Advertisement

Costs of Providing Infusion Therapy for Rheumatoid Arthritis in a Hospital-based Infusion Center Setting

Open AccessPublished:July 14, 2017DOI:https://doi.org/10.1016/j.clinthera.2017.06.007

      Abstract

      Purpose

      Many hospital-based infusion centers treat patients with rheumatoid arthritis (RA) with intravenous biologic agents, yet may have a limited understanding of the overall costs of infusion in this setting. The purposes of this study were to conduct a microcosting analysis from a hospital perspective and to develop a model using an activity-based costing approach for estimating costs associated with the provision of hospital-based infusion services (preparation, administration, and follow-up) in the United States for maintenance treatment of moderate to severe RA.

      Methods

      A spreadsheet-based model was developed. Inputs included hourly wages, time spent providing care, supply/overhead costs, laboratory testing, infusion center size, and practice pattern information. Base-case values were derived from data from surveys, published studies, standard cost sources, and expert opinion. Costs are presented in year-2017 US dollars. The base case modeled a hospital infusion center serving patients with RA treated with abatacept, tocilizumab, infliximab, or rituximab.

      Findings

      Estimated overall costs of infusions per patient per year were $36,663 (rituximab), $36,821 (tocilizumab), $44,973 (infliximab), and $46,532 (abatacept). Of all therapies, the biologic agents represented the greatest share of overall costs, ranging from 87% to $91% of overall costs per year. Excluding infusion drug costs, labor accounted for 53% to 57% of infusion costs.

      Implications

      Biologic agents represented the highest single cost associated with RA infusion care; however, personnel, supplies, and overhead costs also contributed substantially to overall costs (8%–16%). This model may provide a helpful and adaptable framework for use by hospitals in informing decision making about services offered and their associated financial implications.

      Key words

      Introduction

      Rheumatoid arthritis (RA) is an autoimmune inflammatory disease predominantly affecting the joints.
      • Kent P.D.
      • Matteson E.L.
      Clinical features and differential diagnosis.
      It tends to be symmetrical, and severity can vary, with some people having only occasional flares and others having constant symptoms, including fatigue and fever. Serious joint damage and disability can result from moderate to severe disease.
      • Singh J.A.
      • Furst D.E.
      • Bharat A.
      • et al.
      2012 update of the 2008 American College of Rheumatology recommendations for the use of disease-modifying antirheumatic drugs and biologic agents in the treatment of rheumatoid arthritis.
      • Smolen J.S.
      • Landewe R.
      • Breedveld F.C.
      • et al.
      EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2013 update.
      In addition to joint deterioration, RA has been associated with significant systemic comorbidities, including increased prevalences of cardiovascular disease, infections, and malignancies, as well as a higher mortality risk, compared with those in the general population.
      • Dougados M.
      • Soubrier M.
      • Antunez A.
      • et al.
      Prevalence of comorbidities in rheumatoid arthritis and evaluation of their monitoring: results of an international, cross-sectional study (COMORA).
      In the United States, the estimated prevalence of RA in the general population ranges from 0.5% to 1%,
      • Lawrence R.C.
      • Helmick C.G.
      • Arnett F.C.
      • et al.
      Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States.
      with a higher rate of disease observed in women.
      • Kent P.D.
      • Matteson E.L.
      Clinical features and differential diagnosis.
      • Lawrence R.C.
      • Helmick C.G.
      • Arnett F.C.
      • et al.
      Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States.
      The majority of affected individuals are older adults; 67 years is the mean age of patients with RA.
      • Lawrence R.C.
      • Helmick C.G.
      • Arnett F.C.
      • et al.
      Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States.
      With significant growth projected in the US population aged >60 years, it is likely that rates of RA-related functional impairment, disability, morbidity, and mortality will increase.
      • Lawrence R.C.
      • Helmick C.G.
      • Arnett F.C.
      • et al.
      Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States.
      • Helmick C.G.
      • Lawrence R.C.
      • Pollard R.A.
      • et al.
      Arthritis and other rheumatic conditions: who is affected now, who will be affected later? National Arthritis Data Workgroup.
      Treatment guidelines from the American College of Rheumatology
      • Singh J.A.
      • Furst D.E.
      • Bharat A.
      • et al.
      2012 update of the 2008 American College of Rheumatology recommendations for the use of disease-modifying antirheumatic drugs and biologic agents in the treatment of rheumatoid arthritis.
      and the European League Against Rheumatism
      • Smolen J.S.
      • Landewe R.
      • Breedveld F.C.
      • et al.
      EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2013 update.
      emphasize that RA therapy should be commenced early to stem the progression of disease, prevent irreversible joint damage, and halt further functional decline. Recommended treatments are targeted toward achieving clinical remission or low disease activity. Patients become eligible for treatment with biologic disease-modifying antirheumatic drugs (DMARDs) if they continue to experience moderate to high disease activity following nonresponse to conventional DMARD regimens.
      • Singh J.A.
      • Furst D.E.
      • Bharat A.
      • et al.
      2012 update of the 2008 American College of Rheumatology recommendations for the use of disease-modifying antirheumatic drugs and biologic agents in the treatment of rheumatoid arthritis.
      • Smolen J.S.
      • Landewe R.
      • Breedveld F.C.
      • et al.
      EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2013 update.
      Several biologic agents have been approved for use in treating moderate to severe RA; some are administered orally (tofacitinib) or are self-injected (eg, adalimumab and etanercept), while others require monitored infusion in a hospital or infusion center setting. In the United States, the currently marketed biologic DMARDs that require monitored infusion include abatacept, infliximab, rituximab, golimumab, and tocilizumab.
      Although self-injectables have the potential to reduce the administration costs associated with DMARDs, in the United States there remains a strong financial incentive to direct Medicare patients toward infusion-based products administered at a hospital or outpatient center. Medicare Part B directly reimburses physicians for the cost of infused therapies, with patients eligible for a copayment. Although self-injectables are subsidized by the Medicare Part D program, there exists a coverage gap, whereby patients are responsible for 100% of the drug cost until the "catastrophic" phase of coverage.
      • Ward M.M.
      Medicare reimbursement and the use of biologic agents: incentives, access, the public good, and optimal care.
      In a recent study, Yazdany et al
      • Yazdany J.
      • Dudley R.A.
      • Chen R.
      • et al.
      Coverage for high-cost specialty drugs for rheumatoid arthritis in Medicare Part D.
      found that, in patients with RA and Medicare Part D, the mean out-of-pocket cost of self-injected biologic DMARD therapy was $835/mo.
      However, monitored infusion of biologic DMARDs in the hospital setting encompasses costs beyond drug acquisition. To facilitate a comparison of the costs of RA infusion therapy to those of treatment alternatives, it is important to understand the constellation of relevant costs in addition to drug acquisition. Therefore, the purposes of the present study were to conduct a microcosting analysis from a hospital perspective and to develop a model that accounts for all identifiable costs associated with the provision of hospital-based infusion services (preparation, administration, and follow-up) for maintenance treatment of moderate to severe RA in the United States. As such, the model provides infusion center administrators with a reliable framework and tool for identifying and assessing the multitude of costs related to providing infusion therapy at their facilities.

      Materials and Methods

      A model was developed in Excel (Microsoft Corp, Redmond, Washington) for estimating the costs per year associated with providing monitored infusions of biologic DMARDs to patients for the treatment of moderate to severe RA in a US hospital setting. In this analysis, costs were estimated using an activity-based costing framework approach, which allows for the estimation of direct costs of a specific activity by thorough identification of all resources required for completing that activity in order to calculate the overall cost of care.
      • Brimson J.A.
      Activity Accounting: An Activity-Based Costing Approach.
      • Kaplan R.S.
      • Witkowski M.
      • Abbott M.
      • et al.
      Using time-driven activity-based costing to identify value improvement opportunities in healthcare.
      The model analyses were conducted from a provider perspective in a US hospital setting.
      Base-case model input estimates were based on data from multiple sources. First, individuals at 5 community hospital–based infusion centers completed a survey about their patient volume, patient mix, and infusion center–specific overhead and payroll costs (data on file, Survey of Community-Based US Infusion Center Administrators, 2007). Second, published literature provided estimates on the time required for reconstitution and administration of infusion drugs
      • Brixner D.I.
      • Oderda G.M.
      • Nickman N.A.
      • et al.
      Documentation of chemotherapy infusion preparation costs in academic- and community-based oncology practices.
      • Pierce C.A.
      • Baker J.J.
      A nursing process model: quantifying infusion therapy resource consumption.
      and rates of infusion reactions.
      Actemra (tocilizumab) [package insert].
      • Greenwald M.W.
      • Shergy W.J.
      • Kaine J.L.
      • et al.
      Evaluation of the tolerability of rituximab in combination with a tumor necrosis factor inhibitor and methotrexate in patients with active rheumatoid arthritis: results from a randomized controlled trial.
      • Schiff M.
      • Keiserman M.
      • Codding C.
      • et al.
      Efficacy and tolerability of abatacept or infliximab vs placebo in ATTEST: a phase III, multi-centre, randomised, double-blind, placebo-controlled study in patients with rheumatoid arthritis and an inadequate response to methotrexate.
      Third, product package inserts informed most of the assumptions around the administration, dosing, and schedules of monitored infusions.
      Actemra (tocilizumab) [package insert].
      Orencia (abatacept) [package insert].
      Rituxan (rituximab) [package insert].
      Remicade (infliximab) [package insert].
      Fourth, standard costing sources were used for medical staff wages
      US Bureau of Labor Statistics
      US Bureau of Labor Statistics
      and medications,
      Wolters Kluwer Health
      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 the infusion) and to the infusion center itself (for administration of the infusion and management of the patient). Costs obtained from the published literature were adjusted to year-2017 US dollars using the overall Medical Care component of the Consumer Price Index.
      US Bureau of Labor Statistics
      An overview of the model structure is provided in Figure 1. A complete listing of all base-case model estimates (time, costs, facility characteristics, and practice patterns) is displayed in Table I. An overview of data sources selected for informing the base-case estimate is provided subsequently.
      Figure 1
      Figure 1Conceptual framework for the hospital infusion cost mode. RA = rheumatoid arthritis.
      Table IBase-case model inputs.
      ItemBase-Case EstimateSource/Notes
      Infusion drug
       Maintenance infusions administered per yearAbatacept, 14; infliximab, 7; rituximab, 4; tocilizumab, 14Sourced from product package inserts.
      Actemra (tocilizumab) [package insert].
      Orencia (abatacept) [package insert].
      Rituxan (rituximab) [package insert].
      Remicade (infliximab) [package insert].
      Schedule of maintenance infusions assumed is every 4 wk for abatacept and tocilizumab; every 8 wk for infliximab; infusions given 2 wk apart every 24 wk (4 total) for rituximab.
       Drug cost per infusionAbatacept, $2961.09; infliximab, $5839.10; rituximab, $8352.20; tocilizumab, $2276.95Model assumes a stable maintenance dose over a 1-year period. For products with weight-based dosing, the model assumes an average adult weight of 75 kg (all but rituximab).
      Actemra (tocilizumab) [package insert].
      Orencia (abatacept) [package insert].
      Rituxan (rituximab) [package insert].
      Remicade (infliximab) [package insert].
      Infliximab and tocilizumab package inserts indicate an appropriate range of doses; a patient׳s actual maintenance dose is based on clinical response. A mean infliximab dose of 5.5 mg/kg was based on findings from the US CORRONA registry study.
      • Greenberg J.D.
      • Reed G.
      • Decktor D.
      • et al.
      A comparative effectiveness study of adalimumab, etanercept and infliximab in biologically naive and switched rheumatoid arthritis patients: results from the US CORRONA registry.
      The tocilizumab package insert
      Actemra (tocilizumab) [package insert].
      indicates a range of acceptable maintenance doses of 4-8 mg/kg; the model assumes a median dose (6 mg/kg). WAC cost of each drug obtained from Medi-Span Price Rx.
      Wolters Kluwer Health
      Abatacept, 750 mg/infusion, 3 U ($987.03/U); infliximab, 413 mg/infusion, 5 U ($1167.82/U); rituximab, 1000 mg/infusion, 2 U($4176.10/U); tocilizumab, 420 mg/infusion, 2 U ($948.73/U) + 1 U ($379.49).
       Cost of coadministered drug, per infusionAbatacept, $38.94; infliximab, $77.87; rituximab, $136.28; tocilizumab, $0Use of MTX and folic acid in patients receiving abatacept, infliximab, or rituximab is drawn from the package inserts
      Orencia (abatacept) [package insert].
      Rituxan (rituximab) [package insert].
      Remicade (infliximab) [package insert].
      and current guideline recommendations.
      • Smolen J.S.
      • Landewe R.
      • Breedveld F.C.
      • et al.
      EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2013 update.
      • Cipriani P.
      • Ruscitti P.
      • Carubbi F.
      • et al.
      Methotrexate in Rheumatoid Arthritis: Optimizing Therapy Among Different Formulations. Current and Emerging Paradigms.
      • Visser K.
      • Katchamart W.
      • Loza E.
      • et al.
      Multinational evidence-based recommendations for the use of methotrexate in rheumatic disorders with a focus on rheumatoid arthritis: integrating systematic literature research and expert opinion of a broad international panel of rheumatologists in the 3E Initiative.
      WAC cost of MTX and folic acid obtained from Medi-Span Price Rx.
      Wolters Kluwer Health
      Methotrexate (MTX) 10 mg/wk; cost per year, $520; folic acid, 5 mg/wk, cost per year, $25.12.
      Costs of infusion services (nonlabor—supplies)
      Pharmacy, per infusionAbatacept, $9.83; infliximab, $13.19; rituximab, $21.59; tocilizumab, $9.83Unit costs drawn from online sources: gloves, $0.26; 10 mL sterile water, $0.87; syringe with 21-gauge or smaller needle, $0.76; alcohol swab, $0.05; 0.9% sodium chloride injection, 250 mL infusion bottle/bag, $1.55; impervious gown, $1.57; mask, $1.41; for all infusions, assume 1 U each of gloves, mask, gown and sodium chloride bottle; units of sterile water, syringes and alcohol swabs vary by product (3 U for abatacept and tocilizumab, 5 U for infliximab, 10 U for rituximab).
       Premedication, per infusionAcetaminophen, $0.02; diphenhydramine, $0.02; hydrocortisone, $6.93; methylprednisolone, $19.41Only applies to patients receiving infliximab or rituximab. Use of premedication is assumed per the product package inserts.
      Remicade (infliximab) [package insert].
      Mean WAC cost and WAC cost of premedications obtained from Medi-Span Price-Rx.
      Wolters Kluwer Health
      Infliximab: assumes use diphenhydramine (25 mg PO), acetaminophen (1000 mg PO) and hydrocortisone (75 mg IV) in infliximab patients. Rituximab: assumes use of diphenhydramine (25 mg PO), acetaminophen (1000 mg PO), and methylprednisolone (100 mg IV).
       Infusion, per infusionAbatacept, $16.64; infliximab, $16.64; rituximab, $9.85; tocilizumab, $9.85Unit costs drawn from online sources: IV start kit, nonlatex, $2.60; basic IV set, nonlatex, $5.90; IV administration set with 1.2 microfilter, $6.79; 1000 mL IV bag, $15 per case of 12; mean WAC cost and WAC cost of premedications obtained from Medi-Span Price-Rx
      Wolters Kluwer Health
      ; costs adjusted to year-2017 US $ from year-2014 US $ based on the medical care component of the CPI (inflation factor = 1.0792) ($15/12 * 1.0792 = $1.35).
      US Bureau of Labor Statistics
      All infusions were assumed to require an IV start kit, basic IV set and 0.9% sodium chloride bag (1 U each); infusions of abatacept and infliximab also include the use of a 1.2-micron filter per the instructions in the product package inserts.
      Orencia (abatacept) [package insert].
      Remicade (infliximab) [package insert].
       Infusion reaction, per eventHeadache, $0.01; dyspnea, $0.08; hypertension, $1.09; hypotension, $0.08; nausea, $1.15; urticaria, $1.15Only some infusion reactions require additional treatment. Choice of treatment was informed by Lenz.
      • Lenz H.J.
      Management and preparedness for infusion and hypersensitivity reactions.
      WAC cost of drug treatment was obtained from Medi-Span Price Rx.
      Wolters Kluwer Health
      Treatments for infusion reactions included: headache, acetaminophen 1000 mg PO; dyspnea, epinephrine 0.5 mg (1:1000 IV push); hypertension, captopril 25 mg PO; hypotension, epinephrine 0.5 mg (1:1000 IV push); nausea, diphenhydramine (50 mg IV push); and urticaria, diphenhydramine (50 mg IV push)
      Infusion services
       Labor
        Wages and benefits for clinical staffMean hourly wage, including benefits, assumed for hospital staff: physician, $147.48; nurse–practitioner, $76.44; nurse, $52.74; pharmacist, $87.87; pharmacy tech, $23.51Average hourly wage estimates
      US Bureau of Labor Statistics
      and benefits rate (34.20%) for all hospital workers
      US Bureau of Labor Statistics
      were obtained from BLS for Q2 and Q4, 2017; staff roles correspond to the following categories in the 2017 BLS wage file: internists, general (29-1063), nurse–practitioners (29-1171), pharmacists (29-1051), pharmacy technicians (29-2052), and registered nurses (29-1141).
        Pharmacy (per infusion)Pharmacist time, 12.43 min; pharmacy tech time, 11.55 minTime assumed for reconstitution of infusion drug (constant across all products). Estimate drawn from Brixner et al
      • Brixner D.I.
      • Oderda G.M.
      • Nickman N.A.
      • et al.
      Documentation of chemotherapy infusion preparation costs in academic- and community-based oncology practices.
      and adjusted assuming 5% of hospital pharmacy time during 8-h day would be spent on RA infusion-related activities.
        Infusion administration (per infusion)49.8 min for setup, documentation, discontinuation; per infusion (all products); infusion of RA drug: abatacept, 30 min; infliximab, 120 min; rituximab, 195 min; tocilizumab, 60 minThe model assumes 49.8 min/infusion for tasks informed by the process model suggested by Pierce and Baker.
      • Pierce C.A.
      • Baker J.J.
      A nursing process model: quantifying infusion therapy resource consumption.
      Tasks included are: preservice (11 min); initial stage (8.1 min); treatment admin (stage 1) (20.5 min); completion stage (3.4 min); postservice (6.8 min).
      Product package inserts informed the time estimates for the duration of each product-specific infusion
      Actemra (tocilizumab) [package insert].
      Orencia (abatacept) [package insert].
      Rituxan (rituximab) [package insert].
      Remicade (infliximab) [package insert].
      ; the assumed mix of staff completing these tasks: physician (4.5%), nurse–practitioner (4.5%), and nurse (91%)
        Management of infusion reactions (per infusion)Grade 1 reactions, 10 min; grade 2 reactions, 15 min + infusion flow rate is assumed to be reduced by halfAssumption informed by Lenz.
      • Lenz H.J.
      Management and preparedness for infusion and hypersensitivity reactions.
      Grade 1 infusion reactions are mild and are assumed not to require interruption of the infusion. The model assumed 10 min of additional monitoring time for a patient who experienced a grade 1 reaction. The total time to manage a grade 2 infusion reaction is product specific. The model assumed 15 min of additional time to interrupt the infusion and treat the reaction plus a 50% reduction in the flow rate midway through the infusion. Reduction in infusion flow rate drawn from instructions in the rituximab package insert.
      Rituxan (rituximab) [package insert].
       Infusion reaction rates
        Grade 1Dizziness (abatacept, 0.60%; infliximab, 2.40%; rituximab, 6.00%; tocilizumab, 0%)Infusion reaction rates for abatacept and infliximab drawn from Schiff et al
      • Schiff M.
      • Keiserman M.
      • Codding C.
      • et al.
      Efficacy and tolerability of abatacept or infliximab vs placebo in ATTEST: a phase III, multi-centre, randomised, double-blind, placebo-controlled study in patients with rheumatoid arthritis and an inadequate response to methotrexate.
      ; rituximab infusion reaction rates obtained from Greenwald et al
      • Greenwald M.W.
      • Shergy W.J.
      • Kaine J.L.
      • et al.
      Evaluation of the tolerability of rituximab in combination with a tumor necrosis factor inhibitor and methotrexate in patients with active rheumatoid arthritis: results from a randomized controlled trial.
      ; tocilizumab infusion reaction rates came from the product package insert.
      Actemra (tocilizumab) [package insert].
      Fatigue (abatacept, 0.00%; infliximab, 0%; rituximab, 6.00%; tocilizumab, 0%)
      Flushing (abatacept, 0.60%; infliximab, 3.00%; rituximab, 0%; tocilizumab, 0%)
      General infusion-related reaction (nonspecific) (abatacept, 0%; infliximab, 0%; rituximab, 6.00%; tocilizumab, 0%)
      Headache (abatacept, 1.30%; infliximab, 4.20%; rituximab, 6.00%; tocilizumab, 1.00%)
      Skin reactions, including rash, pruritus (abatacept, 0%; infliximab, 3.00%; rituximab, 12.00%; tocilizumab, 1.00%)
        Grade 2Dyspnea (abatacept, 0%; infliximab, 3.00%; rituximab, 0%; tocilizumab, 0%)Infusion reaction rates for abatacept and infliximab drawn from Schiff et al
      • Schiff M.
      • Keiserman M.
      • Codding C.
      • et al.
      Efficacy and tolerability of abatacept or infliximab vs placebo in ATTEST: a phase III, multi-centre, randomised, double-blind, placebo-controlled study in patients with rheumatoid arthritis and an inadequate response to methotrexate.
      ; rituximab infusion reaction rates obtained from Greenwald et al
      • Greenwald M.W.
      • Shergy W.J.
      • Kaine J.L.
      • et al.
      Evaluation of the tolerability of rituximab in combination with a tumor necrosis factor inhibitor and methotrexate in patients with active rheumatoid arthritis: results from a randomized controlled trial.
      ; and tocilizumab infusion reaction rates came from the product package insert.
      Actemra (tocilizumab) [package insert].
      Hypertension (abatacept, 0%; infliximab, 0%; rituximab, 0%; tocilizumab, 1.00%)
      Hypotension (abatacept, 0%; infliximab, 4.80%; rituximab, 0%; tocilizumab, 0%)
      Nausea (abatacept, 1.90%; infliximab, 4.20%; rituximab, 9.00%; tocilizumab, 0%)
      Urticaria (abatacept, 0%; infliximab, 4.80%; rituximab, 6.00%; tocilizumab, 1.00%)
      Facility: nonlabor
       Facility characteristicsNo. of infusion patients served weekly, 100; size of infusion facility, 2500 sq ft; weeks open per year, 52No. of patients includes all diagnoses; input assumptions informed by an activity-based costing survey to infusion center administrators (2007)
       Years to depreciate capital5 yInternal Revenue Service publication 946
      Internal Revenue Service
       Non-labor, pharmacy (per infusion)$6.86Estimate informed by Brixner et al.
      • Brixner D.I.
      • Oderda G.M.
      • Nickman N.A.
      • et al.
      Documentation of chemotherapy infusion preparation costs in academic- and community-based oncology practices.
      Items included are: storage (refrigerators, cabinets, freezers; $0.20), space rental (patient and nonpatient rooms, utilities, maintenance; $0.72), equipment (biologic tolerability cabinets, ventilation equip, telecommunications and information technology equipment; $0.58), supplies ($2.80), and information resources (drug information resources; $0.03); adjusted from 2003 USD to 2017 USD (inflation factor: 1.5811).
       Nonlabor, facility (per infusion)$22.65Rent, utilities, facilities: estimate from DesHarnais et al,
      • DesHarnais Castel L.
      • Bajwa K.
      • Markle J.P.
      • et al.
      A microcosting analysis of zoledronic acid and pamidronate therapy in patients with metastatic bone disease.
      of $25/sq ft adjusted from 1999 USD to 2017 USD (inflation factor = 1.8746). Adjusted cost per infusion for rent, utilities, facilities was calculated as (rent per sq ft * # sq ft)/(weekly infusion patients * weeks open per year); Estimated equipment cost culled from online sources (2017): infusion pump ($805) infusion pump stand ($242.99); electronic blood pressure machine ($995); infusion chair ($796.77); electronic thermometer ($229). All equipment adjusted (depreciated) cost of equipment per infusion = (equipment cost/years to depreciate capital)/(weekly infusion patients * weeks open per year)
      Facility: labor
       Labor, administration and management, pharmacy (per infusion)$18.20Estimate drawn from Brixner et al
      • Brixner D.I.
      • Oderda G.M.
      • Nickman N.A.
      • et al.
      Documentation of chemotherapy infusion preparation costs in academic- and community-based oncology practices.
      Includes inventory management ($5.47), insurance management ($11.80) and labor for the disposal of waste material ($0.93); adjusted from 2003 USD to 2017 USD (inflation factor: 1.5811).
       Labor, administration and management, facility (per infusion)$52.75Based on Brixner et al.
      • Brixner D.I.
      • Oderda G.M.
      • Nickman N.A.
      • et al.
      Documentation of chemotherapy infusion preparation costs in academic- and community-based oncology practices.
      Estimate assumed labor overhead would be 2.33 times the nonlabor overhead of $22.64 (see the entry for nonlabor, facility [per infusion]).
      Laboratory testing
       Laboratory monitoring (annual)Laboratory test (CPT code); unit cost; frequency of annual testingSchedule of annual laboratory tests was informed by product package inserts.
      Actemra (tocilizumab) [package insert].
      Orencia (abatacept) [package insert].
      Rituxan (rituximab) [package insert].
      Remicade (infliximab) [package insert].
      Rheumatrex (methotrexate) [package insert].
      The unit costs for laboratory testing were obtained from the Centers for Medicare & Medicaid Services 2017 Clinical Laboratory Fee Schedule; the unit cost of TB intradermal test came from the CMS 2017 Physician Fee Schedule.
      Centers for Medicare & Medicaid Services
      Metabolic panel (80048); unit cost, $11.60; frequency of annual testing (abatacept 6.0; infliximab 6.0; rituximab 6.0; tocilizumab 1.0)
      Lipid panel (80061); unit cost, $17.86; frequency of annual testing (abatacept 0.0; infliximab 0.0; rituximab 0.0; tocilizumab 3.0)
      Hepatic function panel (80076); unit cost, $11.21; frequency of annual testing, (abatacept 6.0; infliximab 6.0; rituximab 6.0; tocilizumab 9.0)
      Complete CBC w/auto diff WBC (85025); unit cost, $10.66; frequency of annual testing (abatacept 12.0; infliximab 12.0; rituximab 12.0; tocilizumab 9.0)
      Hepatitis B core antibody total (86704); unit cost, $16.53; frequency of annual testing (abatacept 2.0; infliximab 2.0; rituximab 2.0; tocilizumab 2.0)
      TB intradermal test (86580); unit cost, $9.02; frequency of annual testing (abatacept 1.0; infliximab 1.0; rituximab 1.0; tocilizumab 1.0)
      Hospital system
       Allocated overhead30%Assumption
      Patient costs
       Labor, wages and benefits for patientsMean hourly wage, including benefits, $34.90Mean hourly wage estimate and benefits rate (31.60%) for civilian workers
      • Kent P.D.
      • Matteson E.L.
      Clinical features and differential diagnosis.
      from December 2016 was obtained from the US Bureau of Labor Statistics (BLS).
      US Bureau of Labor Statistics
       Out-of-pocket costs for transportation to/from infusion center (per infusion)$15.40Assumption informed by Luce et al.
      • Luce B.R.
      • Zangwill K.M.
      • Palmer C.S.
      • et al.
      Cost-effectiveness analysis of an intranasal influenza vaccine for the prevention of influenza in healthy children.
      Assumed patients lived ~15 miles from the site of infusion therapy. Transportation included 50% private vehicle mileage and parking ($0.56/mile × 15 miles and $8 parking), 50% public transportation estimated at $6/round trip.
      BPS = US Bureau of Labor Statistics; CBC = complete blood count; CPI = Consumer Price Index; CMS = Centers for Medicare & Medicaid Services; CPT = Current Procedural Terminology; MTX = methylprednisolone; RA = rheumatoid arthritis; TB = tuberculosis; WAC = wholesale acquisition cost; WBC = white blood cells.

       Time Estimates

      The process model detailed in an observational study by Pierce and Baker
      • Pierce C.A.
      • Baker J.J.
      A nursing process model: quantifying infusion therapy resource consumption.
      was used for estimating the time spent by nursing staff on administering an infusion. For each of the modeled infusion therapies, the times spent on preservice and postservice activities were assumed to be the same (~50 minutes), whereas time spent on administration of infusion therapy varied by product and matched specifications for total infusion time in each package insert. The model assumed that infusion services were delivered predominantly by nurses (91%), followed by nurse–practitioners and physicians (4.5% each), as reported by Pierce and Baker.
      • Pierce C.A.
      • Baker J.J.
      A nursing process model: quantifying infusion therapy resource consumption.
      Data from an observational study by Brixner et al
      • Brixner D.I.
      • Oderda G.M.
      • Nickman N.A.
      • et al.
      Documentation of chemotherapy infusion preparation costs in academic- and community-based oncology practices.
      that measured chemotherapy preparation time were used for estimating pharmacist and technician times per infusion, in addition to the overhead costs related to infusion preparation.
      The administration of medication, such as diphenhydramine, prior to infusion therapy was assumed in patients treated with infliximab and rituximab, as outlined in the package insert.
      Actemra (tocilizumab) [package insert].
      Rituxan (rituximab) [package insert].
      Where possible, infusion reaction rates were obtained from the package inserts (eg, tocilizumab
      Actemra (tocilizumab) [package insert].
      ). The package inserts for abatacept, infliximab, and rituximab did not report rates of infusion reactions; thus, data from clinical trial publications
      • Greenwald M.W.
      • Shergy W.J.
      • Kaine J.L.
      • et al.
      Evaluation of the tolerability of rituximab in combination with a tumor necrosis factor inhibitor and methotrexate in patients with active rheumatoid arthritis: results from a randomized controlled trial.
      • Schiff M.
      • Keiserman M.
      • Codding C.
      • et al.
      Efficacy and tolerability of abatacept or infliximab vs placebo in ATTEST: a phase III, multi-centre, randomised, double-blind, placebo-controlled study in patients with rheumatoid arthritis and an inadequate response to methotrexate.
      were used for obtaining estimates of the frequencies of grades 1 and 2 infusion reactions. A published review
      • Lenz H.J.
      Management and preparedness for infusion and hypersensitivity reactions.
      informed time estimates of medical staff who manage hypersensitivity reactions during a patient׳s infusion. The model assumed that grade 1 infusion reactions were mild, with no infusion interruption, but required an extra 10 minutes of nurse monitoring per event. Grade 2 infusion reactions were assumed to require treatment (eg, antihistamines, epinephrine) and an interruption of a single infusion session for 15 minutes. In the base-case scenario, all grade 2 events were assumed to occur midway through a single infusion session, to be resolved following treatment, and patients were expected to resume their infusion at a 50% slower flow rate until the single infusion session was completed. Time estimates of a nurse׳s management of infusion reactions were weighted by frequency of events.

       Cost Estimates

       Infusion Drug

      The model considered maintenance infusion therapy for RA with biologic agents approved for marketing by the US Food and Drug Administration as of January 1, 2015 (ie, abatacept, infliximab, rituximab, or tocilizumab); package inserts informed the dosing and schedules of infusions.
      Actemra (tocilizumab) [package insert].
      Orencia (abatacept) [package insert].
      Rituxan (rituximab) [package insert].
      Remicade (infliximab) [package insert].
      A patient weight of 75 kg was assumed for calculating total drug used for infusions that required weight-based dosing (abatacept, infliximab, and tocilizumab). The tocilizumab and infliximab package inserts stated a range of doses for the treatment of RA and indicated that each patient׳s dose is individualized based on clinical response.
      Actemra (tocilizumab) [package insert].
      Remicade (infliximab) [package insert].
      For infliximab, published estimates from the Corrona registry
      • Greenberg J.D.
      • Reed G.
      • Decktor D.
      • et al.
      A comparative effectiveness study of adalimumab, etanercept and infliximab in biologically naive and switched rheumatoid arthritis patients: results from the US CORRONA registry.
      informed the choice of modeling a maintenance dose of 5.5 mg/kg. Published estimates of dosing of tocilizumab in clinical practice were sparse; thus, the model assumed a dose of 6 mg/kg, which represents the middle of the dosing range specified in the package insert.
      Actemra (tocilizumab) [package insert].
      Per current treatment guidelines, abatacept, infliximab, and rituximab were assumed to be administered in conjunction with methotrexate 10 mg/wk and folic acid 5 mg/wk.
      • Smolen J.S.
      • Landewe R.
      • Breedveld F.C.
      • et al.
      EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2013 update.
      • Cipriani P.
      • Ruscitti P.
      • Carubbi F.
      • et al.
      Methotrexate in Rheumatoid Arthritis: Optimizing Therapy Among Different Formulations. Current and Emerging Paradigms.
      • Visser K.
      • Katchamart W.
      • Loza E.
      • et al.
      Multinational evidence-based recommendations for the use of methotrexate in rheumatic disorders with a focus on rheumatoid arthritis: integrating systematic literature research and expert opinion of a broad international panel of rheumatologists in the 3E Initiative.
      The overall cost of infusions per year was based on the number of maintenance infusions of each product administered in a given year, which varied by product. Patients receiving abatacept or tocilizumab were assumed to receive 14 infusions (ie, every 4 weeks), infliximab-treated patients received 7 infusions (ie, every 8 weeks), and rituximab-treated patients received 4 infusions per year. The model did not consider switching or adding drug treatments or dose escalation; thus, the calculated costs per year of each infusion drug assumed a stable maintenance dose over the course of a year.

       Wages

      Hourly salary values (wages) were based on US national estimates for personnel involved in activities related to providing infusion services in the hospital. The latest estimates available from the Bureau of Labor Statistics
      US Bureau of Labor Statistics
      at the time of this study were used, and the salaries (wages plus benefits) of health care providers were calculated using an amount of 34.2% of total compensation for benefits, as suggested by data from the Bureau of Labor Statistics.
      US Bureau of Labor Statistics

       Supplies

      Prescribing information and expert opinion were used for developing a comprehensive list of disposable supplies and the number of units required for the reconstitution and administration of infusion therapy. Data on supply costs were derived from searches of medical supply websites or obtained from Medi-Span Price Rx on-line (member access only; Wolters Kluwer Health, Alphen aan den Rijn, the Netherlands).
      Wolters Kluwer Health
      Overall supply costs were determined by multiplying the number of each type of supply used, by the unit cost.

       Pharmacy Nonlabor Facility Costs

      Pharmacy nonlabor facility costs included facility, utility, and equipment costs that are not specifically related to the preparation and administration of infusions, but that are necessary for the general performance and operation of the pharmacy. In the model, costs associated with the facility, utility, and equipment are considered sunk costs. Estimated nonlabor facility costs included in the model were the costs of storage (refrigerators, cabinets, freezers), space rental (patient and nonpatient rooms, utilities, maintenance), equipment (biologic tolerability cabinets, ventilation equipment, telecommunications, information technology equipment), and other drug information resources. In the pharmacy, these costs per dose of infusion therapy were estimated based on findings from a published study.
      • Brixner D.I.
      • Oderda G.M.
      • Nickman N.A.
      • et al.
      Documentation of chemotherapy infusion preparation costs in academic- and community-based oncology practices.

       Pharmacy Labor Facility Costs

      Published estimates of labor overhead in the pharmacy and the time required for reconstituting infusion drugs were measured as personnel time required for maintaining the pharmacy but not related to direct patient care. These activities included managing annual physical inventory and current inventory levels, insurance management, coding, reimbursement tasks, and disposal of waste materials.
      • Brixner D.I.
      • Oderda G.M.
      • Nickman N.A.
      • et al.
      Documentation of chemotherapy infusion preparation costs in academic- and community-based oncology practices.

       Infusion Center (Facility) Nonlabor Costs

      In the infusion center, nonlabor overhead included costs related to rent, utilities, facilities (per square foot), and capital equipment. The base-case value for class A medical space, including rent, utilities, and maintenance, was taken from an article by DesHarnais et al.
      • DesHarnais Castel L.
      • Bajwa K.
      • Markle J.P.
      • et al.
      A microcosting analysis of zoledronic acid and pamidronate therapy in patients with metastatic bone disease.
      For the base case, the size of the infusion center was estimated at 2500 sq ft, which was informed by a survey of community-based infusion center administrators (data on file, AbbVie, 2007).
      Prices of infusion equipment such as infusion chairs, pumps, pump stands, electronic blood pressure monitors, and electronic thermometers were estimated based on an Internet search and multiplied by the percentage of patients seeking infusion therapy among all patients at the site expected to use each resource. Capital depreciation was taken over 5 years per the Internal Revenue Service׳s guideline on high-technology medical equipment.
      Internal Revenue Service
      While other categories of nonlabor overhead may exist, there were no published estimates available, and, as such, no other categories were included in the model. The model aimed to represent all categories of nonlabor costs found in the published literature.

       Infusion Center (Facility) Labor Costs

      Costs of the infusion center labor overheads were estimated based on multiplying the infusion center nonlabor overhead by a multiplier (2.33) reported by Brixner et al.
      • Brixner D.I.
      • Oderda G.M.
      • Nickman N.A.
      • et al.
      Documentation of chemotherapy infusion preparation costs in academic- and community-based oncology practices.
      In other words, the labor overhead was 2.33-fold that of the nonlabor overhead.

       Laboratory

      Estimates of the cost per year of diagnostic tests performed by the laboratory were included in this analysis. The choice and frequency of laboratory testing were drawn from the package inserts.
      Actemra (tocilizumab) [package insert].
      Orencia (abatacept) [package insert].
      Rituxan (rituximab) [package insert].
      Remicade (infliximab) [package insert].
      Rheumatrex (methotrexate) [package insert].
      The costs of routine laboratory testing were obtained from fee schedules published by the Centers for Medicare and Medicaid Services.
      Centers for Medicare & Medicaid Services

       Allocated Overhead

      Apart from the direct overhead costs associated with the 2 relevant facilities (pharmacy and infusion center), hospitals incur other overhead costs that are spread across each business unit. These allocated overheads account for services unrelated to providing infusion services (eg, maintenance and repairs, medical records, laundry), but nonetheless are costs from a hospital׳s perspective. Previously published studies have reported overhead costs of US hospitals in the range of 22.8% to 46.1%.
      • Himmelstein D.U.
      • Jun M.
      • Busse R.
      • et al.
      A comparison of hospital administrative costs in eight nations: US costs exceed all others by far.
      • Kalman N.
      • Hammill B.
      • Schulman K.
      • Shah B.
      Hospital overhead costs: the neglected driver of health care spending?.
      The allocated overhead for the base-case analysis in this model was assumed to be 30%. Costs modeled are presented with and without the allocated overheads (Table II). Allocated overhead was applied to all cost categories, with the exception of infusion drug.
      Table IIBase-case results—annual cost of infusion.
      Values have been rounded to the nearest dollar. Total number of maintenance infusions assumed over a 1-year period were: abatacept, 14; infliximab; 7; rituximab, 4; and tocilizumab, 14.
      Data are given as mean cost per patient per year, in year-2017 US $.
      ItemAbataceptInfliximabRituximabTocilizumab
      Infusion drug
       Cost of infusion drug41,45540,87433,40931,877
       Cost of co-administered drug (eg, methotrexate)5455455450
       Allocated overhead (not applied)0000
      Infusion services
       Nonlabor (supplies)
        Pharmacy1389286138
        Premedication049780
        Infusion (IV set, IV bag, etc.)23311639138
        Infusion reactions (medication)0110
       Labor
        Pharmacy30515387305
        Infusion administration1,0811,1509481,488
        Infusion reactions (stop IV; monitoring)1516213623
       Allocated overhead (30.0%)532517412627
      Facility
       Nonlabor
        Pharmacy96482796
        Facility31715991317
       Labor (administration and management)
        Pharmacy25512773255
        Facility739369211739
       Allocated overhead (30.0%)422211121422
      Laboratory
       Tests307307307304
       Allocated overhead (30.0%)92929291
      Subtotals
       Infusion drug42,00041,41933,95431,877
       Labor2,3951,9621,4552,809
       Nonlabor784465322689
       Laboratory307307307304
       Allocated overhead (30.0%)1,0468206251,141
      Total46,53244,97336,66336,821
      low asterisk Values have been rounded to the nearest dollar. Total number of maintenance infusions assumed over a 1-year period were: abatacept, 14; infliximab; 7; rituximab, 4; and tocilizumab, 14.

      Results

      The overall modeled costs per patient per year of infusion therapy with abatacept, infliximab, tocilizumab, and rituximab were $46,532, $44,973, $36,821, and $36,663, respectively. These modeled costs were attributed to the categories of infusion drug, infusion services, facility, laboratory, and allocated overhead. Infusion services costs represented the actual costs of the supplies and personnel wages required for infusing the drug into the patient. Facilities costs encompassed nonlabor and labor overhead for the pharmacy and infusion center necessary for supporting infusion services in the hospital setting. Results of the base-case analysis by each category of costs are presented in Table II.

       Infusion Drug

      The model used for estimating the cost per year of infusion therapy included the cost of the infused RA product and any coadministered drugs (ie, methotrexate, folic acid). Across all products, drug accounted for the greatest share of overall costs of infusion, ranging from 86.6% per year in tocilizumab-treated patients to 92.6% per year in rituximab-treated patients.

       Infusion Services Costs

      The costs of infusion services (pharmacy plus infusion center costs) before the application of the 30% allocated overhead were $1375 (rituximab), $1723 (infliximab), $1772 (abatacept), and $2092 (tocilizumab) per year. With the addition of allocated overhead, these costs ranged from $1787 to $2719 over a 1-year period. Infusion services costs were heavily influenced by patients׳ schedule of infusions over the modeled period. Tocilizumab- and abatacept-treated patients had the most infusions (14) over a 1-year period, resulting in the highest supply and labor costs of all of the RA infusion therapies modeled. The estimated infusion services costs were lowest in rituximab-treated patients due to the assumed schedule of 4 infusions per year. Table II and Figure 2 present the breakdown of individual cost components in this category.
      Figure 2
      Figure 2Costs per year of infusion services, not considering infusion drug cost. Total number of maintenance infusions assumed per year: abatacept (14), infliximab (7), rituximab (4), and tocilizumab (14).
      Infusion services consisted of pharmacy and infusion center staff labor for administering infusions, nonlabor items (ie, supplies), and allocated overhead. Labor accounted for the greatest percentage of infusion services costs (61%–67%), followed by allocated overhead (23%) and nonlabor costs (10%–16%).
      Time required by the pharmacist and pharmacy technician for preparing the infusion drug was valued at ~$22 per dose for each product. All patients were assumed to require nearly 50 minutes of time for preservice, intraservice (ie, infusion setup), and postservice activities; the total estimated cost per infusion was $48. This time estimate was combined with the product-specific time necessary for infusing the treatment, to determine the total time spent by medical staff for infusion services. From preservice through the postservice period, the total time that a center׳s staff spent on administering infusions ranged from 81 minutes (abatacept; $77) to 280 minutes (rituximab; $236) per infusion. Finally, the model considered staff time spent on the management of infusion reactions of grade 1 (10 minutes per event) and grade 2 (15 minutes to stop infusion; 50% flow rate thereafter). Time spent on additional monitoring as a consequence of these events varied by product. The additional cost of staff time spent on monitoring adverse events ranged from $1 to $34 per infusion.
      Supplies associated with reconstituting product (eg, gloves, gown, mask, alcohol swabs, needles, sterile water, infusion bags) and administering the infusion (eg, IV set, tubing, 1.2 micron filter) were estimated to cost from $20 (tocilizumab) to $31 (rituximab) per infusion and accounted for the greatest share of nonlabor, nondrug infusion costs. Cost differences between products were due to the number of vials of sterile water and associated supplies required for reconstituting drug and were dependent on whether a 1.2-micron filter was needed for the administration of the infusion drug.
      Other nonlabor costs included premedication and supplies in patients who experienced infusion reactions. The base-case scenario assumed that infliximab- and rituximab-treated patients received medication prior to infusion, as specified in the package inserts.
      Rituxan (rituximab) [package insert].
      Remicade (infliximab) [package insert].
      Premedication consisted of diphenhydramine and acetaminophen with both infliximab and rituximab, with the addition of hydrocortisone in infliximab-treated patients, and methylprednisone in rituximab-treated patients. Estimated premedication costs per infusion were $7 with infliximab and $19 with rituximab. Supplies for treating infusion reactions (eg, acetaminophen, epinephrine) were of minimal cost, estimated at <$1 per infusion across all infused RA products.

       Nonlabor Facilities and Overhead Costs

      On a per-dose basis, nonlabor facility and overhead costs were assumed to be the same for all products modeled. Before the application of allocated overhead, estimated pharmacy facility costs were $25 per dose, and approximated infusion center facility costs were $75 per dose. In both cases, roughly 70% of these costs were associated with labor, while the remainder of costs represented nonlabor overhead (eg, capital equipment, space, pharmacy infrastructure). Together, the estimated facility costs were $131 per infusion once the 30% allocated overhead was applied.
      As with infusion services costs, patients with more frequent dosing accumulated greater facility costs over the 1-year period. The calculated costs, including allocated overhead, were $523 (rituximab; 4 infusions per year), $914 (infliximab; 7 infusions per year), and $1829 (abatacept and tocilizumab; 14 infusions per year).

       Laboratory Costs

      Patients prescribed infusion therapy with a biologic agent require periodic laboratory monitoring. Tests considered included a complete blood count, tuberculin skin test, hepatitis B antibody, a metabolic panel, and a hepatic function panel. The costs per year of laboratory testing were similar across treatments ($304–$307; $395–399 with allocated overhead).

       Sensitivity Analysis Findings

      Results from sensitivity analyses are presented in Figure 3. One-way sensitivity analyses with allocated overhead costs (20%–40%), wages (±20%), and patient weight (±25 kg) were conducted. Changes in patient weight resulted in the greatest impact on overall costs of infusion by affecting the cost of drugs with weight-based dosing (abatacept, infliximab, and tocilizumab). Varying wage estimates had the second greatest impact, modifying costs per year by $304 to $472.
      Figure 3
      Figure 3One way sensitivity analyses of allocated overhead, wages, and patient weight; change in overall cost per year. Sensitivity analyses values are for allocated overhead, 20%/40%; wages, ±20%; weight, ±25 kg. Bars are centered on base-case costs for individual treatments.

      Discussion

      This analysis demonstrated that providing infusion services is associated with substantial costs that may have a significant impact on a hospital׳s bottom line. Although the cost of administration is a small percentage of the costs associated with the biologic agent being infused, such costs can accumulate to substantial amounts with large patient volume, resulting in significant financial implications for the hospital if adequate reimbursement cannot be obtained.
      Currently, there are 8 approved subcutaneously injected and infused biologic therapies that may be prescribed for the management of moderate to severe RA.
      • Singh J.A.
      • Furst D.E.
      • Bharat A.
      • et al.
      2012 update of the 2008 American College of Rheumatology recommendations for the use of disease-modifying antirheumatic drugs and biologic agents in the treatment of rheumatoid arthritis.
      Owing to the steady increase in spending on this category of drugs, payers have focused on developing strategies (eg, negotiated cost discounts or utilization limits) for helping to stem the rising costs of specialty drug use, which may, in turn, affect reimbursement of the provision of infusion therapy in some therapeutic areas. Step-authorization programs may also be used for ensuring that patients with RA try and fail more cost-effective therapies before being permitted to receive a more expensive therapy.
      • Kalman N.
      • Hammill B.
      • Schulman K.
      • Shah B.
      Hospital overhead costs: the neglected driver of health care spending?.
      Therefore, an understanding of the individual components of infusion center costs is vitally important for the cost-effective provision of infusion services.
      Published cost-effectiveness analyses may inform payer decision making with regard to the selection of preferred infused agents and, subsequently, the choice between directing patients toward a hospital-based infusion center versus subcutaneous injection for RA. Since 2012, several studies have reported lower costs per effectively treated patient with RA with subcutaneous-injected biologic agents relative to infused biologic therapy.
      • Brimson J.A.
      Activity Accounting: An Activity-Based Costing Approach.
      • Kaplan R.S.
      • Witkowski M.
      • Abbott M.
      • et al.
      Using time-driven activity-based costing to identify value improvement opportunities in healthcare.
      • Himmelstein D.U.
      • Jun M.
      • Busse R.
      • et al.
      A comparison of hospital administrative costs in eight nations: US costs exceed all others by far.
      • Smolen J.S.
      • Breedveld F.C.
      • Burmester G.R.
      • et al.
      Treating rheumatoid arthritis to target: 2014 update of the recommendations of an international task force.
      • Bonafede M.M.
      • Gandra S.R.
      • Fox K.M.
      • Wilson K.L.
      Tumor necrosis factor blocker dose escalation among biologic naive rheumatoid arthritis patients in commercial managed-care plans in the 2 years following therapy initiation.
      In one example, Liu et al
      • Liu Y.
      • Wu E.Q.
      • Bensimon A.G.
      • et al.
      Cost per responder associated with biologic therapies for Crohn’s disease, psoriasis, and rheumatoid arthritis.
      published results from mixed-treatment comparison meta-analyses in RA and other indications. The investigators estimated the 6-month costs per responder from pooled effectiveness data drawn from Phase III randomized controlled trials. Costs per responder, defined as a patient who achieved ≥50% improvement in the American College of Rheumatology score, were lowest among subcutaneously injected biologic agents (adalimumab, certolizumab, etanercept, golimumab, or tocilizumab) compared with infused therapies (abatacept, infliximab, or rituximab). The cost-effectiveness of subcutaneously injected biologic agents compared with that of infusion therapy has since been confirmed in multiple retrospective analyses of different commercial claims data from groups of insured patients with RA.
      • Kaplan R.S.
      • Witkowski M.
      • Abbott M.
      • et al.
      Using time-driven activity-based costing to identify value improvement opportunities in healthcare.
      • Himmelstein D.U.
      • Jun M.
      • Busse R.
      • et al.
      A comparison of hospital administrative costs in eight nations: US costs exceed all others by far.
      • Smolen J.S.
      • Breedveld F.C.
      • Burmester G.R.
      • et al.
      Treating rheumatoid arthritis to target: 2014 update of the recommendations of an international task force.
      • Bonafede M.M.
      • Gandra S.R.
      • Fox K.M.
      • Wilson K.L.
      Tumor necrosis factor blocker dose escalation among biologic naive rheumatoid arthritis patients in commercial managed-care plans in the 2 years following therapy initiation.
      If infusion therapy remains the appropriate clinical choice for managing a patient with RA, institutions should expect greater payer scrutiny of the setting where infusion services are delivered. Most infused anti-RA drugs can be administered at a physician׳s office, an outpatient center, or in the home, with quality outcomes similar to those with infusions administered in the hospital.
      • Einodshofer M.T.
      • Duren L.N.
      Cost management through care management, part 2: the importance of managing specialty drug utilization in the medical benefit.
      • Einodshofer M.
      • Ellis M.
      However, infusions performed in these alternative settings yield substantial cost savings; an analysis by Walgreens in 2008–2012 found that the mean cost per claim of hospital outpatient infusion of infliximab was $5790, compared with $3134 at alternate treatment sites.
      • Motheral B.
      • Fairman K.
      Specialty pharmacy: you can׳t manage what you can׳t measure.
      Furthermore, a report commissioned by the National Home Infusion Association found that initiating a $120 (2014) per diem payment for home-based infusion could save Medicare $79.9 million over a period of 10 years (2015–2024).
      • Drozd E.
      • Hoban N.
      Impact on Medicare Expenditures from Expanding Coverage of Infusion Therapy of Anti-Infective Drugs to the Home Setting.
      Although the funding structure of Medicare incentivizes users to seek hospital-based infusion, increasingly private payers are seeking to reduce the cost of infusion therapies, through both use of alternate treatment sites, and by providing medications to infusion sites rather than reimbursing physician billing.
      • Baldini C.G.
      • Culley E.J.
      Estimated cost savings associated with the transfer of office-administered specialty pharmaceuticals to a specialty pharmacy provider in a Medical Injectable Drug program.
      For many therapeutic indications in which infusion is preferable for select patients such as those with RA, it may become challenging to justify the provision of infusion services in the hospital when more cost-effective settings exist. An understanding of resource use patterns may make an individual biologic more or less attractive, and may influence a provider׳s choice of therapy given the availability of alternative treatments and treatment settings.
      Our analysis highlights the value to hospitals of assessing their infusion patient mix by disease or indication—based on whether other effective drugs are available in different dosage formulations (subcutaneous or oral) that can free up infusion center facilities and personnel to provide care to more patients in other therapy areas. While the use of estimates from national and public sources have helped to 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 "what if" scenarios for evaluating the impact of adjusting services to help inform decision making for the delivery of infusions services.
      The base-case analysis has certain limitations. First, the base case makes the simplifying assumption that all patients maintain a stable dose of infusion drug and a set infusion schedule over a 1-year time horizon. No additional resource use is assumed due to episodes of worsening symptoms or switching biologic therapies. In clinical practice, dose escalation,
      • Wu E.
      • Chen L.
      • Birnbaum H.
      • et al.
      Retrospective claims data analysis of dosage adjustment patterns of TNF antagonists among patients with rheumatoid arthritis.
      switching,
      • Greenberg J.D.
      • Reed G.
      • Decktor D.
      • et al.
      A comparative effectiveness study of adalimumab, etanercept and infliximab in biologically naive and switched rheumatoid arthritis patients: results from the US CORRONA registry.
      changes in the frequency of infusions,
      • Ogale S.
      • Hitraya E.
      • Henk H.J.
      Patterns of biologic agent utilization among patients with rheumatoid arthritis: a retrospective cohort study.
      worsening symptoms,
      • Singh J.A.
      • Furst D.E.
      • Bharat A.
      • et al.
      2012 update of the 2008 American College of Rheumatology recommendations for the use of disease-modifying antirheumatic drugs and biologic agents in the treatment of rheumatoid arthritis.
      • Smolen J.S.
      • Landewe R.
      • Breedveld F.C.
      • et al.
      EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2013 update.
      as well as treatment discontinuation
      • Greenberg J.D.
      • Reed G.
      • Decktor D.
      • et al.
      A comparative effectiveness study of adalimumab, etanercept and infliximab in biologically naive and switched rheumatoid arthritis patients: results from the US CORRONA registry.
      are to be expected among a segment of treated patients and would result in modifications to resource use. Second, as the data gathered in the primary source for nurse time
      • Pierce C.A.
      • Baker J.J.
      A nursing process model: quantifying infusion therapy resource consumption.
      were published prior to more recent dosing guidelines, it is assumed that the time estimates reflect standard practice, although it has not been validated. Third, 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. Fourth, 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 (eg, infusion chairs, monitors), yielding lower costs per infusion because associated overhead costs would be distributed among more individuals. While a host of different scenarios could lead to different costs for the patient, infusion interruption and inconvenience are an important consideration in the choice of treatment.
      • Ansari R.H.
      • Glaspy J.A.
      • Lu J.
      • et al.
      Assessing medication related disruption of life: results from a survey of patients receiving prophylactic filgrastim to prevent chemotherapy-induced neutropenia.
      Finally, neither of the studies from which time estimates were derived included extensive documentation about the role of non–health care personnel (office administrative staff, billing and insurance staff), which suggests that the model underestimates overall hospital costs.

      Conclusions

      Although the biologic agent is the greatest single cost associated with infusion center–based therapy for patients with RA, costs incurred by a hospital-based infusion center in administering any infusion drug may have a substantial financial impact. In an increasingly regulated and payer-constrained reimbursement environment, the model framework presented may assist administrators in exploring costs associated with their clinical practice patterns and case mix in order to inform choices about provision of infusion services for patients with RA and their associated financial implications. Future research to update and validate the infusion cost model assumptions with data from clinical practice would be helpful.

      Conflicts of Interest

      Design, study conduct, and financial support for the study were provided by AbbVie Inc.
      N.N. has received consultant׳s fees from Exponent Inc. J.S. is an employee of Exponent Inc. K.O. is a former employee of ICON PLC. M.H. is a former employee of Exponent Inc, has received funding for research from AbbVie, and has received consultant׳s fees from Medscape and Clinical Care Options. M.C., A.G., Y.B., and V.G. are employees of AbbVie and may hold shares in AbbVie. The authors have indicated that they have no other conflicts of interest with regard to the content of this article.

      Acknowledgments

      The authors express their appreciation to Michael L Friedman, MA, for his editorial contributions.
      All of the authors contributed to the interpretation of the data and review and approval of the manuscript. All of the authors contributed to the development of the publication and maintained control over the final content.

      References

        • Kent P.D.
        • Matteson E.L.
        Clinical features and differential diagnosis.
        in: St. Clair E.W. Pisetsky D.S. Haynes B.F. Rheumatoid arthritis. Lippincott Williams & Wilkins, Philadelphia, PA2004: 14-25
        • Singh J.A.
        • Furst D.E.
        • Bharat A.
        • et al.
        2012 update of the 2008 American College of Rheumatology recommendations for the use of disease-modifying antirheumatic drugs and biologic agents in the treatment of rheumatoid arthritis.
        Arthritis care & research. 2012; 64: 625-639
        • Smolen J.S.
        • Landewe R.
        • Breedveld F.C.
        • et al.
        EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2013 update.
        Ann Rheum Dis. 2014; 73: 492-509
        • Dougados M.
        • Soubrier M.
        • Antunez A.
        • et al.
        Prevalence of comorbidities in rheumatoid arthritis and evaluation of their monitoring: results of an international, cross-sectional study (COMORA).
        Ann Rheum Dis. 2014; 73: 62-68
        • Lawrence R.C.
        • Helmick C.G.
        • Arnett F.C.
        • et al.
        Estimates of the prevalence of arthritis and selected musculoskeletal disorders in the United States.
        Arthritis Rheum. 1998; 41: 778-799
        • Helmick C.G.
        • Lawrence R.C.
        • Pollard R.A.
        • et al.
        Arthritis and other rheumatic conditions: who is affected now, who will be affected later? National Arthritis Data Workgroup.
        Arthritis Care & Research. 1995; 8: 203-211
        • Ward M.M.
        Medicare reimbursement and the use of biologic agents: incentives, access, the public good, and optimal care.
        Arthritis Care & Research. 2010; 62: 293-295
        • Yazdany J.
        • Dudley R.A.
        • Chen R.
        • et al.
        Coverage for high-cost specialty drugs for rheumatoid arthritis in Medicare Part D.
        Arthritis Rheumatology. 2015; 67: 1474-1480
        • Brimson J.A.
        Activity Accounting: An Activity-Based Costing Approach.
        John Wiley, New York1991
        • Kaplan R.S.
        • Witkowski M.
        • Abbott M.
        • et al.
        Using time-driven activity-based costing to identify value improvement opportunities in healthcare.
        J Healthc Manag. 2014; 59: 399-412
        • Brixner D.I.
        • Oderda G.M.
        • Nickman N.A.
        • et al.
        Documentation of chemotherapy infusion preparation costs in academic- and community-based oncology practices.
        J Natl Compr Canc Netw. 2006; 4: 197-208
        • Pierce C.A.
        • Baker J.J.
        A nursing process model: quantifying infusion therapy resource consumption.
        J Infus Nurs. 2004; 27: 232-244
      1. Actemra (tocilizumab) [package insert].
        Genentech, Inc, South San Francisco, CA2013
        • Greenwald M.W.
        • Shergy W.J.
        • Kaine J.L.
        • et al.
        Evaluation of the tolerability of rituximab in combination with a tumor necrosis factor inhibitor and methotrexate in patients with active rheumatoid arthritis: results from a randomized controlled trial.
        Arthritis Rheum. 2011; 63: 622-632
        • Schiff M.
        • Keiserman M.
        • Codding C.
        • et al.
        Efficacy and tolerability of abatacept or infliximab vs placebo in ATTEST: a phase III, multi-centre, randomised, double-blind, placebo-controlled study in patients with rheumatoid arthritis and an inadequate response to methotrexate.
        Ann Rheum Dis. 2008; 67: 1096-1103
      2. Orencia (abatacept) [package insert].
        Bristol-Myers Squibb Company, Princeton, NJ2013
      3. Rituxan (rituximab) [package insert].
        Genentech, Inc, South San Francisco, CA2013
      4. Remicade (infliximab) [package insert].
        Janssen Biotech Inc, Horsham, PA2013
        • US Bureau of Labor Statistics
        Employer Costs for Employee Compensation Historical Listing March 2004 - December 2016. 2017; (Accessed April 26, 2017)
        • US Bureau of Labor Statistics
        National Occupational Employment and Wage Estimates (United States) - Healthcare Practitioner and Technical Occupations (May 2016). 2017; (Accessed April 26, 2017)
        • Wolters Kluwer Health
        Medi-Span Price Rx [member access only]. 2017; (Accessed April 26, 2017)
        • US Bureau of Labor Statistics
        CPI Detailed Report (March 2017), Table 25. 2017; (Accessed April 26, 2017)
        • Cipriani P.
        • Ruscitti P.
        • Carubbi F.
        • et al.
        Methotrexate in Rheumatoid Arthritis: Optimizing Therapy Among Different Formulations. Current and Emerging Paradigms.
        Clin Ther. 2014; 36: 427-435
        • Visser K.
        • Katchamart W.
        • Loza E.
        • et al.
        Multinational evidence-based recommendations for the use of methotrexate in rheumatic disorders with a focus on rheumatoid arthritis: integrating systematic literature research and expert opinion of a broad international panel of rheumatologists in the 3E Initiative.
        Ann Rheum Dis. 2009; 68: 1086-1093
        • Wolters Kluwer Health
        Medi-Span Price Rx [member access only]. 2014; (Accessed May 23, 2014)
        • Lenz H.J.
        Management and preparedness for infusion and hypersensitivity reactions.
        Oncologist. 2007; 12: 601-609
        • Internal Revenue Service
        Publication 946. How to Depreciate Property. 2013; (Accessed April 26, 2017)
        • DesHarnais Castel L.
        • Bajwa K.
        • Markle J.P.
        • et al.
        A microcosting analysis of zoledronic acid and pamidronate therapy in patients with metastatic bone disease.
        Support care cancer. 2001; 9: 545-551
      5. Rheumatrex (methotrexate) [package insert].
        DAVA Pharmaceuticals, Inc, Fort Lee, NJ2010
        • Centers for Medicare & Medicaid Service
        Clinical Laboratory Fee Schedule. 2017; (Accessed April 26, 2017)
        • Centers for Medicare & Medicaid Services
        Physician Fee Schedule. 2017; (Accessed April 26, 2017)
        • Luce B.R.
        • Zangwill K.M.
        • Palmer C.S.
        • et al.
        Cost-effectiveness analysis of an intranasal influenza vaccine for the prevention of influenza in healthy children.
        Pediatrics. 2001; 108: E24
        • Greenberg J.D.
        • Reed G.
        • Decktor D.
        • et al.
        A comparative effectiveness study of adalimumab, etanercept and infliximab in biologically naive and switched rheumatoid arthritis patients: results from the US CORRONA registry.
        Ann Rheum Dis. 2012; 71: 1134-1142
        • Himmelstein D.U.
        • Jun M.
        • Busse R.
        • et al.
        A comparison of hospital administrative costs in eight nations: US costs exceed all others by far.
        Health affairs (Millwood). 2014; 33: 1586-1594
        • Kalman N.
        • Hammill B.
        • Schulman K.
        • Shah B.
        Hospital overhead costs: the neglected driver of health care spending?.
        J Healthc Finance. 2015; 41: 1-15
        • Smolen J.S.
        • Breedveld F.C.
        • Burmester G.R.
        • et al.
        Treating rheumatoid arthritis to target: 2014 update of the recommendations of an international task force.
        Ann Rheum Dis. 2016; 75: 3-15
        • Bonafede M.M.
        • Gandra S.R.
        • Fox K.M.
        • Wilson K.L.
        Tumor necrosis factor blocker dose escalation among biologic naive rheumatoid arthritis patients in commercial managed-care plans in the 2 years following therapy initiation.
        J Med Econ. 2012; 15: 635-643
        • Liu Y.
        • Wu E.Q.
        • Bensimon A.G.
        • et al.
        Cost per responder associated with biologic therapies for Crohn’s disease, psoriasis, and rheumatoid arthritis.
        Advances in Therapy. 2012; 29: 1-15
        • Einodshofer M.T.
        • Duren L.N.
        Cost management through care management, part 2: the importance of managing specialty drug utilization in the medical benefit.
        American Health Drug Benefits. 2012; 5: 359-364
        • Einodshofer M.
        • Ellis M.
        The Challenge of Specialty Pharmacy: Mitigating Costs While Providing Quality Care Secondary The Challenge of Specialty Pharmacy: Mitigating Costs While Providing Quality Care. 2013; (Accessed March 20, 2016)
        • Motheral B.
        • Fairman K.
        Specialty pharmacy: you can׳t manage what you can׳t measure.
        Employee Benefit News. 2013; 27: 13-15
        • Drozd E.
        • Hoban N.
        Impact on Medicare Expenditures from Expanding Coverage of Infusion Therapy of Anti-Infective Drugs to the Home Setting.
        National Home Infusion Association. 2014;
        • Baldini C.G.
        • Culley E.J.
        Estimated cost savings associated with the transfer of office-administered specialty pharmaceuticals to a specialty pharmacy provider in a Medical Injectable Drug program.
        Journal of Managed Care Pharmacy. 2011; 17: 51-59
        • Wu E.
        • Chen L.
        • Birnbaum H.
        • et al.
        Retrospective claims data analysis of dosage adjustment patterns of TNF antagonists among patients with rheumatoid arthritis.
        Curr Med Res Opin. 2008; 24: 2229-2240
        • Ogale S.
        • Hitraya E.
        • Henk H.J.
        Patterns of biologic agent utilization among patients with rheumatoid arthritis: a retrospective cohort study.
        BMC Musculoskelet Disord. 2011; 12: 204
        • Ansari R.H.
        • Glaspy J.A.
        • Lu J.
        • et al.
        Assessing medication related disruption of life: results from a survey of patients receiving prophylactic filgrastim to prevent chemotherapy-induced neutropenia.
        Paper presented at: American Society of Clinical Oncology 38th Annual Meeting, Orlando, FL2002