Infusion administration billing for vedolizumab and infliximab in inflammatory bowel disease

Abstract

Aims: Increasing use of biologics has led to interest in treatment components with potential for cost savings. This study was aimed at comparing administration times and associated costs of infliximab and vedolizumab infusions for inflammatory bowel disease (IBD).

Materials and methods: This study used claims data from the Symphony Health Integrated Dataverse to identify IBD patients using infliximab or vedolizumab between 20 May 2014 and 29 February 2016. Use of Current Procedural Terminology administration codes was evaluated and costs calculated using the 2016 Center for Medicare and Medicaid Services Physician Fee Schedule. Assessments included infusion times, associated costs, productivity loss using average wage estimates from the United States Bureau of Labor Statistics, and home infusion adoption.

Results: A total of 10,051 infliximab and 3114 vedolizumab patients with first-hour claims were identified; 52.0% were female and 64.5% had Crohn’s disease. There were 48,377 infliximab first-hour claims (mean 4.8 infusions per patient); 46,462 (96.0%) had a second-hour claim. In comparison, there were 14,717 vedolizumab claims (mean 4.7 infusions per patient), with only 411 (2.8%) second-hour claims, resulting in vedolizumab cost savings of approximately $1.27 million. The difference in second-hour infusions resulted in 46,051 additional hours of productivity loss with infliximab, and lost wages averaging $1.18 million (range $0.68–$1.77 million).

Limitations: Administration costs were inferred as charge costs and not directly assessed. Productivity loss assessed time spent on infusion only, and included a small proportion of patients beyond working age.

Conclusions: Second-hour infusion billing was significantly lower with vedolizumab than with infliximab, corresponding to cost savings and reduced productivity loss.

Keywords:

• Vedolizumab
• infliximab
• inflammatory bowel disease
• infusion administration
• cost

JEL classification codes:

• I11
• I00

Introduction

Management of specialty drug spending in the US is an important consideration for payers and employers^1,2. Biologic drugs have transformed the landscape of inflammatory bowel disease (IBD) therapy during the past two decades, and the cost to treat IBD patients has risen considerably as a result³. With the majority of biologics requiring intravenous (IV) infusion, the non-drug costs associated with the use of these drugs to treat IBD are relatively high⁴. The recommended duration of IV infusion varies from drug to drug, leading to differences in the costs associated with the administration of biologic therapies. Infusion drugs can also be administered at various sites of care, including the physician’s office, free standing ambulatory infusion centers, hospital inpatient or outpatient departments, or at a patient’s home^5,6. Many patients in the US receive outpatient infusion therapy to treat a range of conditions; however, to date, home infusion is still relatively limited with biologic therapies for the treatment of IBD. Infusion administration in IBD is therefore an attractive area to assess potential cost savings.

Infliximab is a tumor necrosis factor (TNF) alpha inhibitor that is approved by the US Food & Drug Administration for the treatment of Crohn’s disease (CD) and ulcerative colitis (UC), as well as for several other conditions⁷. Infliximab prescribing information recommends IV infusion over a period of at least 2 hours for the treatment of patients with moderately to severely active CD and UC⁷. Vedolizumab is a gut-selective humanized immunoglobulin monoclonal antibody that selectively targets α₄β₇ integrin, expressed on activated lymphocytes, and inhibits its interaction with mucosal addressin cell adhesion molecule 1 (MAdCAM-1) that is primarily localized in the intestinal mucosa and gut-associated lymphoid tissue. Vedolizumab was approved in May 2014 for the treatment of adult patients with moderately to severely active CD or UC⁸. Administration of vedolizumab is recommended as an IV infusion over approximately 30 minutes⁸. Health plans typically reimburse for the infusion administration on an hourly basis, most commonly using Current Procedural Terminology (CPT) codes for IV or chemotherapy infusion^9,10. Based on the administration instructions in the package insert, both drugs are infused at Weeks 0, 2, and 6 and then every 8 weeks thereafter^7,8. The longer recommended infusion time for infliximab compared with vedolizumab may result in increased infusion administration costs for health plans. In some cases, the longer time allocated to receiving infusion therapy may have a significant impact on patients and their employers in terms of loss of work productivity.

No studies could be identified that assessed the impact of infusion administration time and related costs for infliximab and vedolizumab in IBD, or how they might affect the payer, the patient and the patient’s employer. In addition to the risk–benefit considerations of these two treatment options in IBD, decision making may be influenced by costs associated with infusion from payer and patient perspectives. From a payer perspective, this study aimed to examine the utilization of vedolizumab and infliximab in IBD patients in a real-world setting using claims data to compare infusion administration times. Infusion times and related costs were also assessed from the perspective of the patient and their employer to estimate the potential productivity loss associated with administration of the two drugs. Finally, adoption of home infusion of infliximab or vedolizumab was investigated throughout the study period.

Methods

Study design

This was a retrospective cohort study that used nationally representative medical and pharmacy data from the Symphony Health Integrated Dataverse (“Symphony”)¹¹, a comprehensive data set containing longitudinal pharmacy and medical claims from multiple health plans, hospital systems and pharmacy networks. Incident users of vedolizumab and infliximab for the treatment of CD or UC were identified from 20 May 2014 to 29 February 2016. We defined index as incident users of vedolizumab or infliximab for CD or UC treatment and index date as the first use of vedolizumab or infliximab between 20 May 2014 and 29 February 2016.

Patients were included in the study if they met the following criteria: aged between 18 and 89 years at the year of index; provided an index claim for infliximab or vedolizumab in the medical or pharmacy file during the identification period and with no use of the index drug in the 12 months before the index date; presence of ≥2 diagnoses for CD or UC at least 30 days apart at any time during the pre-index observation period (12 months up to and including the index date); showed pharmacy or medical claims activity in the 12 months before the index date and at least one claim for either of the study drugs. Patients were excluded if they had post-index use of another study drug, or a diagnosis for ankylosing spondylitis, psoriasis, psoriatic arthritis or rheumatoid arthritis; or if they were being treated with infused chemotherapy at any time during the study period.

Healthcare Common Procedure Coding System (HCPCS) codes and National Drug Codes (NDCs) for infliximab and vedolizumab were used to identify patients (Supplementary Table S1). Specific HCPCS codes for vedolizumab (C9026, J3380) became available on 1 October 2014 and 1 January 2016, respectively; before these dates, non-specific codes were identified and their use inferred using an internally validated algorithm: patients were first required to have used a vedolizumab-specific HCPCS or 11-digit NDC to be confirmed as vedolizumab-treated patients; then all claims using the identified non-specific codes were included only for these patients, and were subjected to a minimum service charge limit of $4500 to remove non-vedolizumab claims and possible billing errors. Diagnosis codes from the 9th and 10th revisions of the International Classification of Diseases (ICD-9 and ICD-10, respectively) were used to confirm diagnoses of CD or UC, as well as to identify diseases leading to the exclusion of patients. Patients receiving infused chemotherapy during the study period were identified using the HCPCS codes J9000 to J9998. Claims with a HCPCS code specific to home infusion were identified as home infusion claims, and home infusion claims with CPT codes were identified using the HCPCS codes for nursing time (Supplementary Table S1). The database queried contained only charged, not paid amounts.

Outcome measures

Second-hour administration and associated costs

All patients with a first-hour billing CPT code (96365 and 96413) were identified, and the use of a second-hour billing code for IV infusion (96366) or chemotherapy (96415) was evaluated. Associated costs were calculated using average fees for each code from the 2016 Center for Medicare and Medicaid Services (CMS) Physician Fee Schedule (Supplementary Table S2)¹². Drug administration costs for second-hour billing were calculated by multiplying the number of infusion billing codes by the unit cost per additional hour of infusion for chemotherapy (CPT 96415: $28.64) and IV infusion (CPT 96366: $18.98) (Supplementary Tables S2 and S3)¹². Difference in cost on a per patient basis was calculated by dividing the total second-hour claims cost for infliximab or vedolizumab by the total number of patients using each type of drug.

Productivity

Productivity loss calculations were based on the hypothesis that infusions were administered during work hours, the patient was at the infusion site of care for the duration of the claimed hours and patients returned directly to work post infusion. Productivity loss was assessed from the perspective of the patient by analyzing the number of additional hours spent receiving infusions and from the perspective of the employer using hourly wage estimates. Productivity loss was defined as the time spent away from work but did not include travel or waiting times. Costs were estimated using the human capital method¹³. Hourly wage estimates from the US Bureau of Labor Statistics (BLS) from June 2016 were used to model productivity loss for an IBD sample¹⁴. To represent a range of hourly earnings, the average hourly wage estimate across all private sectors was used ($25.61), in addition to “Leisure and Hospitality” (lowest hourly wage; $14.89) and “Utilities” workers (highest hourly wage; $38.28). The average hourly earnings were consistent over the quarter ending in June 2016 and from the same month in the previous year (June 2015)¹⁴. Furthermore, the “Leisure and Hospitality” and “Utilities” wages have consistently been the lowest and highest average hourly wage sectors, respectively, which suggests that they are appropriate approximations for the lower and upper wages for the purposes of this study.

Home infusion adoption

As a secondary objective, adoption of infliximab or vedolizumab in a home infusion setting was assessed using all available claims data. The number of claims identified as occurring in a home infusion setting were counted per month and plotted over the course of the study period as a percentage of the total number of claims for the entire period and a polynomial trend line applied. We did not use the exclusion criteria employed for the cost-analysis pool because they were not relevant to this analysis.

Statistical analysis

Descriptive statistics were used to analyze adoption of the drugs in the home infusion setting, and for baseline demographics and clinical characteristics, including frequencies, percentages, mean values and standard deviation. Inferential statistics including Pearson’s chi-square test of independence are reported for second-hour billing outcome measures.

Results

Patient population

A total of 44,170 individuals with an index claim for either infliximab or vedolizumab during the time frame of interest were identified. After sample selection criteria were applied, a total of 10,788 patients were indexed on infliximab, of whom 10,051 (93.2%) patients had a first-hour billing claim. A total of 3377 patients were indexed on vedolizumab, of whom 3114 (92.2%) patients had at least one first-hour infusion administration billing code and were included in the study (Figure 1). The study sample was geographically dispersed, with a mean age of 40.5 years (standard deviation [SD]: 16.1) for infliximab patients and 42.9 years (SD: 15.8) for vedolizumab patients (Table 1). Patient characteristics were similar between infliximab and vedolizumab patient populations, including sex (female patients: 52.0% for infliximab vs. 52.2% for vedolizumab), and diagnosis of CD (64.7% vs. 63.6%, respectively) and UC (35.3% vs. 36.4%, respectively). The study population primarily had commercial insurance coverage, and the types of insurance plans used by patients were also comparable between users of infliximab and users of vedolizumab (commercial: 83.3% vs. 82.0%, respectively; Medicare: 6.5% vs. 10.7%, respectively).

Figure 1. Patient disposition. Abbreviations. AS, ankylosing spondylitis; CD, Crohn’s disease; PsA, psoriatic arthritis; PsO, psoriasis; RA, rheumatoid arthritis; UC, ulcerative colitis.

Table 1. Demographics and clinical characteristics.

Characteristic	Infliximab	Vedolizumab
	(n = 10,051)	(n = 3114)
Female, n (%)	5228 (52.01)	1624 (52.15)
Age, years
Mean ± SD	40.5 ± 16.1	42.9 ± 15.8
18–39, n (%)	5429 (54.01)	1480 (47.53)
40–64, n (%)	3605 (35.87)	1279 (41.07)
65–74, n (%)	751 (7.47)	274 (8.80)
≥75, n (%)	266 (2.65)	81 (2.60)
Insurance plan type, n (%)^a
Commercial	8701 (83.31)	2677 (82.02)
Government^b	247 (2.36)	51 (1.56)
Medicaid	760 (7.28)	172 (5.27)
Medicare	676 (6.47)	348 (10.66)
Cash	8 (0.08)	4 (0.12)
Unknown	52 (0.50)	12 (0.37)
Geographic region, n (%)
Midwest	2995 (29.80)	1045 (33.56)
Northeast	2441 (24.29)	722 (23.19)
South	3240 (32.24)	937 (30.09)
West	1355 (13.48)	393 (12.62)
Unknown	20 (0.20)	17 (0.55)
Clinical characteristics, n (%)
Crohn’s disease	6506 (64.73)	1980 (63.58)
Ulcerative colitis	3545 (35.27)	1134 (36.42)

Abbreviation. SD, standard deviation.

^aPatients can belong to more than one plan type during the study period, so percentages were calculated from the total patient-plan types (n = 13,708).

^bFor example, Veterans Affairs.

Second-hour billing claims

There were a total of 48,377 infliximab first-hour billing claims across all payer types, and of those, 46,462 (96.0%) had an additional hour claim. In comparison, there were 14,717 vedolizumab first-hour claims with 411 (2.8%) having an additional hour claim (Figure 2). The use of second-hour CPT billing codes was significantly higher for infliximab than for vedolizumab using the chemotherapy CPT code (p < .0001) or the IV infusion CPT code (p < .0001).

Figure 2. Percentage of second-hour claims for infliximab and vedolizumab. Billing codes for chemotherapy administration and IV infusion across all sites of care were included. Abbreviations. CPT, Current Procedural Terminology; IV, intravenous. ^aPearson’s chi square test of independence. ^bClaim count.

When the unit cost of each second-hour billing CPT code for either chemotherapy or IV infusion was applied to the utilization estimates, the cost of additional hours of infusion with infliximab was estimated at $1.28 million compared with $10,477 with vedolizumab (Figure 3(A)), resulting in a cost saving of approximately $1.27 million for vedolizumab during this study period. The difference in second-hour billing cost on a per patient basis was $123.84. The billing codes, average costs for chemotherapy and IV infusion, and number of claims for each used to derive this estimate are shown in Supplementary Table S2 and Supplementary Table S3.

Figure 3. Estimated cost savings and reduction in work productivity loss with vedolizumab over infliximab following reduction in infusion hours. ^aTotal additional hours for chemotherapy administration and IV infusion combined. ^bTotal additional hours for chemotherapy administration and IV infusion combined.

Productivity

The disparity between second-hour billing claims can be interpreted as an additional 46,051 hours of loss in productivity for infliximab patients compared with vedolizumab patients. Using BLS estimates of average hourly wage across all private sectors, second-hour infusion of infliximab resulted in lost wages of $1.19 million compared with $10,526 for vedolizumab patients. The difference in second-hour claims translated into an average of $1.18 million (range using lowest and highest average hourly wage sectors: $0.68–$1.77 million) in total work productivity loss in the form of lost wages for infliximab patients compared with that for vedolizumab patients (Figure 3(B)).

Adoption in a home infusion setting

To estimate adoption in a home infusion setting, all available claims were used with no selection criteria applied (55,622 for infliximab and 16,843 for vedolizumab). We acknowledge that the variability between the two treatment groups might be different due to the larger absolute number of patients using infliximab and inherent variability in monthly data. The rate of infliximab adoption in the home setting was greater until the 15th month of the study period, after which it was surpassed by vedolizumab use at home (Figure 4). By the end of the study period, home infusion adoption was 0.75% for vedolizumab (126 out of 16,843 total claims) and 0.48% for infliximab (267 out of 55,622 total claims) as a percentage of total claims over the entire study period (Figure 4).

Figure 4. Home infusion adoption as a percentage of total claims of infliximab and vedolizumab over time. Exact percentages for home infusion adoption of each drug are shown for the final month. Dotted line represents a polynomial smoothed estimation of home infusion adoption rate. Abbreviation. Poly, polynomial.

Discussion

Using representative real-world claims data, this retrospective cohort study was the first to assess the economic impact of the label-indicated 2 hour infliximab and 30 minute vedolizumab infusion administration times. Significantly fewer second-hour infusion claims were associated with vedolizumab than with infliximab for the treatment of IBD, with just 2.8% of vedolizumab claims including an additional hour for drug administration. Although there are no comparative safety analyses between vedolizumab and anti-TNF antagonists, a recent pooled safety analysis of vedolizumab adverse event rates in CD and UC patients across multiple studies supports a favorable long-term benefit–risk profile of vedolizumab in real-world clinical practice, with low rates of infusion-related reactions¹⁵. This favorable safety profile may explain the low rate of second-hour vedolizumab billing and lower productivity loss associated with this drug, as well as some of the increase in vedolizumab home adoption rate observed at the end of the study period.

The financial burden of IBD in the US in 2014 was estimated to be between $14.6 and $31.6 billion¹⁶. There is a lack of direct cost comparisons of IBD drugs in the literature, and in particular any analyses of isolated administration costs. A study comparing the cost of therapy between infliximab and adalimumab in CD patients in the United Kingdom found that administration costs in addition to drug costs were significantly higher in those receiving infliximab¹⁷. These differences in cost of care between the two drugs were largely driven by infliximab’s weight-based dosing schedule and mode of administration that requires IV infusion, compared to the subcutaneous route of adalimumab, which can be administered by the patient¹⁷. A similar trend was observed when health utilization costs were compared between infliximab and adalimumab, with lower healthcare costs associated with adalimumab use¹⁸. The difference in infusion-related costs for infliximab and vedolizumab treatment in this study suggests that drug administration is a component of IBD therapy with the potential for cost savings.

While several studies have investigated the impact of shorter infusion times for infliximab on safety, tolerability and quality of life in IBD patients and implied associated cost savings with accelerated infusion^19–24, the label-approved administration time remains at a minimum of 2 hours⁷. A systematic review of 76 infliximab publications found that immediate infusion reactions, occurring during or within 2 hours of infusion, were reported in 5–23% of IBD patients²⁵. In comparison, ≤5% of vedolizumab-treated patients across two phase 2 and three phase 3 clinical trials in IBD experienced investigator-defined infusion-related reactions²⁶. Fewer infusion-related reactions resulting in less need for medical oversight, in addition to the shorter administration time, may also make vedolizumab a viable option for adoption in a home infusion setting.

Loss of productivity is considered an indirect societal cost of IBD. Because of absence from work due to symptoms of the illness, side effects of treatment, or time spent receiving treatment, from a patient’s or employer’s perspective, productivity loss can be measured in the form of lost earnings²⁷. Interestingly, savings in infusion administration cost for vedolizumab over infliximab were $1.27 million, while savings in financial productivity loss for the same comparison were $1.18 million. Notably, these numbers appear comparable due to the similarity between the CMS Physician Fee Schedule and average hourly wage estimates that were used as the inputs for cost saving and productivity loss, respectively.

The high cost of biologic drugs in particular is under greater scrutiny with substantial increases in their use in recent years; furthermore, the arrival of biosimilars to the IBD market is expected to have a significant impact on the economic burden of biologics to healthcare plans as patient access improves²⁸. While the market for IBD therapies may change with the approval of biosimilar drugs, infusion administration times are expected to be identical to the reference biologic agent and are unlikely to offer cost savings in this respect.

A recent literature review across several disease areas found that, in addition to reduced overall costs and effective care, the use of home infusion was associated with improved quality of life and was the preferred site of care for patients²⁹. The greater adoption of vedolizumab in the home infusion setting offers the potential to improve patient quality of life, an important consideration given the chronic nature of IBD³⁰. Further studies are needed to evaluate the extent of these savings and any improvement in quality of life for IBD patients in the home infusion setting.

Limitations

Limitations of the study include a lack of differentiation between the CMS Physician Fee Schedule for infusion administration CPT codes by “facility” or “non-facility” price, meaning that the fee schedules for these codes were the same regardless of whether the infusion was administered in a hospital or an outpatient hospital setting. Our study assumes the use of good billing practices, which is a limitation of analyzing a claims database. The data accessed for the study did not contain drug cost information. While other studies have investigated the cost-effectiveness of infliximab and vedolizumab³¹, the aim of this study was to identify potential areas for cost saving associated with infliximab and vedolizumab infusion not related to unit drug cost, including administration, financial productivity and site of care. Therefore, values for the infusion administration based on the CMS fee schedule for the CPT codes associated with infusion administration were imputed. Fee schedules may differ by payer and by site of care, and for health plans with a different reimbursement structure for facility and non-facility claims^5,6,32. These variations were not accounted for in this study. Administration costs were therefore inferred as charge costs from the CMS Physician Fee Schedule and not assessed from the Symphony data; future studies could evaluate actual drug and administration costs overall and by site of care. Commercial fee schedules are often higher compared with the CMS fee schedule, so it is likely that the reduction in second-hour infusion billing for vedolizumab would have a more pronounced effect than what is reported in this study^33,34. In addition, the unit of analysis was at the claims level, and patients were not required to be in the study for any amount of time; therefore, the study was not designed to report the yearly cost. Claims for inpatient treatment were included in the initial assessment, but as only two claims were seen, the inpatient setting was excluded from the analysis.

The loss in work productivity estimates did not include time actually spent away from work, but only included an estimate of additional hours billed for infusion after the first hour. Productivity savings assumed that patients’ infusions were administered during work hours, with patients going directly back to work post-infusion, and did not account for those who either took a half-day leave or scheduled infusions outside of work hours, or who were salaried or paid hourly wages. In addition to the fact that a small proportion of patients aged 65 years or above were included in the analysis, which may be beyond the typical working age for some in that age group, some patients under 65 who were included may have been either retired or unemployed dependents. It is likely that taking a half-day leave for infusions would lead to productivity loss. Conversely, infusions that occurred outside of work hours, were given to non-working patients or were taken during work hours by salaried employees would have a lower impact on productivity loss than what we reported. Time away from work, including travel to and from the infusion site but outside the time spent being infused, is extremely variable and therefore difficult to include. This information is not included in the Symphony database and, therefore, was not part of our calculations.

As other countries have sometimes adopted accelerated infusion schedules for infliximab, the applicability of our findings may be limited to centers that more closely adhere to the manufacturer’s instructions. In the US, where this study was conducted, the FDA-approved administration time for infliximab is 2 hours.

Conclusions

Second-hour infusion billing occurred significantly less often with vedolizumab than with infliximab for treating patients with IBD in a real-world setting, corresponding to cost savings for the health care plan and a significant reduction in loss of work productivity in the form of lost wages for the patient. Home use of vedolizumab surpassed infliximab during the study period. Future studies could evaluate the actual costs associated with the relatively short infusion time for vedolizumab across different sites of care.

Transparency

Declaration of funding

This study was funded by Takeda Pharmaceuticals USA Inc.

Author contributions

All authors contributed to the study design and implementation. All authors contributed to the preparation of this manuscript and provided final approval for publication.

Declaration of financial/other relationships

K.N., T.L. and M.L. are employees of Takeda Pharmaceuticals USA Inc., Deerfield, IL, USA. V.K. has disclosed that he was employed as a University of Illinois at Chicago Fellow at Takeda Pharmaceuticals at the time the study was conducted.

JME peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Previous presentation: This work was presented at the American College of Gastroenterology 2016 Annual Scientific Meeting (Null K, et al. Assessing administration and billing of vedolizumab and infliximab across IBD sites of care, Poster P1917); the 2016 Advances in Inflammatory Bowel Diseases, Crohn’s & Colitis Foundation of America’s Clinical & Research Conference (Null K, et al. Comparing productivity between patients on vedolizumab and infliximab patients using infusion administration billing in the United States, Poster P-102. Kumar V, et al. Economic impact of infusion administration time billing for vedolizumab and infliximab for ulcerative colitis and Crohn’s disease: health plan perspective, Poster P-092); and the Academy of Managed Care Pharmacy 2017 Managed Care & Specialty Pharmacy Annual Meeting (Kumar V, et al. Economic impact of infusion administration time billing for vedolizumab and infliximab for ulcerative colitis and Crohn’s disease: health plan perspective, Poster K9).

Acknowledgements

Ann Cameron helped with a component of the literature review and provided background materials. Ann is an employee of Xcenda Inc. Medical writing assistance was provided by Alpa Parmar PhD and Bomina Yu PhD CMPP of inVentiv Medical Communications and funded by Takeda Pharmaceuticals USA Inc.