Advertisement

Cost Savings from Reduced Hospitalizations with Use of Home Noninvasive Ventilation for COPD

Open ArchivePublished:November 11, 2016DOI:https://doi.org/10.1016/j.jval.2016.09.2401

      Abstract

      Background

      Although evidence suggests significant clinical benefits of home noninvasive ventilation (NIV) for management of severe chronic obstructive pulmonary disease (COPD), economic analyses supporting the use of this technology are lacking.

      Objectives

      To evaluate the economic impact of adopting home NIV, as part of a multifaceted intervention program, for severe COPD.

      Methods

      An economic model was developed to calculate savings associated with the use of Advanced NIV (averaged volume assured pressure support with autoexpiratory positive airway pressure; Trilogy100, Philips Respironics, Inc., Murrysville, PA) versus either no NIV or a respiratory assist device with bilevel pressure capacity in patients with severe COPD from two distinct perspectives: the hospital and the payer. The model examined hospital savings over 90 days and payer savings over 3 years. The number of patients with severe COPD eligible for home Advanced NIV was user-defined. Clinical and cost data were obtained from a quality improvement program and published reports. Scenario analyses calculated savings for hospitals and payers covering different COPD patient cohort sizes.

      Results

      The hospital base case (250 patients) revealed cumulative savings of $402,981 and $449,101 over 30 and 90 days, respectively, for Advanced NIV versus both comparators. For the payer base case (100,000 patients), 3-year cumulative savings with Advanced NIV were $326 million versus no NIV and $1.04 billion versus respiratory assist device.

      Conclusions

      This model concluded that adoption of home Advanced NIV with averaged volume assured pressure support with autoexpiratory positive airway pressure, as part of a multifaceted intervention program, presents an opportunity for hospitals to reduce COPD readmission-related costs and for payers to reduce costs associated with managing patients with severe COPD on the basis of reduced admissions.

      Keywords

      Introduction

      Chronic obstructive pulmonary disease (COPD) is a progressive disease characterized by airflow limitation and loss of lung function that is not fully reversible [
      • Pauwels R.A.
      • Buist A.S.
      • Calverley P.M.
      • et al.
      Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease. NHLBI/WHO Global Initiative for Chronic Obstructive Lung Disease (GOLD) Workshop summary.
      ,
      • Celli B.R.
      • MacNee W.
      • Force A.E.T.
      Standards for the diagnosis and treatment of patients with COPD: a summary of the ATS/ERS position paper.
      ,
      Global Initiative for Chronic Obstructive Lung Disease. Global Strategy for the Diagnosis, Management and Prevention of COPD.
      ]. In the United States, COPD is a major cause of morbidity and mortality, and a cause of substantial increases in health care costs, mainly as a result of inpatient admissions [
      Global Initiative for Chronic Obstructive Lung Disease. Global Strategy for the Diagnosis, Management and Prevention of COPD.
      ,
      • Dalal A.A.
      • Shah M.
      • D’Souza A.O.
      • et al.
      Costs of COPD exacerbations in the emergency department and inpatient setting.
      ,
      • Chapman K.R.
      • Mannino D.M.
      • Soriano J.B.
      • et al.
      Epidemiology and costs of chronic obstructive pulmonary disease.
      ,
      • Breunig I.M.
      • Shaya F.T.
      • Scharf S.M.
      Delivering cost-effective care for COPD in the USA: recent progress and current challenges.
      ]. National data indicate that more than 20% of patients hospitalized because of COPD are rehospitalized within 30 days [
      • Jennings J.H.
      • Thavarajah K.
      • Mendez M.
      • et al.
      Predischarge bundle for patients with acute exacerbations of chronic obstructive pulmonary disease to reduce readmissions and emergency department visits: a randomized controlled trial.
      ,
      • Shah T.
      • Churpek M.M.
      • Coca Perraillon M.
      • et al.
      Understanding why patients with COPD get readmitted: a large national study to delineate the Medicare population for the readmissions penalty expansion.
      ]. In an effort to reduce COPD readmissions, the Centers for Medicare & Medicaid Services (CMS) Hospital Readmission Reduction Program (HRRP) now includes penalties for all-cause unplanned readmissions within 30 days of an index admission for a COPD exacerbation [
      • Feemster L.C.
      • Au D.H.
      Penalizing hospitals for chronic obstructive pulmonary disease readmissions.
      ]. Thus, implementation of innovative measures to reduce readmissions among patients with COPD represents an important objective for hospitals and payers [
      • Jennings J.H.
      • Thavarajah K.
      • Mendez M.
      • et al.
      Predischarge bundle for patients with acute exacerbations of chronic obstructive pulmonary disease to reduce readmissions and emergency department visits: a randomized controlled trial.
      ,
      • Dalal A.A.
      • Chiristensen L.
      • Liu F.
      • et al.
      Direct costs of chronic obstructive pulmonary disease among managed care patients.
      ].
      Noninvasive ventilation (NIV) is considered a standard of care for in-hospital treatment of acute COPD exacerbations [
      Global Initiative for Chronic Obstructive Lung Disease. Global Strategy for the Diagnosis, Management and Prevention of COPD.
      ,
      • Pierson D.J.
      History and epidemiology of noninvasive ventilation in the acute-care setting.
      ]. Nevertheless, its use in long-term management of stable, chronic COPD is less common in the United States, in part because evidence has not demonstrated a consistent improvement in survival [
      • Casanova C.
      • Celli B.R.
      • Tost L.
      • et al.
      Long-term controlled trial of nocturnal nasal positive pressure ventilation in patients with severe COPD.
      ,
      • Clini E.
      • Sturani C.
      • Rossi A.
      • et al.
      The Italian multicentre study on noninvasive ventilation in chronic obstructive pulmonary disease patients.
      ,
      • McEvoy R.D.
      • Pierce R.J.
      • Hillman D.
      • et al.
      Nocturnal non-invasive nasal ventilation in stable hypercapnic COPD: a randomised controlled trial.
      ,
      • Struik F.M.
      • Sprooten R.T.
      • Kerstjens H.A.
      • et al.
      Nocturnal non-invasive ventilation in COPD patients with prolonged hypercapnia after ventilatory support for acute respiratory failure: a randomised, controlled, parallel-group study.
      ,
      • Tsolaki V.
      • Pastaka C.
      • Karetsi E.
      • et al.
      One-year non-invasive ventilation in chronic hypercapnic COPD: effect on quality of life.
      ]. From the lung physician’s perspective, some of these previous studies [
      • Casanova C.
      • Celli B.R.
      • Tost L.
      • et al.
      Long-term controlled trial of nocturnal nasal positive pressure ventilation in patients with severe COPD.
      ,
      • Clini E.
      • Sturani C.
      • Rossi A.
      • et al.
      The Italian multicentre study on noninvasive ventilation in chronic obstructive pulmonary disease patients.
      ] may have failed to show favorable outcomes because they did not achieve a reduction in hypercapnia during ventilation [
      • Kohnlein T.
      • Windisch W.
      • Kohler D.
      • et al.
      Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial.
      ]. This was recently supported by a randomized controlled trial, which demonstrated improved survival with home NIV with high backup rates targeted to markedly reduce hypercapnia versus optimized COPD therapy without NIV, but not reduced rates of emergency hospital admission [
      • Kohnlein T.
      • Windisch W.
      • Kohler D.
      • et al.
      Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial.
      ]. More recently, a quality improvement program in a severe COPD cohort with a history of two or more admissions in the previous year demonstrated that multifaceted in-home intervention, including use of an advanced NIV modality (averaged volume assured pressure support with autoexpiratory positive airway pressure; Trilogy100, Philips Respironics, Inc., Murrysville, PA; Advanced NIV), which automatically titrates the device to simultaneously maintain airway patency and a target tidal volume, significantly reduced rehospitalization rates [
      • Coughlin S.
      • Liang W.E.
      • Parthasarathy S.
      Retrospective assessment of home ventilation to reduce rehospitalization in chronic obstructive pulmonary disease.
      ]. Collectively, these data suggest additional benefits to using advanced NIV approaches in long-term management of stable chronic COPD.
      Given this data-driven evidence and growing emphasis on reducing inpatient COPD admissions, an economic model was developed to assess the economic impact of using different categories of home NIV for long-term management of severe COPD.

      Methods

      Perspective

      The economic model, constructed using Microsoft Excel 2013, was designed to assess the economic impact of using home NIV for the management of patients with severe COPD within the United States (at the time this article was written) from two different perspectives: the hospital and the payer (health insurance provider).

      Model and Analytic Framework

      The economic model directly compared outcome-driven costs and savings associated with the use of Advanced NIV as part of a multifaceted intervention program, versus either no NIV or a respiratory assist device (RAD), bilevel pressure capacity with backup rate for in-home care.
      For the hospital, the model calculates readmission-associated savings over 0 to 30, 31 to 60, and 61 to 90 days. The first time horizon reflects the period during which hospitals are presently subject to COPD readmission penalties, and the last two time horizons were examined to provide a longitudinal view of COPD readmission costs. For the payer, the model calculates annual and cumulative savings over a 3-year period.

      Model Inputs and Data Sources

      Model inputs are presented in Table 1. In this analysis, Advanced NIV is defined as a pressure support ventilator with volume control mode (Trilogy100; Healthcare Common Procedure Coding System [HCPCS] code E0464) and RAD as a respiratory assist device, bilevel pressure capacity with backup rate (HCPCS code E0471) (see Appendix Table 1 in Supplemental Materials found at doi:10.1016/j.jval.2016.09.2401). Patients with severe COPD who qualified for home NIV were required to fulfill the following CMS criteria for reimbursement coverage of home NIV: diagnosis of severe COPD, obstructive sleep apnea and continuous positive airway pressure therapy ruled out, and interruption or failure of respiratory support potentially leading to death [
      Centers for Medicare & Medicaid Services. Local Coverage Determination (LCD): Respiratory Assist Devices. Baltimore, MD: Centers for Medicare & Medicaid Services, 2015.
      ].
      Table 1Summary of model inputs
      Model inputValueSource
      Number of patients with severe COPD
      Severe COPD defined as GOLD stages II–IV, Bode Index Score of ≥5, and at least one of the following: PaCO2 ≥ 52 mm Hg, PaO2 ≤ 60 mm Hg, or FEV1 ≤ 40%.
      eligible for home NIV
       Hospital, n250Base-case assumption
       Payer, n100,000Base-case assumption
       Event frequency (30-d hospital readmission rate)6.9%
      • Ford E.S.
      Hospital discharges, readmissions, and ED visits for COPD or bronchiectasis among US adults: findings from the Nationwide Inpatient Sample 2001–2012 and Nationwide Emergency Department Sample 2006–2011.
      No NIV
       1-y mortality rate29.0%
      • Struik F.M.
      • Sprooten R.T.
      • Kerstjens H.A.
      • et al.
      Nocturnal non-invasive ventilation in COPD patients with prolonged hypercapnia after ventilatory support for acute respiratory failure: a randomised, controlled, parallel-group study.
       Admissions per patient per year, n3.1
      • Kohnlein T.
      • Windisch W.
      • Kohler D.
      • et al.
      Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial.
      RAD
       1-y mortality rate29.7%
      • Struik F.M.
      • Sprooten R.T.
      • Kerstjens H.A.
      • et al.
      Nocturnal non-invasive ventilation in COPD patients with prolonged hypercapnia after ventilatory support for acute respiratory failure: a randomised, controlled, parallel-group study.
       Admissions per patient per year, n3.1
      Same value as for no NIV, on the basis of finding no difference in readmission rates from published source.
      • Struik F.M.
      • Sprooten R.T.
      • Kerstjens H.A.
      • et al.
      Nocturnal non-invasive ventilation in COPD patients with prolonged hypercapnia after ventilatory support for acute respiratory failure: a randomised, controlled, parallel-group study.
      ,
      • Kohnlein T.
      • Windisch W.
      • Kohler D.
      • et al.
      Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial.
       Dropout rate18.2%Mean from
      • Windisch W.
      • Kostic S.
      • Dreher M.
      • et al.
      Outcome of patients with stable COPD receiving controlled noninvasive positive pressure ventilation aimed at a maximal reduction of Pa(CO2).
      and
      • Clini E.
      • Sturani C.
      • Rossi A.
      • et al.
      The Italian multicentre study on noninvasive ventilation in chronic obstructive pulmonary disease patients.
      ,
      • Struik F.M.
      • Sprooten R.T.
      • Kerstjens H.A.
      • et al.
      Nocturnal non-invasive ventilation in COPD patients with prolonged hypercapnia after ventilatory support for acute respiratory failure: a randomised, controlled, parallel-group study.
      ,
      • Tsolaki V.
      • Pastaka C.
      • Karetsi E.
      • et al.
      One-year non-invasive ventilation in chronic hypercapnic COPD: effect on quality of life.
      ,
      • Windisch W.
      Quality of life in home mechanical ventilation study group
      Impact of home mechanical ventilation on health-related quality of life.
       Patients requiring sleep study100%Base-case assumption
      Advanced NIV
       1-y mortality rate18.3%
      • Coughlin S.
      • Liang W.E.
      • Parthasarathy S.
      Retrospective assessment of home ventilation to reduce rehospitalization in chronic obstructive pulmonary disease.
       Admissions per patient per year, n0.16
      • Coughlin S.
      • Liang W.E.
      • Parthasarathy S.
      Retrospective assessment of home ventilation to reduce rehospitalization in chronic obstructive pulmonary disease.
       Dropout rate18.2%Mean from
      • Windisch W.
      • Kostic S.
      • Dreher M.
      • et al.
      Outcome of patients with stable COPD receiving controlled noninvasive positive pressure ventilation aimed at a maximal reduction of Pa(CO2).
      and
      • Clini E.
      • Sturani C.
      • Rossi A.
      • et al.
      The Italian multicentre study on noninvasive ventilation in chronic obstructive pulmonary disease patients.
      ,
      • Struik F.M.
      • Sprooten R.T.
      • Kerstjens H.A.
      • et al.
      Nocturnal non-invasive ventilation in COPD patients with prolonged hypercapnia after ventilatory support for acute respiratory failure: a randomised, controlled, parallel-group study.
      ,
      • Tsolaki V.
      • Pastaka C.
      • Karetsi E.
      • et al.
      One-year non-invasive ventilation in chronic hypercapnic COPD: effect on quality of life.
      ,
      • Windisch W.
      Quality of life in home mechanical ventilation study group
      Impact of home mechanical ventilation on health-related quality of life.
      Admission costs$7946
      • Ford E.S.
      Hospital discharges, readmissions, and ED visits for COPD or bronchiectasis among US adults: findings from the Nationwide Inpatient Sample 2001–2012 and Nationwide Emergency Department Sample 2006–2011.
      Hospital reimbursement
      Calculations are based on 2015 reimbursement rates. Given that reimbursement rates are expected to change over time and may vary by geographical areas, consulting payers for present rates is advised.
       Admission (outside of 30 d)$7570
      Centers for Medicare & Medicaid Services
      Medicare Charge In-Patient Summary. National and State Summary.
       Sleep study$657
      Centers for Medicare & Medicaid Services
      2015 Fee Schedule.
      Device reimbursement
      Calculations are based on 2015 reimbursement rates. Given that reimbursement rates are expected to change over time and may vary by geographical areas, consulting payers for present rates is advised.
       RAD
        Device (per month)$475
      Centers for Medicare & Medicaid Services
      2015 Fee Schedule.
        Reimbursement cap13 mo
      Centers for Medicare & Medicaid Services
      2015 Fee Schedule.
        Supplies (per year)$462
      Includes humidification (water chamber) at $18.29 (2 times/year), filters (ultrafine, disposable, 2-pack) at $4.74 (6 times/year), NIV mask (full face) at $180.47 (2 times/year), tubing (System One Performance Tubing, 15 mm) at $38.00 (4 times/year).
      Centers for Medicare & Medicaid Services
      2015 DME Fee Schedule..
      $462 based on the assumption that the payer is responsible for payment of 80% of total supplies reimbursement amount ($578) listed in the CMS 2015 DME Fee Schedule.
       Advanced NIV
        Device (per month; no cap)$1592
      Includes device, supplies, and supportive care.
      Centers for Medicare & Medicaid Services
      2015 Fee Schedule.
      CMS, Centers for Medicare & Medicaid Services; COPD, chronic obstructive pulmonary disease; DME, durable medical equipment; FEV1, forced expiratory volume in 1 s; GOLD, Global Initiative for Chronic Obstructive Lung Disease; NIV, noninvasive ventilation; RAD, respiratory assist device.
      low asterisk Severe COPD defined as GOLD stages II–IV, Bode Index Score of ≥5, and at least one of the following: PaCO2 ≥ 52 mm Hg, PaO2 ≤ 60 mm Hg, or FEV1 ≤ 40%.
      Same value as for no NIV, on the basis of finding no difference in readmission rates from published source.
      Calculations are based on 2015 reimbursement rates. Given that reimbursement rates are expected to change over time and may vary by geographical areas, consulting payers for present rates is advised.
      § Includes humidification (water chamber) at $18.29 (2 times/year), filters (ultrafine, disposable, 2-pack) at $4.74 (6 times/year), NIV mask (full face) at $180.47 (2 times/year), tubing (System One Performance Tubing, 15 mm) at $38.00 (4 times/year).
      || $462 based on the assumption that the payer is responsible for payment of 80% of total supplies reimbursement amount ($578) listed in the CMS 2015 DME Fee Schedule.
      Includes device, supplies, and supportive care.
      The number of admissions per patient per year for Advanced NIV was derived from an unpublished analysis of a quality improvement program demonstrating significantly reduced readmissions associated with the use of Trilogy100 NIV with averaged volume assured pressure support with autoexpiratory positive airway pressure, as part of a multifaceted program including oxygen therapy, respiratory therapist (RT)-led respiratory care, patient education, and medication reconciliation, in patients with severe COPD with two or more exacerbations in the previous year [
      • Coughlin S.
      • Liang W.E.
      • Parthasarathy S.
      Retrospective assessment of home ventilation to reduce rehospitalization in chronic obstructive pulmonary disease.
      ]. Severe COPD was defined as Global Initiative for Chronic Obstructive Lung Disease (GOLD) stages II to IV, Bode Index Score of 5 or higher, and at least one of the following: PaCO2 of 52 mm Hg or higher, PaO2 of 60 mm Hg or less, or forced expiratory volume in 1 second of 40% or less. In total, there were 397 patients in the program, out of which 9 (2.3%) were in GOLD stage II, 292 (73.6%) were in GOLD stage III, and 96 (24.2%) were in GOLD stage IV [
      • Coughlin S.
      • Liang W.E.
      • Parthasarathy S.
      Retrospective assessment of home ventilation to reduce rehospitalization in chronic obstructive pulmonary disease.
      ]. Advanced NIV with Trilogy100 was used in target tidal volume mode, with an expiratory positive airway pressure (EPAP) range of 5 to 15 cm H2O, a pressure support range of 2 to 26 cm H2O, and an automatic backup rate. Settings enabled automatic EPAP titration to treat events of obstructive sleep apnea, if present; automatically adjust pressure support to achieve a physician-defined target tidal volume based on ideal body weight (set at 5–7 ml/kg); and maximize expiratory time by adjusting the backup rate on the basis of the patient’s spontaneous rate to help avoid breath-stacking [
      • Coughlin S.
      • Liang W.E.
      • Parthasarathy S.
      Retrospective assessment of home ventilation to reduce rehospitalization in chronic obstructive pulmonary disease.
      ]. Mortality rate among patients treated with Advanced NIV was based on unpublished results from the same quality improvement program [
      • Coughlin S.
      • Liang W.E.
      • Parthasarathy S.
      Retrospective assessment of home ventilation to reduce rehospitalization in chronic obstructive pulmonary disease.
      ].
      Mortality rates for RAD and no NIV were based on a study in which each device was used at home along with oxygen therapy and pharmacotherapy for adult patients in GOLD stage III or IV with more than 48 hours independence from ventilatory support (invasive or noninvasive) for acute respiratory failure, prolonged hypercapnia (PaCO2 > 45 mm Hg), and no significant heart failure or neuromuscular disease. RAD with bilevel positive airway pressure in the spontaneous/timed mode was initiated at a low backup frequency of 12 breaths per minute, with the aim to achieve normocapnia and maximally support respiration. Inspiration-to-expiration time was set at 1:3. Inspiratory positive airway pressure was initiated at 14 cm H2O and increased to a maximum tolerated level, whereas EPAP was initiated at 4 cm H2O and increased when patients used respiratory muscles to trigger the ventilator, or if auto-positive end expiratory pressure was present [
      • Struik F.M.
      • Sprooten R.T.
      • Kerstjens H.A.
      • et al.
      Nocturnal non-invasive ventilation in COPD patients with prolonged hypercapnia after ventilatory support for acute respiratory failure: a randomised, controlled, parallel-group study.
      ]. Struik et al. [
      • Struik F.M.
      • Sprooten R.T.
      • Kerstjens H.A.
      • et al.
      Nocturnal non-invasive ventilation in COPD patients with prolonged hypercapnia after ventilatory support for acute respiratory failure: a randomised, controlled, parallel-group study.
      ] also reported no difference in the admission rate between patients treated with RAD and no NIV, but failed to report an average frequency for each comparator group.
      The number of admissions per patient per year for no NIV was based on a study in which no NIV consisted of standard treatment [
      • Kohnlein T.
      • Windisch W.
      • Kohler D.
      • et al.
      Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial.
      ] accompanied by oxygen therapy in clinically stable, hypercapnic, GOLD stage IV adult patients with COPD, with a PaCO2 of 51.9 mm Hg or higher, pH of more than 7.35, body mass index of less than 35 kg/m2, and no malignant comorbidities, severe heart failure, unstable angina, or severe arrhythmias. RAD was set up with high backup rates to achieve controlled ventilation, or assisted ventilation if patients did not tolerate high backup rates, with the aim to reduce baseline PaCO2 by 20% or more, or achieve a PaCO2 of less than 48.1 mm Hg [
      • Kohnlein T.
      • Windisch W.
      • Kohler D.
      • et al.
      Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial.
      ]. Because Struik et al. [
      • Struik F.M.
      • Sprooten R.T.
      • Kerstjens H.A.
      • et al.
      Nocturnal non-invasive ventilation in COPD patients with prolonged hypercapnia after ventilatory support for acute respiratory failure: a randomised, controlled, parallel-group study.
      ] reported no difference in the admission rate between RAD and no NIV treatment, the RAD admission frequency was assumed to be the same as the no NIV admission frequency derived by Kohnlein et al. [
      • Kohnlein T.
      • Windisch W.
      • Kohler D.
      • et al.
      Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial.
      ].
      All other event frequencies, including 30-day readmission rates and dropout rates, were derived from previously published studies [
      • Clini E.
      • Sturani C.
      • Rossi A.
      • et al.
      The Italian multicentre study on noninvasive ventilation in chronic obstructive pulmonary disease patients.
      ,
      • Struik F.M.
      • Sprooten R.T.
      • Kerstjens H.A.
      • et al.
      Nocturnal non-invasive ventilation in COPD patients with prolonged hypercapnia after ventilatory support for acute respiratory failure: a randomised, controlled, parallel-group study.
      ,
      • Tsolaki V.
      • Pastaka C.
      • Karetsi E.
      • et al.
      One-year non-invasive ventilation in chronic hypercapnic COPD: effect on quality of life.
      ,
      • Ford E.S.
      Hospital discharges, readmissions, and ED visits for COPD or bronchiectasis among US adults: findings from the Nationwide Inpatient Sample 2001–2012 and Nationwide Emergency Department Sample 2006–2011.
      ,
      • Windisch W.
      • Kostic S.
      • Dreher M.
      • et al.
      Outcome of patients with stable COPD receiving controlled noninvasive positive pressure ventilation aimed at a maximal reduction of Pa(CO2).
      ,
      • Windisch W.
      Quality of life in home mechanical ventilation study group
      Impact of home mechanical ventilation on health-related quality of life.
      ].
      Hospital admission costs were obtained from previously published data [
      • Ford E.S.
      Hospital discharges, readmissions, and ED visits for COPD or bronchiectasis among US adults: findings from the Nationwide Inpatient Sample 2001–2012 and Nationwide Emergency Department Sample 2006–2011.
      ]. For examination of hospital costs, device-related costs (acquisition, repair, setup, supplies, and supportive care) were set to $0, reflecting the present practice in which US hospitals incur no device-related costs. Reimbursements for hospital admission (outside 30 days), sleep study, device, and supplies were defined by 2015 CMS-reported rates [
      Centers for Medicare & Medicaid Services
      Medicare Charge In-Patient Summary. National and State Summary.
      ,
      Centers for Medicare & Medicaid Services
      2015 Fee Schedule.
      ,
      Centers for Medicare & Medicaid Services
      2015 DME Fee Schedule..
      ].

      Analyses and Outputs: Hospital Perspective

      The model estimates hospital readmissions for time periods of 0 to 30, 31 to 60, and 61 to 90 days. Numbers of 30-day readmissions for each comparator were defined as cohort size × admission frequency per patient per year × 30-day readmission rate. Readmissions for 31 to 60 and 61 to 90 days were extrapolated using the number of 30-day readmissions (see Appendix Table 2 in Supplemental Materials found at doi:10.1016/j.jval.2016.09.2401).
      For each comparator, the cost of readmissions during each time frame was defined as the number of readmissions over the time frame × the cost per admission. Total reimbursement amount was defined as the number of readmissions × the reimbursement amount per readmission. Reimbursement was set to $0 for readmissions within 30 days. The total cost of readmissions over each given time frame was calculated as the difference between the readmissions cost and reimbursement. Admissions-related savings were calculated as the total readmissions cost difference between Advanced NIV and either comparator. All calculated costs and savings were adjusted to 2015 US dollars [

      United States Bureau of Labor and Statistics. CPI inflation calculator. Available from: http://www.bls.gov/data/inflation_calculator.htm. [Accessed February 1, 2015]

      ]. Because the same admission frequency was used for no NIV and RAD, as described earlier, cost savings were expected to be similarly impacted across both comparators.

      Analyses and Outputs: Payer Perspective

      From a payer perspective, patient outcomes over the 3-year period included annual numbers of admissions and surviving patients. Annual admissions were calculated from the initial number of patients with severe COPD (year 0) or the number of surviving patients from the previous year (years 1–3), and admission frequency per patient per year with a given intervention (Advanced NIV, no NIV, or RAD). The number of surviving patients in year 1 for each comparator was based on initial numbers of patients with severe COPD × 1-year mortality rate. The number of surviving patients in subsequent years was calculated by applying the 1-year mortality rate to the number of patients surviving the preceding year. This approach led to year 2 and year 3 survival. Total annual cost was calculated as the sum of the following: 1) number of annual admissions × reimbursement per admission (outside 30 days), 2) number of patients eligible for device-related reimbursement × device-related reimbursement, and 3) for year 0 and RAD only, number of patients with severe COPD × reimbursement for one sleep study per patient required before commencement of RAD treatment [

      Gay PC. Nocturnal ventilatory support in COPD. In: Stoller JK, ed., UpToDate. Published by Wolters Kluwer, Alphen aan den Rijn, Netherlands, 2015.

      ]. The number of patients eligible for device-related reimbursement was calculated from the number of surviving patients and the number of dropouts (dropout rate × number of surviving patients). Device-related reimbursement included reimbursement for device acquisition, repair, setup, supplies, and supportive care, as applicable (Table 1), and was capped at 13 months (per CMS guidelines) [
      Centers for Medicare & Medicaid Services
      2015 Fee Schedule.
      ] for RAD, but uncapped for Advanced NIV (see Appendix Table 2 in Supplemental Materials for detailed calculations).

      Base-Case and Scenario Analyses: Hospital Perspective

      The hospital base case (Table 1) was defined as 250 patients with severe COPD, on the basis of available US data regarding hospital size, COPD prevalence, and readmissions [
      • Kohnlein T.
      • Windisch W.
      • Kohler D.
      • et al.
      Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial.
      ,
      • Ford E.S.
      Hospital discharges, readmissions, and ED visits for COPD or bronchiectasis among US adults: findings from the Nationwide Inpatient Sample 2001–2012 and Nationwide Emergency Department Sample 2006–2011.
      ,
      Mount Sinai Health System. Available from: http://www.bethisraelny.org/brooklyn/. [Accessed June 10, 2015].
      ,
      The World Bank Data. Available from: http://data.worldbank.org/indicator/SH.MED.BEDS.ZS. [Accessed June 10, 2015].
      ,
      United States Census Bureau—Population. Available from: http://www.census.gov/topics/population.html. [Accessed June 10, 2015].
      ]. Scenario analyses (see Appendix Table 3 in Supplemental Materials found at doi:10.1016/j.jval.2016.09.2401) examined cumulative savings over 30, 60, and 90 days, for hospitals covering 50, 100, 500, and 1000 patients with severe COPD.

      Base-Case and Scenario Analyses: Payer Perspective

      The payer base case (Table 1) was defined as 100,000 patients with severe COPD, on the basis of available US data regarding payer lives covered and COPD prevalence [
      • Buist A.S.
      • McBurnie M.A.
      • Vollmer W.M.
      • et al.
      International variation in the prevalence of COPD (the BOLD Study): a population-based prevalence study.
      ,
      Fast Facts About Kaiser Permanente. Available from: http://share.kaiserpermanente.org/article/fast-facts-about-kaiser-permanente/. [Accessed June 10, 2015].
      ,
      Aetna Facts. Available from: https://www.aetna.com/about-us/aetna-facts-and-subsidiaries/aetna-facts.html. [Accessed June 10, 2015].
      ]. Scenarios examined cumulative savings during years 0 to 3 for payers covering 25,000, 50,000, 250,000, and 500,000 patients with severe COPD.

      Sensitivity Analyses

      One-way sensitivity analysis was performed for hospital and payer perspectives by individually varying model parameters using reasonable lower and upper bounds for each parameter (Table 2). Lower and upper bounds were based on CMS reimbursement rates over a 3-year period before and including the year of the base-case value for all reimbursement amounts [
      Centers for Medicare & Medicaid Services
      Medicare Charge In-Patient Summary. National and State Summary.
      ,
      Centers for Medicare & Medicaid Services
      2015 Fee Schedule.
      ,
      Centers for Medicare & Medicaid Services
      2015 DME Fee Schedule..
      ,
      Centers for Medicare & Medicaid Services
      2014 Fee Schedule.
      ,
      Centers for Medicare & Medicaid Services
      2013 Fee Schedule.
      ,
      Centers for Medicare & Medicaid Services
      2014 DME Fee Schedule.
      ,
      Centers for Medicare & Medicaid Services
      2013 DME Fee Schedule..
      ,
      Centers for Medicare & Medicaid Services
      Medicare Charge In-Patient Summary. National and State Summary.
      ,
      Centers for Medicare & Medicaid Services
      Medicare Charge In-Patient Summary. National and State Summary.
      ] and on published values from studies that were well powered (sample size ≥ 90) for mortality, dropout rates, and cost of admission, when available [
      • Struik F.M.
      • Sprooten R.T.
      • Kerstjens H.A.
      • et al.
      Nocturnal non-invasive ventilation in COPD patients with prolonged hypercapnia after ventilatory support for acute respiratory failure: a randomised, controlled, parallel-group study.
      ,
      • Kohnlein T.
      • Windisch W.
      • Kohler D.
      • et al.
      Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial.
      ,
      • Coughlin S.
      • Liang W.E.
      • Parthasarathy S.
      Retrospective assessment of home ventilation to reduce rehospitalization in chronic obstructive pulmonary disease.
      ,
      • Ford E.S.
      Hospital discharges, readmissions, and ED visits for COPD or bronchiectasis among US adults: findings from the Nationwide Inpatient Sample 2001–2012 and Nationwide Emergency Department Sample 2006–2011.
      ,
      • Budweiser S.
      • Hitzl A.P.
      • Jorres R.A.
      • et al.
      Impact of noninvasive home ventilation on long-term survival in chronic hypercapnic COPD: a prospective observational study.
      ]. Appropriate studies for variations in RAD mortality rate, readmission rates, and other parameters were not found; these were varied ±20%. The output was savings from reduced 30-day readmission for the hospital perspective, and 3-year cumulative savings for the payer perspective. Sensitivity analysis was performed using @RISK software (Palisade Corp., Ithaca, NY) in the advanced sensitivity analysis mode.
      Table 2Parameters for one-way sensitivity analysis
      Model inputMinimumMaximum
      Number of patients with severe COPD
      Severe COPD defined as GOLD stages II–IV, Bode Index Score of ≥5, and at least one of the following: PaCO2 ≥ 52 mm Hg, PaO2 ≤ 60 mm Hg, or FEV1 ≤ 40%.
      eligible for home NIV
       Hospital, n200300
       Payer, n80,000120,000
       Event frequency (30-d hospital readmission rate)5.52%8.28%
      No NIV
       1-y mortality rate29%33%
       Admissions per patient per year, n2.483.72
      RAD
       1-y mortality rate23.8%35.6%
       Admissions per patient per year, n2.483.72
       Dropout rate11.1%24.8%
       Patients requiring sleep study80%100%
      Advanced NIV
       1-y mortality rate12%18.3%
       Admissions per patient per year, n0.130.19
       Dropout rate11.1%24.8%
       Admission costs$7706$7945.71
       Hospital reimbursement
        Admission (outside of 30 d)$6971.53$7569.69
        Sleep study$651.52$677.40
      Device reimbursement
       RAD
        Device (per month, capped at 13 mo)$460.69$475
        Reimbursement cap10.4 mo14.3 mo
        Supplies (per year)$462$469
       Advanced NIV
        Device (per month; no cap)$1552.72$1592.14
      COPD, chronic obstructive pulmonary disease; FEV1, forced expiratory volume in 1 s; GOLD, Global Initiative for Chronic Obstructive Lung Disease; NIV, noninvasive ventilation; RAD, respiratory assist device.
      low asterisk Severe COPD defined as GOLD stages II–IV, Bode Index Score of ≥5, and at least one of the following: PaCO2 ≥ 52 mm Hg, PaO2 ≤ 60 mm Hg, or FEV1 ≤ 40%.

      Results

      Hospital Perspective

      Because of the same admission frequency used across no NIV and RAD (bilevel pressure capacity with backup rate [HCPCS code E0471]; see Appendix Table 1 in Supplemental Materials), base-case, sensitivity, and scenario analyses were impacted in an identical manner across both no NIV and RAD comparators, and the results described hereafter are applicable for Advanced NIV (pressure support ventilator with volume control mode [Trilogy100; HCPCS code E0464]) versus either comparator.
      For the hospital base case (250 patients with severe COPD), the total cumulative saving associated with the use of in-home Advanced NIV versus either comparator was $402,981 for the period of 0 to 30 days (Table 3). This saving resulted from a reduction in admissions during the 30-day period, from n = 53 to n = 3. Cumulative savings for Advanced NIV versus either comparator were $426,041 for 0 to 60 days and $449,101 for 0 to 90 days. Thus, cost analysis favored Advanced NIV versus either comparator.
      Table 3Base-case economic outcomes, hospital and payer perspectives
      Advanced NIV vs. no NIV or RADDays 0–30Days 31–60Days 61–90
      Hospital perspective (n = 250 patients with severe COPD
      Severe COPD defined as GOLD stages II–IV, Bode Index Score of ≥5, and at least one of the following: PaCO2 ≥ 52 mm Hg, PaO2 ≤ 60 mm Hg, or FEV1 ≤ 40%.
      )
      No NIV or RAD
       Cost of admissions$424,912$513,424$513,424
       Reimbursement$0$489,109$489,109
       Total admissions cost
      Costs of admissions − Reimbursement.
      $424,912$24,315$24,315
      Advanced NIV
       Cost of admissions$21,931$26,499$26,499
       Reimbursement$0$25,244$25,244
       Total admissions cost
      Costs of admissions − Reimbursement.
      $21,931$1,255$1,255
      Admissions savings
      Total admissions cost (Advanced NIV) − Total admissions cost (no NIV or RAD).
      $402,981$23,060$23,060
      Cumulative admissions savings
      Sum of admissions savings for present and all preceding time frames examined.
      $402,981$426,041$449,101
      Payer perspective (n = 100,000 patients with severe COPD)
      Advanced NIV vs. no NIVYear 0Year 1Year 2Year 3
      No NIV
       Total annual cost
      Includes reimbursement for hospital admissions, sleep study, device acquisition and repair, device, and supplies.
      $0$2,184,688,231$1,551,128,644$1,101,301,337
      Advanced NIV
       Total annual cost
      Includes reimbursement for hospital admissions, sleep study, device acquisition and repair, device, and supplies.
      $0$2,023,326,102$1,368,967,427$1,118,442,144
      Annual savings
      Total annual cost (Advanced NIV) − Total annual cost (no NIV or RAD).
      $0$161,362,129$182,161,217($17,140,807)
      Cumulative savings
      Sum of admissions savings for present and all preceding time frames examined.
      $0$161,362,129$343,523,345$326,382,539
      Advanced NIV vs. RADYear 0Year 1Year 2Year 3
      RAD
       Total annual cost
      $462 based on the assumption that the payer is responsible for payment of 80% of total supplies reimbursement amount ($578) listed in the CMS 2015 DME Fee Schedule.
      $65,717,000$2,800,925,031$1,589,739,548$1,098,385,929
      Advanced NIV
       Total annual cost
      $462 based on the assumption that the payer is responsible for payment of 80% of total supplies reimbursement amount ($578) listed in the CMS 2015 DME Fee Schedule.
      $0$2,023,326,102$1,368,967,427$1,118,442,144
      Annual savings
      Total annual cost (Advanced NIV) − Total annual cost (no NIV or RAD).
      $65,717,000$777,598,929$220,772,121($20,056,215)
      Cumulative savings
      Includes humidification (water chamber) at $18.29 (2 times/year), filters (ultrafine, disposable, 2-pack) at $4.74 (6 times/year), NIV mask (full face) at $180.47 (2 times/year), tubing (System One Performance Tubing, 15 mm) at $38.00 (4 times/year).
      $65,717,000$843,315,929$1,064,088,049$1,044,031,835
      Note. Outcomes account for mortality and dropout rates in patients; ( ) indicates a negative number.
      COPD, chronic obstructive pulmonary disease; FEV1, forced expiratory volume in 1 s; GOLD, Global Initiative for Chronic Obstructive Lung Disease; NIV, noninvasive ventilation; RAD, respiratory assist device.
      low asterisk Severe COPD defined as GOLD stages II–IV, Bode Index Score of ≥5, and at least one of the following: PaCO2 ≥ 52 mm Hg, PaO2 ≤ 60 mm Hg, or FEV1 ≤ 40%.
      Costs of admissions − Reimbursement.
      Total admissions cost (Advanced NIV) − Total admissions cost (no NIV or RAD).
      § Sum of admissions savings for present and all preceding time frames examined.
      || Includes reimbursement for hospital admissions, sleep study, device acquisition and repair, device, and supplies.
      Total annual cost (Advanced NIV) − Total annual cost (no NIV or RAD).
      One-way sensitivity analysis of 0- to 30-day savings associated with Advanced NIV use revealed that the number of admissions per patient per year (patients receiving no NIV or RAD), the number of patients with severe COPD, and the 30-day readmission rate all had a similar effect on savings over this period (Fig. 1A). In each case, an increase in the input value resulted in increased savings with Advanced NIV use. Conversely, an increase in the number of admissions per patient per year among patients receiving Advanced NIV resulted in decreased savings; this change, however, was relatively small ($398,595 in savings, vs. $402,981 for base case). As noted earlier, sensitivity analysis of savings associated with the use of Advanced NIV versus RAD produced the same results as for Advanced NIV versus no NIV.
      Fig. 1
      Fig. 1One-way sensitivity analysis of cumulative savings associated with NIV use. Sensitivity analysis was carried out by independently modifying individual model inputs. (A) 30-day readmissions savings, from the hospital perspective, when comparing Advanced NIV with no NIV or RAD. (B) 3-year cumulative savings, from the payer perspective, when comparing Advanced NIV with no NIV. (C) 3-year cumulative savings, from the payer perspective, when comparing Advanced NIV with RAD. Reimbursement rate corresponds to the rate for admissions outside 0–30 days relative to the initial admission. Green bars represent savings associated with an increase in the value of a given model input. Red bars represent savings associated with a decrease in the value of a given model input. COPD, chronic obstructive pulmonary disease; NIV, noninvasive ventilation (device reimbursement $1592/mo, with no reimbursement cap); RAD, respiratory assist device (bilevel pressure capacity with backup rate; device reimbursement $475/mo, with reimbursement capped at 13 months; supplies reimbursement $462/y).
      The defined scenarios involving the use of Advanced NIV resulted in hospital savings over 90 days, with the greatest savings observed over the 0- to 30-day period (Fig. 2A). The greatest cumulative savings were realized with the largest COPD cohort examined (1000 patients with severe COPD: $1.6 million over 0–30 days, $1.8 million over 0–90 days, vs. no NIV or RAD). Even with a cohort of 50 patients with severe COPD, sizeable cumulative savings were observed ($80,596 for 0–30 days and $89,820 for 0–90 days, vs. either comparator).
      Fig. 2
      Fig. 2Cumulative savings scenario analyses. Scenario analyses examined cumulative savings, from the hospital and payer perspectives, associated with varying number of patients with COPD (all other base-case parameters constant). (A) Cumulative savings for 0–30 days, 31–60 days, and 61–90 days, from the hospital perspective, when comparing Advanced NIV with no NIV or RAD. (B) Cumulative savings for years 0–3, from the payer perspective, when comparing Advanced NIV with no NIV. (C) Cumulative savings for years 0–3, from the payer perspective, when comparing Advanced NIV with RAD. COPD, chronic obstructive pulmonary disease; NIV, noninvasive ventilation (device reimbursement $1592/mo, with no reimbursement cap); RAD, respiratory assist device (bilevel pressure capacity with backup rate; device reimbursement $475/mo, with reimbursement capped at 13 months; supplies reimbursement $462/y).

      Payer Perspective

      Advanced NIV resulted in reductions in annual admissions of patients with severe COPD. Using the base case of a payer covering 100,000 patients with severe COPD, Advanced NIV was associated with 16,000 year 1 admissions and 10,679 year 3 admissions, as compared with 310,000 year 1 admissions and 156,271 year 3 admissions among patients receiving no NIV (see Appendix Figure 1 in Supplemental Materials found at doi:10.1016/j.jval.2016.09.2401). There was also a reduction in admissions versus RAD, which was associated with 310,000 and 153,205 admissions in years 1 and 3, respectively. In addition, Advanced NIV was associated with increased survival versus no NIV or RAD; this increase, however, was modest. Use of Advanced NIV resulted in 54,534 surviving patients in year 3, whereas no NIV and RAD resulted in 35,792 and 34,743 surviving patients, respectively, in year 3.
      For a payer covering 100,000 patients with severe COPD, total cumulative savings with Advanced NIV versus no NIV were $161 million for year 1 and $326 million through year 3, whereas savings with Advanced NIV versus RAD were $843 million by year 1 and $1.04 billion by year 3 (Table 3). Use of Advanced NIV resulted in reduced annual costs for year 1 (7% reduction vs. no NIV; 28% reduction vs. RAD) and year 2 (12% reduction vs. no NIV; 14% reduction vs. RAD), and increased costs for year 3 (2% increase vs. no NIV and RAD).
      One-way sensitivity analysis of savings for Advanced NIV versus either no NIV or RAD revealed that 3-year cumulative savings were most strongly impacted by the number of admissions per patient per year for patients receiving either comparator (Figs. 1B and 1C). This was followed by the dropout rate for Advanced NIV, hospital admissions reimbursement for either comparator, number of patients with COPD (vs. RAD only), and mortality rates for Advanced NIV and either comparator. Although an increase in most of these parameters was associated with greater cumulative 3-year savings, changes in the comparator mortality rates had an opposite effect.
      Scenario analyses revealed that the greatest 3-year cumulative savings were realized with the largest covered COPD population examined (500,000 patients: $1.63 billion vs. no NIV and $5.22 billion vs. RAD; Figs. 2B and 2C). Savings associated with the smallest covered COPD cohort tested (25,000 patients) remained sizeable, with 3-year cumulative savings of $82 million versus no NIV and $261 million versus RAD.

      Discussion

      Management of COPD imposes a substantial financial burden on both hospitals and payers [
      Global Initiative for Chronic Obstructive Lung Disease. Global Strategy for the Diagnosis, Management and Prevention of COPD.
      ,
      • Dalal A.A.
      • Shah M.
      • D’Souza A.O.
      • et al.
      Costs of COPD exacerbations in the emergency department and inpatient setting.
      ,
      • Chapman K.R.
      • Mannino D.M.
      • Soriano J.B.
      • et al.
      Epidemiology and costs of chronic obstructive pulmonary disease.
      ]. A recent study found that among COPD discharges, COPD itself was the most common cause for readmissions and that rehospitalization frequency was significantly reduced for patients discharged to home with home care or to a skilled-nursing facility [
      • Shah T.
      • Churpek M.M.
      • Coca Perraillon M.
      • et al.
      Understanding why patients with COPD get readmitted: a large national study to delineate the Medicare population for the readmissions penalty expansion.
      ]. This supports the important role of home care in COPD and suggests potential benefits of novel technologies aimed at reducing exacerbations and rehospitalization.

      Hospital Perspective

      Rehospitalization of patients with COPD represents an important area of concern for health care providers. For hospitals, improving financial outcomes related to managing patients with COPD is expected to rely increasingly on reducing readmissions, and thus avoid reimbursement penalties [
      • Feemster L.C.
      • Au D.H.
      Penalizing hospitals for chronic obstructive pulmonary disease readmissions.
      ]. The present analysis revealed that in-home Advanced NIV had a positive economic impact from the hospital perspective, on the basis of reduced readmission rates associated with the implementation of this technology. Savings were observed for the base case (Table 1) as well as for all scenarios examining different COPD cohort sizes. Although reimbursement adjustments are calculated under the HRRP on the basis of excess readmission ratios, the present model calculated readmission-associated savings assuming no reimbursement for any 30-day COPD readmission. Furthermore, it is important to note that HRRP at present penalizes 30-day COPD readmissions. Nevertheless, expansion of the program to 60-day and 90-day readmissions, for example, could place greater emphasis on COPD readmission reductions.

      Payer Perspective

      For payers, hospital admissions for COPD represent a large coverage expense. On the basis of previously reported data [
      • Struik F.M.
      • Sprooten R.T.
      • Kerstjens H.A.
      • et al.
      Nocturnal non-invasive ventilation in COPD patients with prolonged hypercapnia after ventilatory support for acute respiratory failure: a randomised, controlled, parallel-group study.
      ,
      • Kohnlein T.
      • Windisch W.
      • Kohler D.
      • et al.
      Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial.
      ,
      • Coughlin S.
      • Liang W.E.
      • Parthasarathy S.
      Retrospective assessment of home ventilation to reduce rehospitalization in chronic obstructive pulmonary disease.
      ], the present analysis estimated a sizeable reduction in admissions for years 1 through 3, which resulted in significant savings for all scenarios examined. It is worth noting that, in the model, cumulative savings with Advanced NIV tended to peak in year 2 ($1.06 billion vs. RAD), with an approximately 2% decline in year 3 ($1.04 billion vs. RAD). There are two reasons for this finding: first, the model reflects present practice in which Advanced NIV device reimbursement is uncapped and RAD is reimbursed for up to 13 months [
      • Sunwoo B.Y.
      • Mulholland M.
      • Rosen I.M.
      • et al.
      The changing landscape of adult home noninvasive ventilation technology, use, and reimbursement in the United States.
      ] (contributing to the year 2 peak), and second, in addition to the reduction in admissions, there was a concomitant decrease in mortality (34.4% reduction vs. no NIV, 36.3% reduction vs. RAD, for year 3) and a concomitant increase in the number of covered patients over time, impacting savings negatively in year 3. The 13-month RAD reimbursement cap also contributed to the large difference in year 1 cost savings observed with the use of Advanced NIV versus RAD (28% reduction in costs) compared with Advanced NIV versus no NIV (7% reduction in costs). This difference was observed only in year 1 because the first 12 months of the 13-month cap constituted the bulk of RAD reimbursement. No such difference was seen in years 2 and 3 because both RAD and no NIV received no device-related reimbursement during this period, with the exception of the first month of year 2, as part of the 13-month RAD reimbursement cap.
      Despite the effects of mortality and reimbursement caps, reduction in admissions over this period (93.2% reduction vs. no NIV, 93.0% reduction vs. RAD, over 3 years) drove considerable savings, in line with data identifying admission costs as the key driver of COPD patient coverage costs [
      • Dalal A.A.
      • Chiristensen L.
      • Liu F.
      • et al.
      Direct costs of chronic obstructive pulmonary disease among managed care patients.
      ]. Admissions-related savings also greatly outweighed differences in the monthly reimbursement amounts for NIV (HCPCS code E0464, $1592/month) and RAD (HCPCS code E0471, $475/month; plus $462/year for supplies). On the basis of the sensitivity analysis for Advanced NIV versus RAD, lower Advanced NIV dropout rates corresponded to reduced savings because of associated costs of a larger covered patient population, versus RAD dropout rates, which had an opposite and smaller impact because of the capped device-related reimbursement.
      Device reimbursement reflected present amounts; proposed decreases in reimbursement [
      • Budweiser S.
      • Hitzl A.P.
      • Jorres R.A.
      • et al.
      Impact of noninvasive home ventilation on long-term survival in chronic hypercapnic COPD: a prospective observational study.
      ], however, could result in additional savings for payers. Similarly, although there is at present no cap on E0464 device reimbursement (i.e., Advanced NIV), hypothetical implementation of a cap would be expected to further reduce payer costs. It is important to note, however, that reduction in reimbursement would likely have a negative impact on patient clinical outcomes. A reduction in reimbursement would potentially shift the burden of payment to the patient, resulting in reduced adherence; an inferior device may be substituted or services cut by the equipment provider, or patients may prolong resupply, resulting in lapses in device use. The introduction of a reimbursement cap may decrease the treatment time period, which may, in turn, have an economic impact and influence realized savings.

      Perspective of Other Health Care Providers

      The results from this study also have implications for home care providers supplying durable medical equipment (DME) as part of disease management programs, which can reduce COPD-associated readmissions [
      • Carlin B.
      • Wiles K.
      • Easley D.
      • et al.
      Transition of care and rehospitalization rates for patients with COPD and CHF who require supplemental oxygen therapy.
      ]. Advanced NIV adoption to support these programs would be expected to benefit DME providers. This, in turn, is relevant to hospitals in light of HRRP, and calls for more coordinated efforts between hospitals and DME providers [
      • Messenger R.W.
      Reducing chronic obstructive pulmonary disease readmissions: the role of the durable medical equipment provider.
      ]. Furthermore, the home Advanced NIV technology may provide substantial benefits that are aligned to the needs of accountable care organizations and integrated delivery network health care models, which aim to improve care coordination and reduce costs via, among other measures, reductions in acute care episodes and readmissions [
      • Windisch W.
      • Haenel M.
      • Storre J.H.
      • et al.
      High-intensity non-invasive positive pressure ventilation for stable hypercapnic COPD.
      ,
      • Adamson M.
      How ACOs can prevent avoidable hospital admissions.
      ,
      • Spencer G.
      Making the move to an ACO.
      ].

      Impact of Supportive Home Care Interventions

      The quality improvement initiative used to define readmission and mortality rates for Advanced NIV [
      • Coughlin S.
      • Liang W.E.
      • Parthasarathy S.
      Retrospective assessment of home ventilation to reduce rehospitalization in chronic obstructive pulmonary disease.
      ] used Advanced NIV as part of a multifaceted program including oxygen therapy, medication reconciliation, patient education, and RT-led respiratory care. Comparatively, studies used to derive readmission and mortality rates for no NIV and RAD [
      • Struik F.M.
      • Sprooten R.T.
      • Kerstjens H.A.
      • et al.
      Nocturnal non-invasive ventilation in COPD patients with prolonged hypercapnia after ventilatory support for acute respiratory failure: a randomised, controlled, parallel-group study.
      ,
      • Kohnlein T.
      • Windisch W.
      • Kohler D.
      • et al.
      Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial.
      ] were accompanied by oxygen therapy and pharmacotherapy. The potential contribution of patient education and RT-led care to the positive outcomes seen with Advanced NIV use should hence be considered. Patient education (including medication counseling and smoking cessation) alone has not been found effective in reducing COPD-associated hospitalizations [
      • Jennings J.H.
      • Thavarajah K.
      • Mendez M.
      • et al.
      Predischarge bundle for patients with acute exacerbations of chronic obstructive pulmonary disease to reduce readmissions and emergency department visits: a randomized controlled trial.
      ]. In comparison, the impact of RT-led or specialist-led home intervention is less clear. Individualized treatment action plans managed by specialty nurse case managers [
      • Trappenburg J.C.
      • Monninkhof E.M.
      • Bourbeau J.
      • et al.
      Effect of an action plan with ongoing support by a case manager on exacerbation-related outcome in patients with COPD: a multicentre randomised controlled trial.
      ] or respiratory nurses [
      • Sridhar M.
      • Taylor R.
      • Dawson S.
      • et al.
      A nurse led intermediate care package in patients who have been hospitalised with an acute exacerbation of chronic obstructive pulmonary disease.
      ], with telephone sessions and/or home visits, did not impact exacerbation rates or health care utilization among patients with COPD. In contrast, more intensive home care interventions involving integrated specialist and home care nurse efforts [
      • Titova E.
      • Steinshamn S.
      • Indredavik B.
      • et al.
      Long term effects of an integrated care intervention on hospital utilization in patients with severe COPD: a single centre controlled study.
      ] or a pulmonary specialty team comprising generalist and specialty nurses, therapists, and home care aides [
      • Steinel J.A.
      • Madigan E.A.
      Resource utilization in home health chronic obstructive pulmonary disease management.
      ] resulted in reduced hospital and emergency department admission rates. RT-led care in the recent initiative [
      • Coughlin S.
      • Liang W.E.
      • Parthasarathy S.
      Retrospective assessment of home ventilation to reduce rehospitalization in chronic obstructive pulmonary disease.
      ] was more similar to the former home care interventions [
      • Trappenburg J.C.
      • Monninkhof E.M.
      • Bourbeau J.
      • et al.
      Effect of an action plan with ongoing support by a case manager on exacerbation-related outcome in patients with COPD: a multicentre randomised controlled trial.
      ,
      • Sridhar M.
      • Taylor R.
      • Dawson S.
      • et al.
      A nurse led intermediate care package in patients who have been hospitalised with an acute exacerbation of chronic obstructive pulmonary disease.
      ] in intensity of intervention, suggesting that the observed clinical outcomes may be attributable to the addition of Advanced NIV. Future studies are needed to understand how variations in program characteristics and devices affect cost savings across a range of hospital and payer systems.

      Study Limitations

      Limitations of this analysis include the use of a set of assumptions and data sources that may not be generalizable to all institutions and the limited set of scenarios tested. Furthermore, outcome rates for the Trilogy100 NIV device [
      • Coughlin S.
      • Liang W.E.
      • Parthasarathy S.
      Retrospective assessment of home ventilation to reduce rehospitalization in chronic obstructive pulmonary disease.
      ] may not be reflective of outcome rates with other home NIV approaches.
      Another potential limitation of the quality improvement program [
      • Coughlin S.
      • Liang W.E.
      • Parthasarathy S.
      Retrospective assessment of home ventilation to reduce rehospitalization in chronic obstructive pulmonary disease.
      ] was that it was a single-center initiative evaluating a severe COPD population at risk for rehospitalization. Although all patients with COPD who may benefit from Advanced NIV as part of a multifaceted program may not have been included, the model focused on those representing a substantial burden to hospitals and payers who would be expected to benefit most from such an approach.
      The model also did not account for differences in disease severity (GOLD stage III vs. IV), comorbidities, reason for admission, length of stay, or discharge location within the quality improvement program [
      • Coughlin S.
      • Liang W.E.
      • Parthasarathy S.
      Retrospective assessment of home ventilation to reduce rehospitalization in chronic obstructive pulmonary disease.
      ]. In addition, differences in inclusion/exclusion criteria and device treatment regimen (including usage time and device settings) between the quality improvement program [
      • Coughlin S.
      • Liang W.E.
      • Parthasarathy S.
      Retrospective assessment of home ventilation to reduce rehospitalization in chronic obstructive pulmonary disease.
      ] and the comparator studies [
      • Struik F.M.
      • Sprooten R.T.
      • Kerstjens H.A.
      • et al.
      Nocturnal non-invasive ventilation in COPD patients with prolonged hypercapnia after ventilatory support for acute respiratory failure: a randomised, controlled, parallel-group study.
      ,
      • Kohnlein T.
      • Windisch W.
      • Kohler D.
      • et al.
      Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial.
      ] were not considered. Still, the estimated savings suggest important benefits associated with implementation of home Advanced NIV for severe COPD.

      Conclusions

      This analysis provides evidence for the economic benefits of home Advanced NIV as part of a multifaceted intervention program including oxygen therapy, RT-led respiratory care, patient education, and medication reconciliation, with hospital savings driven by reduced 30-day readmissions and payer savings driven by admission reductions over 3 years. Adoption of such innovative technologies that can reduce readmissions is congruent with the objectives of recent policies, such as HRRP, and new health care models, such as accountable care organizations, all of which aim to improve the quality of care while reducing high health care costs.

      Acknowledgments

      We thank Ms Amy Day for providing access to the quality improvement program data and Dr Sairam Parthasarathy for data analysis and interpretation.
      Source of financial support: Funding for the design of the model, analysis, and manuscript preparation was provided by Philips Respironics (Monroeville, PA).

      Supplemental Materials

      References

        • Pauwels R.A.
        • Buist A.S.
        • Calverley P.M.
        • et al.
        Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease. NHLBI/WHO Global Initiative for Chronic Obstructive Lung Disease (GOLD) Workshop summary.
        Am J Respir Crit Care Med. 2001; 163: 1256-1276
        • Celli B.R.
        • MacNee W.
        • Force A.E.T.
        Standards for the diagnosis and treatment of patients with COPD: a summary of the ATS/ERS position paper.
        Eur Respir J. 2004; 23: 932-946
      1. Global Initiative for Chronic Obstructive Lung Disease. Global Strategy for the Diagnosis, Management and Prevention of COPD.
        Global Initiative for Chronic Obstructive Lung Disease. 2015; (Available from: http://goldcopd.org/. [Accessed October 14, 2016])
        • Dalal A.A.
        • Shah M.
        • D’Souza A.O.
        • et al.
        Costs of COPD exacerbations in the emergency department and inpatient setting.
        Respir Med. 2011; 105: 454-460
        • Chapman K.R.
        • Mannino D.M.
        • Soriano J.B.
        • et al.
        Epidemiology and costs of chronic obstructive pulmonary disease.
        Eur Respir J. 2006; 27: 188-207
        • Breunig I.M.
        • Shaya F.T.
        • Scharf S.M.
        Delivering cost-effective care for COPD in the USA: recent progress and current challenges.
        Expert Rev Pharmacoecon Outcomes Res. 2012; 12: 725-731
        • Jennings J.H.
        • Thavarajah K.
        • Mendez M.
        • et al.
        Predischarge bundle for patients with acute exacerbations of chronic obstructive pulmonary disease to reduce readmissions and emergency department visits: a randomized controlled trial.
        Chest. 2015; 147: 1227-1234
        • Shah T.
        • Churpek M.M.
        • Coca Perraillon M.
        • et al.
        Understanding why patients with COPD get readmitted: a large national study to delineate the Medicare population for the readmissions penalty expansion.
        Chest. 2015; 147: 1219-1226
        • Feemster L.C.
        • Au D.H.
        Penalizing hospitals for chronic obstructive pulmonary disease readmissions.
        Am J Respir Crit Care Med. 2014; 189: 634-639
        • Dalal A.A.
        • Chiristensen L.
        • Liu F.
        • et al.
        Direct costs of chronic obstructive pulmonary disease among managed care patients.
        Int J Chron Obstruct Pulmon Dis. 2010; 5: 341-349
        • Pierson D.J.
        History and epidemiology of noninvasive ventilation in the acute-care setting.
        Respir Care. 2009; 54: 40-52
        • Casanova C.
        • Celli B.R.
        • Tost L.
        • et al.
        Long-term controlled trial of nocturnal nasal positive pressure ventilation in patients with severe COPD.
        Chest. 2000; 118: 1582-1590
        • Clini E.
        • Sturani C.
        • Rossi A.
        • et al.
        The Italian multicentre study on noninvasive ventilation in chronic obstructive pulmonary disease patients.
        Eur Respir J. 2002; 20: 529-538
        • McEvoy R.D.
        • Pierce R.J.
        • Hillman D.
        • et al.
        Nocturnal non-invasive nasal ventilation in stable hypercapnic COPD: a randomised controlled trial.
        Thorax. 2009; 64: 561-566
        • Struik F.M.
        • Sprooten R.T.
        • Kerstjens H.A.
        • et al.
        Nocturnal non-invasive ventilation in COPD patients with prolonged hypercapnia after ventilatory support for acute respiratory failure: a randomised, controlled, parallel-group study.
        Thorax. 2014; 69: 826-834
        • Tsolaki V.
        • Pastaka C.
        • Karetsi E.
        • et al.
        One-year non-invasive ventilation in chronic hypercapnic COPD: effect on quality of life.
        Respir Med. 2008; 102: 904-911
        • Kohnlein T.
        • Windisch W.
        • Kohler D.
        • et al.
        Non-invasive positive pressure ventilation for the treatment of severe stable chronic obstructive pulmonary disease: a prospective, multicentre, randomised, controlled clinical trial.
        Lancet Respir Med. 2014; 2: 698-705
        • Coughlin S.
        • Liang W.E.
        • Parthasarathy S.
        Retrospective assessment of home ventilation to reduce rehospitalization in chronic obstructive pulmonary disease.
        J Clin Sleep Med. 2015; 11: 663-670
      2. Centers for Medicare & Medicaid Services. Local Coverage Determination (LCD): Respiratory Assist Devices. Baltimore, MD: Centers for Medicare & Medicaid Services, 2015.
        • Ford E.S.
        Hospital discharges, readmissions, and ED visits for COPD or bronchiectasis among US adults: findings from the Nationwide Inpatient Sample 2001–2012 and Nationwide Emergency Department Sample 2006–2011.
        Chest. 2015; 147: 989-998
        • Windisch W.
        • Kostic S.
        • Dreher M.
        • et al.
        Outcome of patients with stable COPD receiving controlled noninvasive positive pressure ventilation aimed at a maximal reduction of Pa(CO2).
        Chest. 2005; 128: 657-662
        • Windisch W.
        • Quality of life in home mechanical ventilation study group
        Impact of home mechanical ventilation on health-related quality of life.
        Eur Respir J. 2008; 32: 1328-1336
        • Centers for Medicare & Medicaid Services
        Medicare Charge In-Patient Summary. National and State Summary.
        Centers for Medicare & Medicaid Services, Baltimore, MD2012
        • Centers for Medicare & Medicaid Services
        2015 Fee Schedule.
        Centers for Medicare & Medicaid Services, Baltimore, MD2015
        • Centers for Medicare & Medicaid Services
        2015 DME Fee Schedule..
        Centers for Medicare & Medicaid Services, Baltimore, MD2015
      3. United States Bureau of Labor and Statistics. CPI inflation calculator. Available from: http://www.bls.gov/data/inflation_calculator.htm. [Accessed February 1, 2015]

      4. Gay PC. Nocturnal ventilatory support in COPD. In: Stoller JK, ed., UpToDate. Published by Wolters Kluwer, Alphen aan den Rijn, Netherlands, 2015.

      5. Mount Sinai Health System. Available from: http://www.bethisraelny.org/brooklyn/. [Accessed June 10, 2015].
      6. The World Bank Data. Available from: http://data.worldbank.org/indicator/SH.MED.BEDS.ZS. [Accessed June 10, 2015].
      7. United States Census Bureau—Population. Available from: http://www.census.gov/topics/population.html. [Accessed June 10, 2015].
        • Buist A.S.
        • McBurnie M.A.
        • Vollmer W.M.
        • et al.
        International variation in the prevalence of COPD (the BOLD Study): a population-based prevalence study.
        Lancet. 2007; 370: 741-750
      8. Fast Facts About Kaiser Permanente. Available from: http://share.kaiserpermanente.org/article/fast-facts-about-kaiser-permanente/. [Accessed June 10, 2015].
      9. Aetna Facts. Available from: https://www.aetna.com/about-us/aetna-facts-and-subsidiaries/aetna-facts.html. [Accessed June 10, 2015].
        • Centers for Medicare & Medicaid Services
        2014 Fee Schedule.
        Centers for Medicare & Medicaid Services, Baltimore, MD2014
        • Centers for Medicare & Medicaid Services
        2013 Fee Schedule.
        Centers for Medicare & Medicaid Services, Baltimore, MD2013
        • Centers for Medicare & Medicaid Services
        2014 DME Fee Schedule.
        Centers for Medicare & Medicaid Services, Baltimore, MD2014
        • Centers for Medicare & Medicaid Services
        2013 DME Fee Schedule..
        Centers for Medicare & Medicaid Services, Baltimore, MD2013
        • Centers for Medicare & Medicaid Services
        Medicare Charge In-Patient Summary. National and State Summary.
        Centers for Medicare & Medicaid Services, Baltimore, MD2013
        • Centers for Medicare & Medicaid Services
        Medicare Charge In-Patient Summary. National and State Summary.
        Centers for Medicare & Medicaid Services, Baltimore, MD2014
        • Budweiser S.
        • Hitzl A.P.
        • Jorres R.A.
        • et al.
        Impact of noninvasive home ventilation on long-term survival in chronic hypercapnic COPD: a prospective observational study.
        Int J Clin Pract. 2007; 61: 1516-1522
        • Sunwoo B.Y.
        • Mulholland M.
        • Rosen I.M.
        • et al.
        The changing landscape of adult home noninvasive ventilation technology, use, and reimbursement in the United States.
        Chest. 2014; 145: 1134-1140
        • Carlin B.
        • Wiles K.
        • Easley D.
        • et al.
        Transition of care and rehospitalization rates for patients with COPD and CHF who require supplemental oxygen therapy.
        Chest. 2011; 140: 550A
        • Messenger R.W.
        Reducing chronic obstructive pulmonary disease readmissions: the role of the durable medical equipment provider.
        Prof Case Manag. 2012; 17 (quiz 15–6): 109-114
        • Windisch W.
        • Haenel M.
        • Storre J.H.
        • et al.
        High-intensity non-invasive positive pressure ventilation for stable hypercapnic COPD.
        Int J Med Sci. 2009; 6: 72-76
        • Adamson M.
        How ACOs can prevent avoidable hospital admissions.
        Accountable Care News. 2013; 4: 1-3
        • Spencer G.
        Making the move to an ACO.
        Healthc Financ Manage. 2014; 68: 92-96
        • Trappenburg J.C.
        • Monninkhof E.M.
        • Bourbeau J.
        • et al.
        Effect of an action plan with ongoing support by a case manager on exacerbation-related outcome in patients with COPD: a multicentre randomised controlled trial.
        Thorax. 2011; 66: 977-984
        • Sridhar M.
        • Taylor R.
        • Dawson S.
        • et al.
        A nurse led intermediate care package in patients who have been hospitalised with an acute exacerbation of chronic obstructive pulmonary disease.
        Thorax. 2008; 63: 194-200
        • Titova E.
        • Steinshamn S.
        • Indredavik B.
        • et al.
        Long term effects of an integrated care intervention on hospital utilization in patients with severe COPD: a single centre controlled study.
        Respir Res. 2015; 16: 8
        • Steinel J.A.
        • Madigan E.A.
        Resource utilization in home health chronic obstructive pulmonary disease management.
        Outcomes Manag. 2003; 7: 23-27