Costs of providing infusion therapy for patients with inflammatory bowel disease in a hospital-based infusion center setting

Abstract

Aims: Inflammatory bowel disease (IBD) (e.g. ulcerative colitis [UC] and Crohn’s disease [CD]) severely impacts patient quality-of-life. Moderate-to-severe disease is often treated with biologics requiring infusion therapy, adding incremental costs beyond drug costs. This study evaluates US hospital-based infusion services costs for treatment of UC or CD patients receiving infliximab or vedolizumab therapy.

Materials and methods: A model was developed, estimating annual costs of providing monitored infusions using an activity-based costing framework approach. Multiple sources (published literature, treatment product inserts) informed base-case model input estimates.

Results: The total modeled per patient infusion therapy costs in Year 1 with infliximab and vedolizumab was $38,782 and $41,320, respectively, and Year 2+, $49,897 and $36,197, respectively. Drug acquisition cost was the largest total costs driver (90–93%), followed by costs associated with hospital-based infusion provision: labor (53–56%, non-drug costs), allocated overhead (23%, non-drug costs), non-labor (23%, non-drug costs), and laboratory (7–10%, non-drug costs).

Limitations: Limitations included reliance on published estimates, base-case cost estimates infusion drug, and supplies, not accounting for volume pricing, assumption of a small hospital infusion center, and that, given the model adopts the hospital perspective, costs to the patient were not included in infusion administration cost base-case estimates.

Conclusions: This model is an early step towards a framework to fully analyze infusion therapies’ associated costs. Given the lack of published data, it would be beneficial for hospital administrators to assess total costs and trade-offs with alternative means of providing biologic therapies. This analysis highlights the value to hospital administrators of assessing cost associated with infusion patient mix to make more informed resource allocation decisions. As the landscape for reimbursement changes, tools for evaluating the costs of infusion therapy may help hospital administrators make informed choices and weigh trade-offs associated with providing infusion services for IBD patients.

Keywords:

• Crohn’s disease
• Inflammatory bowel disease
• ulcerative colitis
• large intestine
• small intestine
• health economics
• outcomes research

Introduction

Inflammatory bowel disease (IBD) is a description for multiple chronic, relapsing inflammatory disorders that primarily affect the digestive system. The most well-known and frequently observed IBD conditions are ulcerative colitis (UC) and Crohn’s disease (CD). While the cause of these diseases is not known, they have been linked to genetic and environmental factors¹. IBD severely impact patients’ quality-of-life, causing frequent inconvenient and painful symptoms, and can affect their ability to work and fully participate in life activities².

The burden of disease for IBD is high in the US, with roughly 1–1.3 million people suffering from IBD³. The majority of the cost is attributable to medical care, with further increases in cost due to severe or poorly controlled disease leading to surgery or other expensive interventions. In fact, nearly two-thirds of patients with CD and one-third of UC patients will require surgical intervention for disease control in their lifetime^4–6. Drug therapy to manage IBD is associated with substantial costs, particularly since the introduction of innovative biologic treatments⁷. The annual cost of medical and surgical treatments for CD are estimated at nearly $2 billion USD (2008 USD), and the direct medical costs of UC are estimated at over $4 billion USD (2009 USD)⁸.

In the US, initial disease management for UC and CD patients typically consists of drug therapy. Agents, such as aminosalicylates, may be prescribed to control disease course and manage symptoms, with corticosteroids added during periods of exacerbated symptoms and/or non-responsive disease. Patients who progress or experience consistent moderate-to-severe disease activity are typically transitioned to treatment with biologic therapy. Once a patient is considered eligible to receive biologic therapy, clinicians must determine whether a self-injected or infused therapy would be the best treatment choice. If prescribing infusion therapy, clinicians may select one of two currently approved therapies: infliximab (Remicade) or vedolizumab (Entyvio). At present, the published American College of Gastroenterology (ACG) guidelines are limited to guidance about the use of infliximab for patients with moderate-to-severe UC and CD^7,8.

Institutions that administer monitored infusions for IBD incur costs beyond drug acquisition. The aim of this study was to develop a model framework to undertake a micro-costing analysis to evaluate the majority of identifiable costs associated with the delivery of hospital-based infusion services (preparation, administration, and follow-up) in the US for treatment of patients with UC or CD receiving treatment with either infliximab or vedolizumab. As such, the model provides infusion center administrators and clinicians treating IBD with an adaptable tool to identify and assess the myriad of costs related to the provision of infusion therapy.

Methods

A model was developed in Microsoft Excel to estimate the annual costs associated with providing monitored infusions of either infliximab or vedolizumab to patients for treatment of moderate-to-severe IBD (UC and CD) in a US hospital setting. For this analysis, an activity-based costing (ABC) framework approach was used. This method allows for the estimation of direct costs for a specific activity by comprehensively identifying the set of resources necessary to complete that activity, in order to calculate total cost of care^9,10. The model analyses were conducted from the perspective of a US hospital.

Base-case model input estimates were obtained from multiple sources. First, administrators at five community hospital-based infusion centers were surveyed about their patient volume, patient mix, and overheads¹¹. Second, staff time requirements for reconstitution and administration of infusion drugs^12,13 and treatment for infusion reactions¹⁴ were obtained from published literature. Third, product package inserts informed the majority of the assumptions around the administration, dosing, schedule of monitored infusions, and rates of infusion reactions^15,16. Fourth, standard costing sources were used for medical staff wages¹⁷ and medications¹⁸, while costs for supplies were determined based on internet searches of medical suppliers.

Activities relevant to providing infusion services were identified and limited to the hospital pharmacy or related infusion center pharmacy (for preparation of infusion), and the infusion center itself (for administration of infusion and management of patient). Costs obtained from the published literature were adjusted to 2015 US dollars using the mid-year medical care component of the Consumer Price Index¹⁹.

A complete listing of all base-case model estimates (time, costs, facility characteristics, and practice patterns) is displayed in Table 1; provided below is an overview of data sources chosen to inform the base-case estimate.

Table 1. Base-case model inputs.

Section	Item	Baseline estimate	Annotation
Population	Total # patients new to infusion therapy	CD: 15% UC: 15%	Assumption; informed by chart review study by Tkacz et al.^21
Population	Average patient weight	80 kg	Assumption informed by Rutgeerts et al.^37, Waters et al.^22, and McConnell et al.^20 In the ACT 1 trail, mean weight for the 5 mg/kg group was 80 kg; the ACT 2 trial 5 mg/kg group had a mean weight of 78 kg^37 Mean weight reported in the Waters et al.^22 paper (used for Year 1 infliximab treatment pattern assumptions) was 81 kg for CD and 80 kg for UC In McConnell et al.^20 (used to populate Year 2+ assumptions for infliximab treatment patterns), mean weight for CD patients was 82 kg and for UC was 86 kg
Health system	Allocated overhead (%)	Allocated overhead: 30%	Assumption; Components of the overhead cost are: Administrative & General, Maintenance & Repairs, Operation of Plant, Laundry & Linen, Housekeeping, Dietary, Cafeteria, Maintenance of Personnel, Nursing Admin, Central Svc & Supply, Pharmacy, Medical Records, Social Services, Total Capital (including Buildings & Fixtures and Major Movable Equipment) The model default assumes allocated overhead is applied to all non-drug costs
Infliximab drug cost, dose, and infusion duration (per infusion)	Infliximab 5 mg/kg	Dose description (average dose for a 80 kg adult): 5 mg/kg =400 mg Unit supplied (mg/vial): 100 mg Average unit(s) per infusion: 4 Unit cost: $1021.43 Administration time (min): 120.00	Unused drug, if any, is assumed to be discarded at the time of the infusion Total cost of infusion drug (per infusion, 80 kg patient): $1,021.43 * 4 vials = $4,085.72^16^,^18
Infliximab drug cost, dose, and infusion duration (per infusion)	Infliximab 10 mg/kg	Dose description (average dose for a 80 kg adult): 10 mg/kg =800 mg Unit supplied (mg/vial): 100 mg Average unit(s) per infusion: 8 Unit cost: $1021.43 Administration time (min): 120.00	Unused drug, if any, is assumed to be discarded at the time of the infusion Total cost of infusion drug (per infusion, 80 kg patient): $1,021.43 * 8 vials = $8,171.44^16^,^18
Infliximab infusion schedule	# of induction infusions, infliximab	infliximab: 3 infusions at a dose of 5 mg/kg on Weeks 0, 2, and 6	Number and dose of induction infusions drawn from the USPI^16
Infliximab infusion schedule	Schedule of maintenance infusions, Infliximab	Standard schedule: infusions every 8 weeks Increased frequency (infliximab only): infusions every 6 weeks	Standard schedule of maintenance infusions drawn from the USPI. The standard maintenance regimen for infliximab is administered every 8 weeks (7 infusions annually). Patients who receive more frequent infusions (every 6 weeks) are assumed to receive 9 infusions annually^16 Increased frequency schedule informed by reports of real-world use of infliximab in which patients who experience loss of response may be offered more frequent infusions to address attenuated response to infliximab therapy for CD or UC^20–22
Practice patterns with infliximab, by disease	Time to loss of response with Infliximab (weeks)	CD: 38 weeks UC: 38 weeks	Median time to loss of response reported in ACCENT I trial for the 5 mg/kg group was 38 weeks (IQR =17 to >54 weeks). No data found for loss of response in UC; thus, the model assumes the same estimate as CD^34
Practice patterns with infliximab, by disease	% patients who dose escalate to 10 mg/kg dose, Infliximab	CD: 10% UC: 19%	Tables 3 and 4 of Waters et al., values for patients initiated on an 8-week maintenance schedule^22
Practice patterns with infliximab, by disease	% patients who require increased frequency of administration (i.e. change to 6-week schedule)	CD: 6% UC: 9%	Tables 3 and 4 of Waters et al., values for patients initiated on an 8-week maintenance schedule^22
Practice patterns with infliximab, by disease	% patients who both increase dose and require increased frequency of administration (i.e. change to 6-week schedule)	CD: 1% UC: 9%	Tables 3 and 4 of Waters et al., values for patients initiated on an 8-week maintenance schedule^22
Practice patterns with infliximab, by disease	Distribution of patients by dose	CD: 5 mg/kg =51%; 10 mg/kg =49% UC: 5 mg/kg =56%; 10 mg/kg =44%	Informed by IBD patient data reported in Table 3 of McConnell et al. Study of 46 IBD patients (37 CD; 9 UC) on infliximab therapy who had >4 infliximab infusions and were considered for accelerated infliximab infusions. Paper reported detailed characteristics (weight, maintenance dose, infusion schedule, # prior infusions) for all patients enrolled in the study. UC patients had an average of 20 prior infliximab infusions and the CD group had an average of 29 prior infusions. Patients receiving doses >5 mg/kg were considered as having a dose of 10 mg/kg^20
Practice patterns with infliximab, by disease	Schedule of infusions (5 mg/kg dose)	CD: 8-weeks =84%; 6-weeks =16% UC: 8-weeks =80%; 6-weeks =20%	Informed by IBD patient data reported in Table 3 of McConnell et al. Calculated the distribution of infusion schedules for patients on 5 mg/kg therapy for ≥1 year by dose assuming patients with 7- and 8-week maintenance schedules were on a stable 8-week schedule. Patients receiving infusions every 6 weeks (or less) were considered as having infusion every 6 weeks^20
Practice patterns with infliximab, by disease	Schedule of infusions (10 mg/kg dose)	CD: 8-weeks =55%; 6-weeks =45% UC: 8-weeks =50%; 6-weeks =50%	Informed by IBD patient data reported in Table 3 of McConnell et al. Calculated the distribution of infusion schedules for patients on >5 mg/kg therapy for ≥1 year by dose assuming patients with 7- and 8-week maintenance schedules were on a stable 8-week schedule. Patients receiving infusions every 6 weeks (or less) were considered as having infusion every 6 weeks^20
Acute infusion reactions with infliximab	Acute infusion reaction rates	Infliximab: Mild-to-moderate reactions: 18.0% Severe reactions: 0.5%	Section 6.1 of the Prescribing Information: An infusion reaction was defined in clinical trials as any adverse event occurring during an infusion or within 1 h after an infusion. In phase 3 clinical studies, 18% of REMICADE-treated patients experienced an infusion reaction compared to 5% of placebo-treated patients. Serious infusion reactions occurred in <1% of patients and included anaphylaxis, convulsions, erythematous rash, and hypotension^16 For serious infusion reactions, a rate of 0.5% was assumed (mid-point of 0–1%)^16
Vedolizumab drug cost, dose, and infusion duration (per infusion)	Infusion drug (per infusion) - Vedolizumab	Dose description: 300 mg Unit supplied (mg/vial): 300 mg Average unit(s) per infusion: 1 Unit cost: $4,819.00 Administration time (min): 30.0	Standard maintenance dose is 300 mg administered every 8 weeks. No data available for alterative dosing and/or administration schedules^15 Total cost of infusion drug (per infusion): $4,819.00 * 1 vials = $4,819.00^15^,^18
Vedolizumab infusion schedule	# of induction infusions, Vedolizumab	vedolizumab: 3 infusions at a dose of 300 mg on Weeks 0, 2 and 6	Number and dose of induction infusions drawn from the USPI^15
Vedolizumab infusion schedule	Schedule of maintenance infusions, Vedolizumab	Standard schedule: infusions every 8 weeks	Standard schedule of maintenance infusions drawn from the USPI^15 Vedolizumab: Standard maintenance regimen is administered every 8 weeks (7 infusions annually). No data available concerning the use of alternative schedules for infusions^15
Acute infusion reactions with vedolizumab	Acute infusion reaction rates	Vedolizumab: Mild to moderate reactions: 4% Severe reactions: 0.1%	Rates of acute infusion reactions are patient rates rather than rates across infusions^15
Infusion services	Personnel - Wages & Benefits	(1) Physician - (Hourly Wage: $93.57) (2) Nurse Practitioner (ARNP) - (Hourly Wage: $48.12) (3) Nurse (RN) - (Hourly Wage: $34.27) (4) Pharmacist - (Hourly Wage: $58.18) (5) Pharmacy Tech/Aide - (Hourly Wage: $15.27) Proportion of total compensation for benefits: 33.7%	Wage estimates represent the average hourly wage reported by the National Occupational Employment and Wage Estimates website (2014 US$)^17: (1) Internists, General (29-1063): $91.60 (2014 US$) (2) Nurse Practitioners (29-1170): $47.11 (2014 US$) (3) Registered Nurses (29-1140): $33.55 (2014 US$) (4) Pharmacists (29-1050): $56.96 (2014 US$) (5) Pharmacy Technicians (29-2052): $14.95 (2014 US$) The proportion of total compensation for benefits is drawn from Table 28 of US Bureau of Labor Statistics (2015) which reports the percent of total compensation benefits rate for hospitals^23 Wages updated from 2014 US$ to 2015 US$ using the inflation rate of 1.0144 based on changes to the Annual Medical Services CPI Index^19
Infusion services	Labor, Pharmacy (per infusion)	Value of pharmacy labor per infusion: $21.66 Component tasks with time estimates for pharmacists and pharmacy techs: Treatment evaluation: RPh (2.86 min); tech (3.07 min) Consultation: RPh (2.85 min); tech (3.07 min) Patient care: RPh (1.33 min); tech (1.46 min) Order entry: RPh (2.96 min); tech (3.17 min) Production/evaluation: RPh (1.07 min); tech (1.14 min) Interruption/other: RPh (0.48 min); tech (0.52 min)	Calculated based on Table 4 of Brixner et al.^12 adjusted to percent of patients receiving a biologic infusion
Infusion services	Labor Infusion Services - Staff providing infusion services (% utilization)	Weighted average hourly wage: $56.55 Components: Staff role and % utilization providing infusions to patients Physician: 4.5% utilization Nurse Practitioner (ARNP): 4.5% utilization Nurse (RN): 91.0% utilization	Pierce and Baker^13 indicate 9% of infusions are performed by physicians and ARNPs (we assume 4.5% utilization for each role). The remaining 91% of infusions are assumed to be administered by RNs
Infusion services	Infusion Administration Common Activities	Pre-service: (11 min) Initial stage: (8.1 min) Treatment Admin (Stage 1): (20.5 min) Treatment Admin (Stage 2): (product-specific, informed by USPI) Completion stage: (3.4 min) Post-service: (6.8 min)	Figure 5 of Pierce and Baker^13 Treatment Admin (Stage 2) is the amount of time spent infusing the patient. These estimates are a drug-specific input located on the drug-specific inputs (Infusion Drug) page in the model; default administration times are pulled directly from the USPI for each product. The model assumes that only a portion of the infusion administrative time is spent in direct patient care activities (64%)^13
Infusion services	Labor, Management of Acute Infusion Reactions (per infusion)	Additional intra-service time due to stop the infusion in order to manage the patient’s reaction: 20 min (mild/moderate) and 30 min (severe) In addition, we assume the acute reaction occurs half-way through the infusion. For mild/moderate acute infusion reactions, when the infusion is resumed, the model assumes it takes double the remaining time to complete the infusion. For severe reactions, the model assumes the infusion is not re-started	Calculation: Time (min) to manage an acute infusion reaction is product-specific due to differences in administration times. For example, the model assumes it takes 80 min of RN time to administer one infusion. If an infliximab patient experienced an acute reaction, the model would assume 20 min of additional time to interrupt the infusion and treat the reaction +40 min of extra RN administration time because the infusion flow rate would be reduced by 50% (i.e. one additional hour of RN time would be required for patients who experience an acute infusion reaction with infliximab)^14
Facility	Administration & Management, Pharmacy (per infusion)	Total: $17.22 Component costs: (1) Inventory management (annual physical inventory & managing current inventory): $5.18 (2) Insurance management (insurance management, coding, reimbursement tasks): $11.16 (3) Disposal of wasted material (labor only): $0.88	Table 3 of Brixner et al.^12. Costs adjusted to 2014 US$ from 2003 US$ based on the medical care component of the CPI (inflation factor =1.451,6) Component costs: (1) Inventory management (annual physical inventory & managing current inventory): $3.46 (2003 US$)^12 (2) Insurance management (insurance management, coding, reimbursement tasks): $7.46 (2003 US$)^12 (3) Disposal of wasted material (labor only): $0.59 (2003 US$) Pharmacy labor that is not associated with direct patient care^12
Facility	Administration & Management, Facility (per infusion)	None; calculated field	Brixner et al.^12
Infusion services	Supplies, Pharmacy (per infusion)	infliximab 5 mg/kg: $13.76 infliximab 10 mg/kg: $20.48 vedolizumab 300 mg: $8.51 Component supplies for reconstitution, Units Cost/Unit Total (1) Gloves (pair) 1.0 unit*$0.19/unit = $0.19 (2) 10 mL sterile water (infliximab only) 5 mg/kg: 4.0 units*$0.87/unit = $3.48 10mg/kg: 8.0 units*$0.87/unit = $6.96 (3) Syringe with 21 gauge or smaller needle (infliximab only) 5 mg/kg: 4.0 units*$0.76/unit = $3.07 10 mg/kg: 8.0 units*$0.76/unit = $6.13 (4) Alcohol swab 5 mg/kg: 4.0 units*$0.05/unit = $0.20 10mg/kg: 8.0 units*$0.05/unit = $0.40 (5) 0.9% NaCl injection; 250 mL infusion bottle/bag 1.0 unit*$4.06/unit = $4.06 (6) Impervious gown 1.0 unit*$1.57/unit = $1.57 (7) Mask 1.0 unit*$1.22/unit = $1.22 (8) 5 mL sterile water (vedolizumab only). 1.0 unit*$1.05/unit = $1.05 (9) 5 mL syringe with 21 gauge or smaller needle (vedolizumab only). 1 unit * $0.37/unit = $0.37	(1) Gloves ($9.49 per 100)^38 (2) Sterile Water, 10 mL for injection (WAC $21.75 for case of 25)^18 (3) 10 mL Safety-Lok Syringe 21 g × 1-1/2" Detachable Needle (Box of 50 - $38.32)^39 (4). Alcohol swab ($60.13 for case of 1,200)^40 (5) 0.9% NaCl; 250 mL prefilled infusion bottle/bag (Average WAC $97.49 for case of 24)^18 (6) Impervious Gown ($15.66 for case of 10)^41 (7) Disposable particulate respiratory mask ($42.54 for case of 35)^42 (8) Sterile water, 5 mL ($26.25 for a case of 25)^18 (9) 5 mL Luer-Lok Syringe 21 g × 1" PrecisionGlide Detachable Needle (Box of 100 - $37.00)^43 Baseline values are estimates only, and are intended to be used as a starting point in the model. Individual health facilities may be able to obtain more favorable pricing based on contracting relationships and/or bulk purchases; if so, baseline estimates may over-estimate costs. Users are encouraged to enter facility-specific information throughout the model
Infusion services	Supplies, Pre-medication (per infusion)	Total cost for infliximab: $1.60 (75% diphenhydramine, acetaminophen; 20% hydrocortisone) Total cost for vedolizumab: $0.39 (5% diphenhydramine, acetaminophen, hydrocortisone) Component costs: (1) Diphenhydramine (25 mg p.o.) 1 unit: $0.02 each (2) Acetaminophen (650 mg p.o) 2.0 units. 325 mg unit: $0.01 each = $0.02 (3) Hydrocortisone (100 mg IV) 1 unit: $7.83 each	Infliximab: USPI recommends pre-treatment with an anti-histamine, acetaminophen, and/or corticosteroid. Model assumes 75% receive treatment with diphenhydramine and acetaminophen and 20% receive hydrocortisone (limited to the proportion of patients who experience acute infusions reactions)^16 Vedolizumab: USPI does not recommend pre-treatment. Pre-treatment with diphenhydramine, acetaminophen, and hydrocortisone assumed only in patients who experience acute infusion reactions (∼ 5%)^15 (1) Diphenhydramine (25 mg tablet): Average WAC price $17.48 for 1000 tablets = $0.017 each^18 (2) Acetaminophen (325 mg tablet): Average WAC price $7.77 for 1000 tablets = $0.01 each^18 (3) Hydrocortisone (100 mg IV): WAC price for 100 mg vial is $7.83^18 Baseline values are estimates only, and are intended to be used as a starting point in the model. Individual health facilities may be able to obtain more favorable pricing based on contracting relationships and/or bulk purchases; if so, baseline estimates may over-estimate costs. Users are encouraged to enter facility-specific information throughout the model
Infusion services	Supplies, Infusion (per infusion)	infliximab: $16.11 vedolizumab: $11.09 Component costs: (1) IV Start Kit (non-latex) 1.0 units * $3.42/unit = $3.42 (2) Basic IV Set (non-latex) 1.0 units * $5.90/unit = $5.90 (3) IV Admin Set, Lifeshield W/1.2microfilter. 1.0 units * $6.79/unit = $6.79 (infliximab only) (4) 10 mL Sodium Chloride Flush Intravenous Solution 0.9% 3.0 units * $0.59 = $1.77 (vedolizumab only)	(1) IV Start Kit (non-latex): Medical Supply Group ($170.96 per case of 50; includes 2 gloves, 1 label, 1 transport tape, 1 PVP prep pad, 1 alcohol prep pad, 1 gauze, 1 tourniquet, 1 tegaderm)^44 (2) Basic IV Set (non-latex; $295.00 for a case of 50)^45 (3) IV Admin Set, Lifeshield W/1.2 microfilter (infliximab only; $325.85 per case of 48)^46 (4) 10 ml Sodium Chloride Flush Intravenous Solution 0.9% (vedolizumab only; $469.60 for a case of 800)^18 Baseline values are estimates only, and are intended to be used as a starting point in the model. Individual health facilities may be able to obtain more favorable pricing based on contracting relationships and/or bulk purchases; if so, baseline estimates may over-estimate costs. Users are encouraged to enter facility-specific information throughout the model
Infusion services	Supplies, Treatment of Acute Infusion Reactions (mild/moderate)	Total cost of supplies per event: $0.60 Component costs: (1) Acetaminophen 650 mg p.o.: $0.02 (2) Diphenhydramine 25 mg IV: $0.58	Choice of treatment drawn from Figure 2A of Cheifetz et al.^14 (1) Acetaminophen (325 mg tablet): Average WAC price $7.7 for 1000 tablets = $0.01 each^18 (2) WAC price for diphenhydramine HCl injection (50 mg/mL) supplied in a 10 mL vial (pkg of 10 vials) = $115.00. Cost per 50 mg dose: ($115/10)/10 = $1.15^18
Infusion services	Supplies, Treatment for Acute Infusion Reactions (severe)	Total cost of supplies per event: $11.56 Component costs: 50% use of: (1) Acetaminophen 650 mg p.o.: $0.02 (2) Diphenhydramine 25 mg IV: $0.58 100% use of: (3) Epinephrine 0.3 mg (1:1000 IV PUSH) 1 unit @ $0.05 (4) Hydrocortisone 100 mg IV PUSH; 1 unit @ $7.83 (5) NS flush (NaCl 0.9% 500 mL IV); 1 unit @ $3.38	Choice of treatment drawn from Figure 2A of Cheifetz et al.^14 (1) Acetaminophen (325 mg tablet): Average WAC price $7.77 for 1000 tablets = $0.01 each^18 (2) WAC price for diphenhydramine HCl injection (50 mg/mL) supplied in a 10 mL vial (pkg of 10 vials) = $115.00. Cost per 50 mg dose: ($115/10)/10 = $1.15^18 (3) Hydrocortisone (100 mg IV): WAC price for 100 mg vial is $7.83^18 (4) WAC price for epinephrine HCl injection (/mL) supplied in 30 mL vial (pkg of 10 vials) = $45.00. Cost per 1 mg dose: ($45/30)/10 =$0.15^18 (5) NS (NaCl 0.9% 500 mL IV bag); Average WAC price $81.01 for 24 bags = $3.38^18
Facility	Facility Characteristics	(1) Weekly number of infusion patients (all diagnoses) = 100 (2) Size of infusion facility (sq feet) = 2500 (3) Weeks open per year =52	(1) Weekly number of infusion patients (all diagnoses): data from survey^11 (2) Size of infusion facility (sq feet): assumption based on the usual size for an infusion center (3) Weeks open per year: data from survey^11
Facility	Years to Depreciate Capital	5 years	Recovery period for ADS high technology medical equipment is assumed to be 5 years^25
Facility	Non-Labor, Pharmacy (per infusion)	Total: $6.49 Components costs: (1) Storage (refrigerators, cabinets, freezers): $0.30 (2) Space rental (patient & non-patient rooms, utilities, maintenance): $1.08 (3) Equipment (biologic safety cabinets, ventilation equip, telecommunications & IT equip): $0.87 (4) Supplies: $4.19 (5) Information resources (drug information resources): $0.05	Table 3 of Brixner et al.^12, adjusted from 2003 to 2015 US$ based on the medical care component of the CPI Costs adjusted to 2015 US$ from 2003 US$ based on the Medical Care Component of the CPI (inflation factor =1.451,6)^19 Components costs: (1) Storage (refrigerators, cabinets, freezers): $0.20 (2003 US$)^12 (2) Space rental (patient & non-patient rooms, utilities, maintenance): $0.72 (2003 US$)^12 (3) Equipment (biologic safety cabinets, ventilation equip, telecommunications & IT equip): $0.58 (2003 US$)^12 (4) Supplies: $2.80 (2003 US$)^12 (5) Information resources (drug information resources): $0.03 (2003 US$)^12
Facility	Non-Labor, Facility (per infusion)	Total: $21.48 Unit costs are listed below. Detail on calculations for adjusted per infusion cost are entered in the Annotation field. (1) Rent, utilities, facilities (per sq foot) = $44.36 (2) Infusion pump (Abbott Plum A Plus) = $1,299.00 (3) Infusion pump stand = $231.99 (4) Electronic blood pressure machine = $1,522.50 (5) Infusion chair = $699.27 (6) Electronic thermometer = $229.00	(1) Calculation: Adjusted cost per infusion for rent, utilities, facilities = (rent per sq ft * # sq feet)/(weekly infusion patients * weeks open per year)^19^,^24 Calculation: Adjusted (depreciated) cost of equipment per infusion = (equipment cost/years to depreciate capital)/(weekly infusion patients * weeks open per year)^25 Costs (per sq. foot) for rent, utilities, facilities adjusted to 2015 US$ from 1999 US$ based on the medical care component of the CPI (inflation factor =1.720,9)^19 (2) Infusion pump ($1,299)^47 (3) Infusion pump stand ($231.99)^48 (4) Blood pressure monitor (portable, $1,522.50 (mid-point of range of $245–$2,800), shipping unspecified)^49 (5) Infusion chair ($699.27)^50 (6) Electronic thermometer ($229.00)^51 Baseline values are estimates only, and are intended to be used as a starting point in the model. Individual health facilities may be able to obtain more favorable pricing based on contracting relationships and/or bulk purchases; if so, baseline estimates may over-estimate costs. Users are encouraged to enter facility-specific information throughout the model
Laboratory testing	Laboratory testing (annual)	Annual laboratory testing: $250.11 Components: Laboratory Test (CPT Code); unit cost; Frequency of annual testing (1) Metabolic panel (80048); unit cost: $11.51; Frequency of annual testing: 4.0 (2) Hepatic function panel (80076); unit cost: $11.11; Frequency of annual testing: 4.0 (3) Complete CBC w/auto diff WBC (85025); unit cost: $10.61; Frequency of annual testing: 4.0 (4) Sedimentation rate (85652); unit cost: $3.68; Frequency of annual testing: 4.0 (5) C-reactive protein (86140); unit cost: $7.04; Frequency of annual testing: 4.0 (6) Hep A antibody total (86708); unit cost: $16.86; Frequency of annual testing: 2.0 (7) Hep B antibody total (86704); unit cost: $16.40; Frequency of annual testing: 2.0 (8) TB intradermal test (86580); unit cost: $7.04; Frequency of annual testing: 1.0	Patients receiving infusion therapy are assumed to have hematology, renal function, and liver function tests every 12 weeks^16 ESR and CRP every 12 weeks are included per UC/CD guidelines and KOL expert opinion^7^,^52 The schedule of 2 hepatitis A and B antibody tests and 1 TB intra-dermal test annually are informed by the USPI for both infusion products^15^,^16 (1–7) For CPTs 80048, 80076, 85025, 85652, 86140, 86704, 86708: used the national limit as the default cost^26 (8) For CPT 86580: used the non-facility charge^27
Patient	Patient: Wages & Benefits	(1) Hourly Wage: $22.72 (2) Proportion of total compensation for benefits: 31.5%	(1) Wage estimates drawn from Table 1 of US Bureau of Labor Statistics (2015). Wages represent the cost per hour worked for civilian workers (June 2015): $22.72^23 (2) Benefits rate drawn from Table 1 of US Bureau of Labor Statistics (2015). Benefits rate for all occupations, civilian workers for June 2015^23
Patient	Patient Employment, by disease and year of therapy	CD: Year 1: 62% Subsequent year: 68% UC: Year 1: 65% Subsequent year: 68%	The value of lost work is only calculated for the proportion of employed patients in each treatment group^7^,^32
Patient	Time Required on the Day of Infusion Therapy	Travel to/from infusion center (roundtrip): 120 min (both treatments) Wait time in infusion center (total): 120 min (both treatments	Buisson et al.^30 report that infliximab patients spent a median of 2 h in transit to/from the infusion center and a median of 2 h in wait time at an infusion center on the day of an infusion. This time did not include the time to infuse drug or time required to administer pre-treatment as prophylaxis for acute infusion reactions
Patient	Time Required for Pre-Treatment (patient time)	30 min	Both product package inserts and a manuscript summarizing time spent by patients who receive infusions for IBD indicate the standard time required to complete pre-treatment is 30 min^15^,^16^,^30 Rate of pre-treatment is drug-specific^15^,^16
Patient	Transportation	$15.40 per infusion	Assumption informed by Luce et al.^33. Assumed 50% private vehicle mileage and parking at $6/round trip ($.56/mile ×30 miles, $8 parking = $24.80), 50% public transportation at $6/round trip^33

ADS: alternative depreciation system; ARNP: advanced registered nurse practitioner; CBC: complete blood count; CD: Crohn’s disease; CPI: consumer price index; CPT: current procedural terminology; CRP: c-reactive protein; ESR: erythrocyte sedimentation rate; g: gram; HCI: hydrochloride; Hep A: hepatitis A; Hep B: hepatitis B; IBD: inflammatory bowel disease; IQR: interquartile range; IV: intravenous; kg: kilograms; mg: milligrams; mL: milliliter; NaCl: sodium chloride; NS: normal saline; pkg: package; p.o.: per os (oral administration); PVP: povidone prep; RN: registered nurse; RPh: registered pharmacist; WAC: wholesale acquisition cost; sq: square; TB: tuberculosis; UC: ulcerative colitis; US: United States; USPI: United States product insert.

Time estimates

An observational study by Pierce and Baker¹³ detailed the process model that was used to estimate time spent by nursing staff to administer an infusion. Time spent for administration of infusion therapy varied by product, and matched the product package inserts’ specifications for total infusion time^15,16. The model assumed infusion services were predominantly delivered by nurses¹³. An observational study by Brixner et al.¹², that quantified chemotherapy preparation time, informed estimates of pharmacist and technician time per infusion, in addition to overhead costs related to infusion preparation.

Pre-medication as prophylaxis for acute hypersensitivity reactions was assumed for treated patients, as outlined in the product package inserts (e.g. use of diphenhydramine prior to administration of therapy) and confirmed with a clinical expert^15,16. Infusion reaction rates for infliximab and vedolizumab were obtained from the package insert^15,16. A published review informed time estimates for medical staff managing hypersensitivity reactions during a patient’s infusion¹⁴. Time estimates for nurse management of infusion reactions were weighted by frequency of events.

Cost estimates

Infusion drug

The model considered infusion therapy for UC and CD with the biologics approved for marketing by the US Food and Drug Administration (FDA) of January 1, 2015 (infliximab or vedolizumab); product package inserts informed the dosing and schedule of infusions^15,16. An average patient weight of 80 kg was assumed to calculate the total dose for infliximab infusions.

The infliximab package insert identifies a range of doses for the treatment of CD and indicates that patients’ dose is individualized based on their clinical response, noting that patients who initially respond to treatment but later lose their response may benefit from the increased dose of 10 mg/kg infliximab¹⁶. Published literature on real-world treatment patterns with infliximab informed assumptions around dose escalation to 10 mg/kg for UC and CD patients receiving infliximab therapy in Year 1 or a stable maintenance dose (5 mg/kg or 10 mg/kg) in subsequent years (Year 2+) for patients who experienced an inadequate response to infliximab therapy^20–22.

The total annual cost of infusions was based on the number of infusions for each product in a given year, with users specifying the number of patients in their first year of therapy. In Year 1, patients receiving both vedolizumab and infliximab were assumed to receive eight infusions in total (i.e. three induction doses by Week 6, followed by an infusion every 8 weeks thereafter)^15,16. In Year 2 or later, vedolizumab patients were assumed to receive seven infusions annually (i.e. infusions every 8 weeks). This assumption differed for infliximab, based on real world data that suggests a proportion of patients receive dose modifications to manage inadequate treatment response during maintenance therapy. The model considered two schedules of infusions for infliximab patients, every 8 weeks or every 6 weeks, based on published reports^16,22.

Staff compensation

Average hourly salary values (wages) and the proportion of total compensation for benefits were based on the most recent estimates published by the Bureau of Labor Statistics (BLS)¹⁷ at the time of this study for hospital employees. Salaries (wages plus benefits) were calculated as suggested by BLS for healthcare providers²³.

Supplies

A comprehensive list of disposable supplies and the number of units required for reconstitution and administration of infusion therapy was developed utilizing prescribing information and expert opinion. Data on supply costs were derived from searches of medical supply websites or obtained from Wolters Kluwer Health’s Medi-Span Price Rx online (member access only)¹⁸. To calculate overall supply costs, the number of each type of supply used was multiplied by the unit cost.

Pharmacy non-labor facility costs

These costs were not specifically related to preparation and administration of infusions, but are required for the general performance and operation of the pharmacy, and included facility, utility, and equipment costs. Estimated non-labor facility costs included in the model were storage, space rental, equipment, and other drug information resources. In the pharmacy, these costs were estimated per dose of infusion therapy grounded in findings from a published study¹².

Pharmacy labor facility costs

A published estimate of labor overhead in the pharmacy measured personnel time required to maintain the pharmacy, but not related to direct patient care (e.g. managing annual physical inventory and current inventory levels, insurance management, coding, reimbursement tasks, and disposal of waste materials)¹².

Infusion center (facility) non-labor costs

In the infusion center, non-labor overhead incorporated costs were related to rent, utilities, facilities (per square foot), and capital equipment. The base-case value for Class A medical space (e.g., rent, utilities, and maintenance) was obtained from DesHarnais Castel et al.²⁴. For the base-case, a survey of community-based infusion center administrators informed the size of the infusion center (estimated at 2,500 square feet)¹¹.

Prices for infusion equipment such as infusion chairs, pumps, and electronic blood pressure monitors were estimated based on an internet search and multiplied by the proportion of patients seeking infusion therapy among all patients at the site expected to use each resource. Capital depreciation was taken over 5 years in accordance with IRS guidelines²⁵. While other categories of non-labor overhead may exist, no published estimates were available and, as such, no other categories were included in the model.

Infusion center (facility) non-labor costs

Costs of the infusion center labor overheads were estimated at 2.33-times the non-labor overhead reported in Brixner et al.¹².

Laboratory

Although the cost calculation for hospital infusion administration does not include estimates of the annual cost of diagnostic tests performed by the laboratory, these costs were included in this analysis. The choice and frequency of laboratory testing was drawn from the product package inserts^15,16, and supplemented with clinical expert opinion. Fee schedules published by the Centers for Medicare and Medicaid Services^26,27 informed the costs for routine laboratory testing.

Allocated overhead

Apart from direct overhead costs associated with the two relevant facilities (pharmacy and infusion center), hospitals incur added overhead costs that are spread across each business unit. These allocated overheads are costs administrators must consider and account for services unrelated to providing infusion services (e.g. maintenance and repairs, medical records, laundry). Previously published studies report overhead costs for US hospitals in the in the range of 22.8–46.1%^28,29. The allocated overhead for the base-case analysis in this model was assumed to be 30% and was applied to all cost categories with the exception of infusion drug. Costs modeled are presented with and without the allocated overheads (Table 2).

Table 2. Base-case results—annual cost of infusion (per patient).

	Infliximab		Vedolizumab
Cost of infusion, CD and UC combined	Year 1	Year 2+	Year 1	Year 2+
Population
Average # infusions annually	8.1	7.6	8.0	7.0
% of infusions with dose escalation (IFX: 10 mg/kg)	5.4%	48.4%	n/a	n/a
Infusion drug
Cost of infusion drug	$34,994	$46,290	$38,552	$33,733
Labor
Infusion services
Labor, pharmacy	$176	$165	$173	$152
Labor, infusion administration	$996	$936	$545	$477
Labor, infusion reactions (stop IV; monitoring)	$458	$431	$239	$209
Facility
Administration and management, Pharmacy	$140	$131	$138	$121
Administration and management, Facility	$407	$382	$400	$350
Allocated overhead (30.0%)	$653	$614	$449	$393
Non-labor
Supplies
Supplies, pharmacy	$115	$130	$68	$60
Supplies, pre-medication	$13	$12	$3	$3
Supplies, infusion (IV set, IV bag, etc.)	$131	$123	$89	$78
Supplies, infusion reactions (medication)	$0	$0	$0	$0
Facility
Non-labor, Pharmacy	$53	$50	$52	$45
Non-labor, Facility	$175	$164	$172	$150
Allocated overhead (30.0%)	$146	$144	$115	$101
Laboratory
Laboratory tests (Hep A/B, CRP, CBC, TB test, etc.)	$250	$250	$250	$250
Allocated overhead (30.0%)	$75	$75	$75	$75
Sub-totals, by category
Infusion drug	$34,994	$46,290	$38,552	$33,733
Labor (infusion center, facility)	$2,177	$2,045	$1,495	$1,309
Non-labor costs (supplies, other non-labor)	$487	$479	$384	$336
Laboratory	$250	$250	$250	$250
Allocated overhead (30.0%)*	$874	$833	$639	$569
Annual total (per patient)	$38,782	$49,897	$41,320	$36,197

* Only applies to costs for the following categories: Labor, Non-Labor, Laboratory.

CBC: complete blood count; CD: Crohn’s disease; CRP: c-reactive protein; Hep A/B: hepatitis A/B; IFX: infliximab; IV: intravenous; TB: tuberculous; UC: ulcerative colitis.

Exploratory analysis of costs to the patient

Patients with IBD (UC and CD) may incur substantial individual and societal costs (in addition to high medical costs) that should be considered by the treating clinician in management of the diseases^2,8,30. To examine the impact of infusion therapy on these individual and societal costs not borne directly by hospitals, the value of annual out-of-pocket costs for transportation and indirect costs of lost time from work (assumed to be up to half a day of missed work) were estimated by performing an exploratory analysis. Estimates of the proportion of employed CD and UC patients were obtained from clinical trials^31,32. The indirect costs of infusion therapy incurred by patients were calculated using the national average wage and benefits estimate for all civilian workers from BLS²³ to estimate the value of lost compensation due to time spent receiving infusion therapy. Finally, literature-based estimates informed the costs and average travel time of patients’ transportation to/from the infusion center³³. These input assumptions are described in Table 1.

Results

Infliximab

The model calculated costs for patients initiating therapy (Year 1) or on a stable maintenance dose of infliximab (Year 2+). In Year 1, all UC and CD patients were assumed to receive a dose of 5 mg/kg¹⁶. In the first year of therapy, it was assumed a portion of treated UC and CD patients may lose response to infliximab ∼38 weeks, as observed in the ACCENT trial, resulting in increasing their infliximab dose to 10 mg/kg, or undergoing more frequent infusions (changing their infusion schedule from every 8 weeks to every 6 weeks), or both escalating dose and having a more frequent schedule of infusions³⁴. For patients on maintenance therapy with infliximab (Year 2+), the proportions of patients receiving a stable regimen of 5 mg or 10 mg/kg and proportions of patients receiving infusions on a 6- or 8-week schedule were informed by publications reporting on real-world treatment patterns with infliximab therapy for IBD. The base-case input assumptions for Year 1 and 2+ infliximab dose and schedule are listed in Table 1. The following section summarizes the annual costs for infliximab-treated patients in Year 1 and Year 2+.

Total costs

The total annual modeled per patient costs of the first year of infusion therapy with infliximab was $38,782, and rose to $49,897 in Year 2+. These costs were categorized to infusion drug, labor, non-labor, laboratory, and allocated overhead. Infusion drug costs represented the greatest share of total costs, followed by labor costs. Results of the base-case analysis by each cost category for infliximab are presented in Table 2.

Infusion drug costs

Infusion drug cost accounted for the greatest percentage of overall infusion therapy costs, $34,994 (90.2%) for infliximab-treated patients in year 1, and $46,290 (92.8%) in Year 2+.

Labor costs

The annual cost of nursing and pharmacy labor to administer infusion therapy before application of the 30% allocated overhead was $2,177 for infliximab in Year 1, and $2,045 for maintenance therapy in subsequent years. With the inclusion of allocated overhead, the costs increased to $2,830 in Year 1, and $2,659 in Year 2+. Labor comprised the greatest proportion of non-drug costs (56% of total), followed by allocated overhead (23% of total), non-labor costs (14% of total), and laboratory costs (7% of total). Table 2 presents the breakdown of individual cost components of this category.

Non-labor and overhead costs

On a per dose basis, non-labor facility and overhead costs were assumed to be the same for all products modeled, and varied based on the number of annual infusions. For infliximab-treated patients, the cost of infusion supplies was $259 and $265 in Year 1 and Year 2+, respectively (without allocated overhead). Pharmacy non-labor cost was $53 and $50, while the cost of infusion center facility non-labor was $175 and $164 in Year 1 and Year 2+, respectively, before application of allocated overhead. With application of 30% allocated overhead, total non-labor costs were ∼ $633 in Year 1 and $623 in Year 2+.

Vedolizumab

All assumptions for the dosing of vedolizumab as initial therapy (Year 1) or maintenance therapy (Year 2+) were drawn from the product package insert¹⁵. Model results for vedolizumab are summarized below.

Total costs

The total annual modeled per patient costs of Year 1 of infusion therapy with vedolizumab were $41,320, and $36,197 in Year 2+. Results of the base-case analysis by each cost category for vedolizumab are presented in Table 2.

Infusion drug costs

For vedolizumab, cost of infusion drug accounted for the greatest percentage of total costs, $38,552 (93.3%) in Year 1, and $33,733 (93.2%) for maintenance therapy in subsequent years (Year 2+).

Labor costs

Labor costs prior to the application of the 30% allocated overhead were $1,495 for vedolizumab in Year 1, and $1,309 in Year 2+. Including allocated overhead, costs increased to $1,944 in Year 1 and $1,702 in Year 2+. As with infliximab infusions, labor made up the greatest proportion of non-drug costs (53% of total), followed by allocated overhead (23% of total), non-labor costs (14% of total), and laboratory (10% of total). Table 2 presents the components of individual costs in this category.

Non-labor and overhead costs

For vedolizumab-treated patients, the cost of infusion supplies without allocated overhead ranged from $160 in Year 1 to $141 in Years 2+. Pharmacy non-labor was $52 in Year 1 and $45 in Year 2, and the cost of facility non-labor was $172 in Year 1 and $150 in Year 2+, prior to the application of allocated overhead. With application of the 30% allocated overhead, the total non-labor costs were ∼ $499 in Year 1 and $437 in Year 2+.

Laboratory costs

Patients prescribed infusion therapy with a biologic agent require periodic laboratory monitoring, included in preparation for both infliximab and vedolizumab therapy. Tests considered included a complete blood count (CBC), comprehensive metabolic panel (CMP), C-reactive protein (CRP), and erythrocyte sedimentation rate (ESR), as well as the tuberculin skin test or a TB quantiferon lab test every 1–2 years, and hepatitis A/B antibody serologies prior to the start of biologic treatment. The annual cost of laboratory testing was similar across both treatments, costing $250 in both years modeled ($325 with inclusion of allocated overhead).

Sensitivity analysis

One-way sensitivity analyses were conducted with allocated overhead costs (20–40%), wages (±20%), drug acquisition cost (±20%), and, for infliximab only, reduction in infusion time to 1 h. Overall, the model parameter most sensitive to change was drug acquisition price. The infliximab scenario was most sensitive to drug acquisition, modifying annual costs by −$7,000 to $9,256, followed by staff wages (−$425 to $423), and total infusion time (−$571 to −$535). The vedolizumab scenario was most sensitive to changes in drug acquisition costs (−$7,710 to $7,710), followed by staff wages (−$250 to $249) and allocated overhead (−$213 to $213). Results of sensitivity analyses are presented in Table 3.

Table 3. Sensitivity analysis.

	Annual per patient cost*				Difference in annual per patient cost*
	Infliximab		Vedolizumab		Infliximab		Vedolizumab
	Year 1	Year 2+	Year 1	Year 2+	Year 1	Year 2+	Year 1	Year 2+
Base-case	$38,782	$49,897	$41,321	$36,196
20% lower wages	$38,357	$49,498	$41,071	$35,978	($425)	($399)	($250)	($218)
20% higher wages	$39,205	$50,294	$41,570	$36,415	$423	$397	$249	$219
20% lower drug acquisition	$31,782	$40,638	$33,611	$29,449	($7,000)	($9,259)	($7,710)	($6,747)
20% higher drug acquisition	$45,780	$59,153	$49,031	$42,943	$6,998	$9,256	$7,710	$6,747
Infliximab infusion time, reduced to 1 h	$38,211	$49,362	N/A	N/A	($571)	($535)
10% lower allocated overhead (20%)	$38,490	$49,619	$41,108	$36,007	($292)	($278)	($213)	($189)
10% higher allocated overhead (40%)	$39,072	$50,174	$41,534	$36,386	$290	$277	$213	$190

* Excludes patient costs. Note: Parenthetical values are negative.

Exploratory analysis of costs to the patient

The calculated cost of transportation to/from the infusion center varied based on the number of infusions over the year. The annual cost of transportation to and from infusion centers is estimated to cost each patient between $108–$125. For patients who were employed, the annual value of lost wages associated with time spent at the infusion center ranged from $842–$1,197 (not shown).

Discussion

Providing infusion services is associated with substantial costs that may impact the bottom financial line of hospitals, as demonstrated by this economic analysis. In this study, the cost of infusion drug was the primary driver of infusion therapy costs with the administration of both infliximab and vedolizumab for the management of moderate-to-severe IBD, representing 90–93% of total costs. However, personnel, supplies, pharmacy/facility non-labor, and allocated overhead costs were not inconsequential, and accounted for $2,400–$3,800 of care costs per patient annually.

While administration costs only make up a small proportion of the total cost to infuse biologic drugs, when presented with a large patient volume, these administrative costs can quickly accumulate into substantial sums. In totality, these costs can cause significant financial implications for the hospital, and may be further exacerbated if adequate reimbursements cannot be obtained from payers. In the last few years, some institutions have stopped providing infusion therapy or looked to find alternative means of delivering infusion care to patients at more cost-effective sites of care (e.g. office infusions, home infusion, etc.)³⁵. In addition, some institutions have begun operating their own specialty pharmacies and using the data they generate to assess which infused therapies are cost-effective and can be delivered efficiently to patients³⁶.

There is currently a lack of published data that examines the costs of infusion care delivery to IBD patients. With the availability of both infused and self-administered biologic therapies to treat IBD patients, there is a need for hospital stakeholders to be able to assess the total costs and trade-offs associated with the alternative means of providing biologic therapy. This model is a first step in developing a general framework to analyze costs associated with infusion therapy for IBD. This analysis highlights the value to hospital administrators of assessing their infusion patient mix by disease or indication—based on whether other effective drugs are available in different dosage forms (subcutaneous) that can free up infusion center facilities and personnel to provide care to more patients in therapy areas for which there are not alternatives to infusions. While the use of estimates from national and public sources have helped create a generalizable model, the hospital infusion cost model presented here can easily be modified to calculate local infusion administration costs based on values provided by an individual hospital. Furthermore, once the model scenario is individualized, stakeholders evaluating the model results may be better able to perform meaningful hypothetical scenarios to evaluate the impact of adjusting services for administrative decision-making.

Despite the strengths of this model, it is also important to note that the decision to treat patients with a specific biologic agent should not be solely influenced by cost analyses of therapies.

Our model does not take into account various important factors that impact treatment decision-making, including disease characteristics/phenotype (e.g. presence of small bowel disease and perianal fistulas) and patient preference (e.g. severe needle-phobia and convenience of self-injectables). All factors influencing clinician decisions must be taken into account, including but not limited to cost, to provide the best possible clinical outcomes for patients.

The base-case analysis has several limitations. A major limitation is reliance on published estimates, many of which are dated, to inform the base-case. There were several sources that could be used to inform current treatment practice with infliximab, including rates of dose escalation with infliximab^20–22. Unfortunately, at the time of this study, no real world data was published for treatment with vedolizumab; thus, the product package insert had to inform all assumptions about treatment with vedolizumab¹⁵. Another limitation is that the model obtained base-case estimates for infusion drug and infusion supplies from publicly available sources; many hospitals may obtain volume pricing and/or drug discounts that are not reflected in the base-case model scenario. Next, the base-case scenario assumes a small hospital infusion center, treating an average of 100 patients per week. A larger infusion center serving more patients weekly may experience some economies of scale associated with greater patient volume and more efficient use of capital equipment (e.g. infusion chairs, monitors), yielding lower costs per infusion, since associated overhead costs would be distributed among more individual patients. A further limitation is that costs to the patient (e.g. transportation costs and lost wages) were not included in the base-case estimate of infusion administration costs given the model adopts the hospital perspective. However, we briefly present these costs in an exploratory analysis to highlight their importance in contributing to the overall cost burden of infused products. Lastly, neither of the studies from which time estimates were derived included extensive documentation about the role of non-health personnel, suggesting that the model may under-estimate total hospital costs. Recognizing this limitation, the model developers plan to undertake a subsequent study to generate real world data that might be used to populate the model with estimates representative of current treatment practices and institutional costs.

Conclusions

Costs incurred by a hospital-based infusion center in administering any infusion drug may have a substantial financial impact to an institution. In this study, drug acquisition was the single greatest driver of costs associated with infusion based therapy for IBD patients; however, personnel, supplies, pharmacy/facility non-labor, and allocated overhead costs represent substantial costs for an institution. As the reimbursement landscape changes, tools for evaluating the costs for the provision of infusion therapy may help hospital administrators make informed choices and weigh the trade-offs associated with providing infusion services for IBD patients.

Transparency

Acknowledgment

No assistance in the preparation of this article is to be declared.

Declaration of funding

Full funding for this research was provided by AbbVie, Inc. AbbVie, Inc. was involved in all stages of the study research and manuscript preparation.

Declaration of financial/other interests

AA is an employee of the University of Washington, Harborview Medical Center, and has served as a consultant for AbbVie, Inc., but did not receive research funding or consulting fees for this project. KO was an employee of ICON plc, Health Economics. MF is an employee of ICON plc, Health Economics. JC was an employee of AbbVie, Inc., and owns stocks and shares in AbbVie, Inc. AW is an employee of AbbVie, Inc., and owns stocks and shares in AbbVie, Inc. JME peer reviewers on this manuscript have no relevant financial or other relationships to disclose.