Translational Pathway for Ventricular Assist Devices

Reimbursement

Author

Robin Bostic

Introduction

Establishing reimbursement (the ability to be paid) early in product development is essential for patients to have long-term access to any therapy. Reimbursement is comprised of coverage, coding, and payment, with coverage being the essential first step that drives subsequent coding and payment. Prior to any product launch, the analysis of each of these reimbursement drivers is essential to market success. The primary path to payer coverage of a product, procedure, or service occurs when a product is deemed a “reasonable and necessary” medical treatment. Recently, payers are demanding a further level of evidence to show effectiveness – does it work in the real world? – and efficiency – does it provide a better value?  The earlier the process is initiated to obtain coverage the sooner reimbursement will be established.  When developing a reimbursement plan, the following questions should be asked:

— Where will this product fit in the larger health care arena?

— Will this product meet not only the U.S. Food and Drug Administration (FDA) “safe and effective” standards required for regulatory approval, but also payers’ “reasonable and necessary” and possibly “effective and efficient” requirements?

— How can the reimbursement adequately cover the cost?

If the product is a new and innovative technology, beyond the excitement that may surround the scientific breakthrough, a new reimbursement structure will need to be constructed and implemented to address coverage, coding, and payment.

Wherever the product “fits,” a reasonable timeline for developing and implementing the reimbursement plan must be anticipated. If coverage, coding, and payment already exist for a similar product, 6 to 12 months is currently typical for getting this product positioned within the existing category. To obtain new or modify existing administrative codes used to identify the product or service to payers, whether related to physician procedures – Current Procedural Terminology (CPT) – or to the device itself – Healthcare Common Procedure Coding System (HCPCS) – the process usually takes approximately 1 to 2 years from the product’s launch into the health care market. For new technologies, Medicare has taken 2 to 5 years to create national medical coverage or to substantially expand existing guidelines, and code creation follows along a similar path.

Overview of Mechanical Circulatory Support Reimbursement

It took more than 3 decades for ventricular assist devices (VADs) to be available to patients. In reverse chronological order are the trials leading to regulatory and reimbursement first approval for HeartMate as destination therapy (DT):

— REMATCH trial DT trial (1998-2003)

— PREMATCH trial DT trial (1996-1998)

— HeartMate VE bridge-to-transplant trial (1991-1998)

— HeartMate IP bridge-to-transplant trial (1986-1994)

— Model 7 – Model 10 LVAD trial (1975-1988)

Left ventricular assist devices (LVADs) utilized as DT are an example of an innovative technology involving the creation of coverage, codes, and payment for not only the hospitals but physicians and outpatient supplies. In 2002, the REMATCH (Randomized Evaluation of Mechanical Assistance for the Treatment of Congestive Heart Failure) trial had been completed showing efficacy of the device over medical therapy, but there was no payer coverage, no administrative codes, and no payment for devices used for DT (1). Every reimbursement structure had to be created, including influencing health technology assessments (HTAs) for commercial payer coverage and creating a Medicare national coverage decision (NCD); establishing coding in every system including the International Classification of Diseases, Clinical Modification, 9th Edition codes, CPT, and HCPCS; creating payment under the Medicare inpatient diagnosis-related group (DRG) system; establishing payment during clinical trials; and influencing payment for physician surgical procedures and replacement outpatient accessories and supplies required to support the LVAD patient in the outpatient setting. Table 1 shows these steps for each provider type.

 

Table 1. Pathways of Reimbursement

  Coverage Coding Payment
Inpatient hospital Medicare national coverage determination, influence HTAs for commercial payers

Unique ICD-9-CM procedure code for implant

 

Appropriate DRG assignment

 

Physician services

Appropriate Correct Coding Initiative

Edits for physician payment

 

Create CPT codes for surgical procedure and VAD interrogation

 

Appropriate relative value units assigned to all CPT codes

 

Replacement accessories and supplies Coverage policies with replacement intervals for outpatient accessories and supplies HCPCS codes for outpatient accessories and supplies Blend manufacturer price with market rate for HCPCS payments

CPT = Current Procedural Terminology; DRG = diagnosis-related group; HCPCS = Healthcare Common Procedure Coding System; HTA = health technology assessment; ICD-9-CM = International Classification of Diseases, Clinical Modification, 9th Edition; VAD = ventricular assist device.

Today, VAD therapy is a well-established treatment for end-stage heart failure (HF), with more than 49,000 VAD implants performed in hospitals globally since early 1980s. VADs used for bridge to transplantation (BTT) have been covered by Medicare and most other commercial payers for post-cardiotomy use since 1993 and for BTT since 1996.

United States Reimbursement for LVADs

In October 2003, the Centers for Medicare and Medicaid Services (CMS) issued an NCD for VADs used for DT (2).  This NCD was based on key studies, which showed significant clinical and quality of life benefits for end-stage HF patients who were implanted with VADs for DT.

In the 1990s, third-party payers were familiar with LVAD technologies but did not have coverage for DT. The device was used solely to assist the left ventricle while patients recovered from heart surgery, and in later years, “bridged” patients’ failed hearts to cardiac transplant while on organ waiting lists.

The 2000s saw a different era in LVADs. With a limited number of donor hearts available, LVADs were presented as the solution for end-stage HF patients who became refractory to medical management. Investigator initiated, REMATCH was conducted at 20 experienced cardiac transplantation centers under a cooperative agreement among Columbia University, the National Institutes of Health, and the manufacturer of the device studied, the HeartMate vented electric VAD (1). It focused on LVADs for DT for end-stage HF patients ineligible for cardiac transplant, either because of age or other complicating factors. The FDA provided expedited review of the premarket approval application based on the landmark REMATCH trial results, and approval was obtained in November 2002 (Figure 1). Blue Cross Blue Shield had conducted its own Technology Evaluation Center (TEC) report on DT and concluded the therapy met its criteria for coverage in October 2002.

 

Figure 1. Timeline for VAD Destination Therapy Coverage by Payers

After REMATCH was published in 2001, discussion began with the Centers for Medicare and Medicaid Services (CMS) about coverage for destination therapy (DT) utilizing a ventricular assist device (VAD). After the technology was positively evaluated by Blue Cross Blue Shield (BCBS) and received approval from the U.S. Food and Drug Administration (FDA), a national coverage decision by CMS occurred within a year. MCAC = Medicare Coverage Advisory Committee; TEC = Technology Evaluation Center.

It was in 2001, however, when the REMATCH trial outcomes had been published in the New England Journal of Medicine (1), the HeartMate’s manufacturer began working with CMS to open a dialogue around the potential for a national coverage decision for DT prior to FDA approval, with a formal request to CMS made in May 2002. Numerous meetings with CMS officials were held to establish the clinically appropriate conditions for coverage, with the support of multiple clinical societies.  The technology received Medicare Coverage Advisory Committee endorsement in March 2003. This led to Medicare’s decision in October 2003 to establish coverage nationally for a narrow population (limiting the definition of DT to the clinical parameters of the REMATCH trial), and Medicare required certification of centers who would perform these surgeries (2).

The body of clinical evidence and how it is utilized significantly affects CMS coverage determinations. The hierarchy of evidence favors randomized, double-blind, placebo-controlled studies published in a peer-reviewed (preferably United States) journal.  Less attractive are open-label trials, where both patients and physicians know the patient is receiving the product, or retrospective studies, where charts are reviewed sometime after a group of patients have received a treatment. Case studies are rarely considered in a technology coverage decision.

In the case of the LVAD, the strength of the published controlled studies supported both CMS and private payers’ DT NCDs.

Pathways to U.S. Payment

After obtaining national coverage for DT in 2003, it was important to establish adequate payment to cover the cost of the therapy. Historically, hospitals were only reimbursed approximately $35,000 per LVAD implant since there was no specific DRG used to identify the procedure. Figure 2 shows the DRG assignment and base rate payment over time.

 

Figure 2.  History of VAD DRGs

As the diagnosis-related groups (DRG) for ventricular assist device system for destination therapy (DT) and transplant support changed over time, so too did the coverage rates from the Centers for Medicare and Medicaid Services (CMS). MS = Medicare severity.

During the coverage process in 2002, 2 years of claims data for DT cases reflecting the need to increase the inpatient hospital DRG payment were provided to CMS. Implanting hospital champions with clinical and cost data worked tirelessly to educate CMS about the clinical benefits and value of the technology. As a result, payment for DT hospitalizations increased over time, culminating in assignment of LVAD procedures to the heart transplant DRG in October 2003.  The current average hospital Medicare payment for centers providing DT or long-term support is $198,000 (3).

Global Reimbursement

Outside of the United States, some countries’ Ministries of Health utilize HTAs to determine whether patients should have access to a therapy or product. The HTA not only includes a review of published clinical trials and reported clinical benefit but also a cost-effectiveness analysis with consideration to health care budget impact.

At this point in time, there is no one global reimbursement pathway.  Each country establishes medical coverage, creates adequate codes, and assigns a payment group for devices. However, there are country clusters with similar mechanisms in establishing reimbursement. Table 2 attempts to unify a noncentralized approach by country reimbursement groupings. For products sold in the European Economic Area (EEA), the CE mark signifies productions have been assessed to meet safety, health, and environmental requirements. Compliance applies to relevant European Union legislation applicable to a product, regardless of the place of manufacture. There is a six-step process for manufacturers to affix a CE mark to the product and for active implantable medical devices, the involvement of a Notified Body is obligatory (4).

 

Table 2.  Device Reimbursement by Payment Clusters and Patient Access Requirements

CED = coverage with evidence development; CFDA = China Food and Drug Administration; FDA = U.S. Food and Drug Administration; HTA = health technology assessment; IDE = investigational device exemption; NCD = national coverage decision.

While clustering countries by type of payment is helpful to understand the approach needed, it is essential that reimbursement strategies be developed for each key country understanding that CE Mark approval or regulatory approval is the first step to establishing reimbursement. Countries such as France, Japan, and the Netherlands are requiring studies completed within their specific populations as well as a cost analysis reporting cost effectiveness for the technology in question.

With many global studies having been completed on the effectiveness of next-generation LVADs, most countries outside the United States cover for both bridge to transplantation and DT.  Table 3 provides a reimbursement overview by country capturing coverage and hospital payment status.

 

Table 3. MCS Global Reimbursement Overview

VI.6 Table 3

Coverage – percent of population with access: red <50%; yellow >70%; green = 100%.

BTD = bridge to destination; BTT = bridge to transplant; CHF = chronic heart failure; DRG = diagnosis-related group; DT = destination therapy; ECMO = extracorporeal membrane oxygenation; LVAD = left ventricular assist device; MS = Medicare severity.

Evolving Cost Effectiveness: How It Is Changing

Many cost-effectiveness studies have been performed with retrospective data on patients supported with the older generation of pulsatile LVADs. These studies conclude improvements in LVAD technology are necessary to improve cost effectiveness and continued reassessment is desired. Over the past decade, there has been a continual improvement in outcomes of durable continuous-flow LVAD support, with 6-month and 1-year survival rates for BTT near 94% and 85%, respectively. These improvements are largely due to better timing of LVAD implant, preoperative optimization, postoperative management protocols, and refinements in the technology. The overall rate of serious adverse events has steadily declined, which has a significant impact on hospitalization time, the number of interventions, and the total cost of care. With the use the current continuous-flow devices incremental cost effectiveness ratios have reduced nearing the $100,000 mark of willingness to pay per life-year saved (5).

During the MOMENTUM 3 trial, an economic analysis of the 2-year outcomes in the trial were reported demonstrating a reduction in rehospitalization-related hospital days and a significant cost saving with the centrifugal HeartMate 3 when compared with the axial HeartMate II continuous-flow LV assist system, again reflecting, that with next-generation LVADs, reducing adverse events moves the technology closer to being cost effective (6). Table 4 reports the reduction in cost by reducing adverse events and overall days in the hospital.

 

Table 4. Health Care Utilization for All Post-Index Hospitalizations

Costs Accumulated

Over Patient Time in Study

HeartMate 3

(n = 177)

HeartMate II

(n = 160)

Difference* p Value
Average cumulative cost per patient-year $37,685 ± 4,251 $76,599 ± 11,889 $38,913 <0.001
Average days in hospital per patient-year 17.1 ± 1.7 25.5 ± 2.7 8.3 0.003
Average hospitalizations per patient-year 12.1 ± 0.2 2.7 ± 0.2 0.6 0.015

Values indicate ± standard error.
*Difference shown is for HeartMate II – HeartMate 3.
†p values derived from bootstrap simulation (x10,000).

Lessons Learned

In the perfect coverage world, steps to reimbursement would seem simple.  Manufacturers would initiate randomized double-blinded, placebo-controlled studies, which report clear statistically significant long-term clinical benefits for patients in need and simultaneously demonstrate cost effectiveness to payers such as Medicare. An independent patient registry would follow, monitored by a medical group with no connections to the manufacturer, and ongoing peer-reviewed publications would be produced from this registry.

Many companies designing innovative devices are start-ups with one technology only. Financial resources are limited. Meetings with investors seem continuous; with reimbursement an important issue in their decision to provide funding.  The need to show some clinical effect early often outweighs the need to construct a comprehensive body of clinical evidence that will satisfy payers’ requirements. Despite early positive trials, payers have required additional studies specific to their population or simply additional studies so that substantial evidence can demonstrate the benefit of the device.  While such requirements may seem at best arbitrary, identifying the likely payers for a product, and even “partnering” with them in planning studies so that the results address their needs, can save years in the reimbursement process. These years translate to financial viability, which can mean life or death to a technology or a field. During product development, attention to clinical, regulatory, sales and marketing should be coupled with attention to reimbursement.  Almost immediately, a company must start defining who will primarily receive the product, and who then will be paying for it.  In addition, other issues such as who will actually “own” the product (the patient, the payer, the hospital, the physician) and who will do the procedure and in what setting, also become important in defining a reimbursement plan.

Ideally, reimbursement should be considered prior to the creation of the first clinical protocol.  Attention should be directed to what potential indications are being studied, and what impacts this product will have on the quality of life for particular populations, such as Medicare and private payer populations. Once this is determined, meetings with medical directors of the payers who will potentially reimburse for the technology are helpful, if not essential. In preparation for meeting with potential payers, the company should consider how resulting indications benefit this payer’s population.

A positive coverage decision also drives the need for a code to identify the product and or procedures. These codes help to establish payment by Medicare and private payers. However, confirmation of existing codes that can be used for the product can be determined prior to FDA approval. Companies must also anticipate costs to the health care system since it is increasingly necessary to prove a product is not only clinically beneficial, but also of value to the health care budget.

If studies do not support reimbursement, payers may be able to describe what additional information would be important for them to reevaluate the device. Can a small retrospective review study or specific outcome analyses affect their initial decision?  A nationally audited registry, maintained by an outside source, may be desirable to them.

In the case of the LVAD, CMS elected to provide “coverage with evidence development,” requiring centers that implant VADs to report specific data into a national registry.  With an initial grant of over $6,000,000 awarded by the National Heart Lung and Blood Institute, the Interagency Registry for Mechanical Assisted Circulatory Support, known as INTERMACS, was created to meet the reporting requirement established by CMS in collaboration with the FDA, CMS, and the clinical community; as of January 1, 2018, INTERMACS became part of the national database of the Society of Thoracic Surgeons (7). Another aspect of the coverage decision was the need for hospitals to become certified in DT by the Joint Commission (8). In 2018, 158 facilities were certified to provide DT therapy in the United States.

Summary

The primary components of a reimbursement strategy are: (1) obtaining medical coverage; (2) defining coding or listing; and (3) establishing adequate payment. Initial reimbursement efforts focus on coverage by determining where a technology fits, what indications are most appropriate or potentially limit coverage, and what coding modifications will be needed to obtain payment given the coverage anticipated. After reviewing the clinical study protocols with targeted payers, revisions are often needed to demonstrate the product is also reasonable and necessary under that payer’s criteria, as well as being cost effective. If these topics are addressed, a reimbursement strategy sensitive to the needs of the provider of care, the patient, the payer, and the company can be developed and successfully implemented.

References

  1. Rose EA, Gelijns AC, Moskowitz AJ, et al. Long-term use of a left ventricular assist device for end-stage heart failure. N Engl J Med. 2001;345:1435-43.
  2. Centers for Medicare and Medicaid Services. Decision Memo for Ventricular Assist Devices as Destination Therapy (CAG-00119R) (March 27, 2007). Available at https://www.cms.gov/medicare-coverage-database/details/nca-decision-memo.aspx?NCAId=187&ver=16&NcaName=Ventricular+Assist+Devices+as+Destination+Therapy+(1st+Recon)&bc=BEAAAAAAEAAA&&fromdb=true. Accessed April 29, 2019.
  3. Centers for Medicare and Medicaid Services. The FY 2010-17 Hospital Inpatient Prospective Payment System Final Rule. Available at https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/AcuteInpatientPPS/FY2017-IPPS-Final-Rule-Home-Page.html. Accessed April 29, 2019.
  4. European Commission. CE Marking – Manufacturers. Available at https://ec.europa.eu/growth/single-market/ce-marking/manufacturers_en. Accessed April 29, 2019.
  5. Rogers JG, Bostic RR, Tong KB, Adamson R, Russo M, Slaughter MS. Cost-effectiveness analysis of continuous-flow left ventricular assist devices as destination therapy. Circ Heart Fail. 2012;5:10-6.
  6. Mehra MR, Salerno C, Cleveland JC, et al. Economic Implications of the MOMENTUM 3 Long-term Outcome Study: A Randomized Controlled Trial of a Magnetically Levitated Cardiac Pump in Advanced Heart Failure. Circulation. 2018;138:1923-34.
  7. UAB School of Medicine. The STS Intermacs Database. Available at https://www.uab.edu/medicine/intermacs/. Accessed April 30, 2019.
  8. Joint Commission. Certification in Ventricular Assist Device. https://www.jointcommission.org/certification/ventricular_assist_device.aspx. Accessed April 30, 2019.
Hide picture