patient with CD had severe disease after initiation of infliximab therapy, compared with one-quarter of the patients before infliximab therapy. Likewise, in patients with UC, 29% of patients had severe disease before infliximab therapy compared with 7% after initiation of therapy. Although the numbers were not adequate to determine statistical significance, they do show encouraging trends.

The findings support those of previous studies showing decreased resource utilization with anti-TNF therapy for the treatment of IBD. Schnitzler et al.32 examined whether the introduction of infliximab had changed the natural course of CD. They compared the number of surgical procedures and hospitalizations that occurred in their 614-patient cohort before and after the start of infliximab therapy. In their study, the rate of major abdominal surgery was reduced by ∼15% after the start of infliximab. Patients who were placed on maintenance (scheduled) treatment from the onset were about half as likely to require hospitalization as patients receiving episodic therapy. In another study, these investigators found that treatment with infliximab (every 8 weeks) was associated with endoscopic mucosal healing in the long-term33.

In a smaller study of similar design to the present study but conducted at only a single institution, Rubenstein et al.34 reviewed medical records for all CD patients (n = 79) for ≥1 full year both before and after the start of infliximab therapy. They reported decreases in the annual incidence of all types of surgery (38%, p < 0.01), GI surgeries (18%, p < 0.05), endoscopies (43%, p < 0.001), ER visits (66%, p < 0.05), all outpatient visits (16%, p < 0.05), outpatient GI visits (20%, p < 0.01), and radiologic examinations (12%, p < 0.01). The benefits were most marked in patients with fistulas.

Jewell et al.35 conducted a study similar to the present study but examined resource utilization outcomes only 6 months before and after infliximab initiation in 205 patients with CD. Nearly half the patients (45%) stopped taking steroids, and a further 34% had a steroid dosage reduction. A decrease of 1093 inpatient days was seen (1435 vs 342) in the 6 months after infliximab administration. There were also 7 fewer operations, 33 fewer examinations under anesthesia, and 99 fewer diagnostic procedures. The total reduction in direct costs amounted to an estimated £591,006.

We found no published data on direct assessments of resource utilization with infliximab therapy for UC; however, in the ACT-1 and ACT-2 trials, 2 randomized placebo-controlled trials, scheduled maintenance therapy with infliximab for 1 year reduced the need for hospitalization24. In a post-hoc analysis of the clinical trial data, these studies also showed improved mucosal healing with infliximab treatment. Infliximab improved mucosal appearance in ∼60% of UC patients after 8 weeks24. Moreover, mucosal healing may be associated with improved long-term remission rates and a lower risk of disease relapse14,36.

There is evidence that treatment with infliximab (along with immunosuppressive therapy) can induce long-term healing of the bowel and that this affects the outcome of both CD and UC15. Few studies have used mucosal healing as a primary endpoint, focusing instead on symptomatic outcome37. With the evidence and clinical experience of infliximab, one might expect that future studies will focus on healing and disease course as primary outcome measures. By monitoring and treating endoscopically apparent disease in addition to symptoms, patients may decrease their chances of long-term complications and the requirement for surgery.

Limitations

This study is a retrospective chart review that carries potential limitations. Results may have been affected by data quality, which depended on the completeness, accuracy, and integrity of the secondary source and data collection process. Data were collected from GI clinic medical records; therefore, resource utilization in other healthcare settings (e.g., hospital, other clinics, etc.) may have been variably recorded. It would not be expected, however, that undocumented resource use would differentially impact the pre-infliximab vs post-infliximab results, as the same patients were followed longitudinally during those 2 time periods. Further, the electronic capture mechanism should ensure consistent capture across charts. The potential limitations could be addressed, and the results of this study validated, via a prospective study design. The retrospective design precluded patient evaluation of disease-related symptom severity or formalized provider assessments, such as the Crohn’s Disease Activity Index. Future studies of resource utilization may include concurrent medications (e.g., intensity of steroid utilization), patient and caregiver time considerations, and all-cause resource utilization, in addition to outcomes associated with recently approved agents for patients with CD (e.g., adalimumab, certolizumab pegol). Disease severity data were limited by the variable data completeness, the lack of consistent timepoint evaluations, and the subjective nature of the assessment (i.e., not based on pre-defined definitions) by an experienced blinded gastroenterologist. Further information about the severity of disease and the extent of mucosal healing could be helpful in determining the effect of therapy on resource utilization. Additionally, a greater number of charts with disease severity information for both pre-treatment and post-treatment could add statistical significance to results that currently look promising. Unfortunately, the number of patients with pre-treatment and post-treatment disease severity information was too small to make comparisons among disease severity groups in this study. Finally, the pre–post study design is associated with limitations as well, such as the lack of a control group (due to inherent difficulty in matching patients based on disease severity, medical history, co-morbid conditions, practice site, etc.) and regression to the mean. Reductions in resource utilization observed here may not be due to infliximab alone but also to chance or natural history of the disease; therefore, no causal inferences can be determined.

Conclusions

The present multi-center medical chart review from 14 US gastroenterology clinics reported substantially reduced GI-related resource utilization, particularly colonoscopy, in patients with either CD or UC during infliximab treatment compared with the pre-infliximab period. Additionally, a smaller percentage of patients in both groups required surgical interventions; UC patients required significantly fewer GI-related hospitalizations, and CD patients had significantly fewer radiology assessments. When estimating the cost savings associated with the decrease in services following infliximab initiation, it is possible that the cost savings would offset the cost of infliximab administration. Concomitantly, disease severity appeared to improve in both patient groups based on a blinded gastroenterologist review of available procedure results. These factors must be taken into consideration when prescribing a treatment regimen for IBD.

Transparency

Declaration of funding

This study was supported by Janssen Scientific Affairs, LLC, Horsham, PA.

Declaration of financial/other relationships

J. E. Vanderpoel, B. Nejadnik, and R. S. McKenzie are employees of Janssen Scientific Affairs, LLC. H. C. Waters was an employee of Janssen Scientific Affairs, LLC, at the time at which this research was conducted and is currently employed by Ethicon, Inc. H. C. Waters was involved in study design, interpretation of data, and preparation and review of the manuscript; R. S. McKenzie was involved in study design, interpretation of data, and review of the manuscript; J. E. Vanderpoel was involved in interpretation of data and preparation and review of the manuscript; B. Nejadnik provided blinded review of endoscopic data, interpreted clinical data, and reviewed the manuscript. O. Lunacsek (statistical analysis) and B. J. Lennert (study coordinator) are employed by Xcenda, LLC, which has a consultant relationship with Janssen Scientific Affairs, LLC. J. Goff and D. H. Augustyn served as scientific advisors to the project and were compensated for their review of the study design and manuscript.

Acknowledgments

The authors wish to thank the following Xcenda, LLC, staff who assisted in the preparation of this article: Alberto Vigo, RN, MSN, and Tim Frank, CPC, LPTA, chart reviewers, and Luther Hill, software engineer, who programmed the Microsoft Access data collection tool. Gianna Paone and Robert Achenbach of Janssen Services, LLC, provided editorial and submission support.