House Ways and Means Subcommittee on Health Hearing

Good afternoon and thank you, Chairman Tiberi, Ranking Member Levin, and Members of the Subcommittee. I am Mark Fendrick, Professor of Internal Medicine and Health Management & Policy at the University of Michigan. I am addressing you today, not as a representative of the University, but as a practicing primary care physician, a medical educator, and a public health professional. I have devoted nearly three decades to studying the United States health care delivery system, and I founded the University's Center for Value-Based Insurance Design [www.vbidcenter.org] in 2005 to develop, implement and evaluate innovative payment initiatives and health insurance designs intended to improve quality of care, enhance the patient experience, and ensure efficient expenditure of health care dollars.

Mr. Chairman, I applaud you for holding this hearing on "Promoting Integrated and Coordinated Care for Medicare Beneficiaries." The provision of patient-centered, high quality health care for our most vulnerable Americans and the containment of health care cost growth are among the most pressing issues for our national well-being and economic security. I strongly concur with your statement that Medicare expenditures should not only serve the best interests of current Medicare members, but must also serve the best interests of American taxpayers and future beneficiaries.

With 18.5M enrollees in 2017 and growing, Medicare Advantage (MA) is at the forefront of developing innovative programs - some of which will be addressed today - to prevent, detect, and treat vulnerable seniors and people living with disabilities, especially those with complex chronic conditions. I will focus my testimony on the importance of providing MA plans increased flexibility to use value-based insurance design (V-BID) principles to create a benefit package that encourages MA members to become smarter health care consumers. V-BID plans work synergistically with the other integrated and coordinated care models discussed today.

There is strong bipartisan agreement that the U.S. spends far more per capita on health care than any other country, yet lags behind other nations that spend substantially less on key health quality and population health measures. Since there is already enough money in the system, patient-centered outcomes can be improved if we reallocate our health care dollars to clinical services for which there is clear evidence for improving health. I believe that the primary goal of the Medicare program is to improve the health of its members, not to save money. Thus, the focus of our discussions should change from how much we spend to how well we spend our limited health care dollars.

FROM A VOLUME-DRIVEN TO VALUE-BASED SYSTEM

Moving from a volume-driven to value-based delivery system requires a change in both how we pay for care (supply-side initiatives) and how we engage consumers to seek care (demand-side initiatives). Other testimonies today and at earlier Subcommittee hearings have focused on the critical importance of reforming care delivery and payment policies. These are important and worthy conversations. Prior to this hearing, little attention has been directed to how we can alter beneficiary behavior to bring about a more effective and efficient Medicare program. Today, I propose that the goals of better health and cost containment are more likely to be achieved if MA plans were provided the flexibility to implement benefit designs that promote personal responsibility and improve member decision-making. I commend the Subcommittee for exploring this matter.

Role of Medicare Beneficiary Cost-sharing

Chairman Tiberi, in the announcement for this hearing, you called for a review of programs designed to deliver integrated and coordinated care for our most vulnerable seniors and people living with disabilities; the potential clinical and financial impacts of these programs are staggering. Of the 57 million people covered by Medicare in 2016; 36% report Functional Impairment (1+ ADL Limitations); 34% Cognitive/Mental Impairment; 30% 5+ Chronic Conditions; and 27% Fair/Poor Health. I have dedicated my career to ensure that at-risk Medicare beneficiaries get the care they need - at a price they can afford - in a fiscally responsible way.

Over the past few decades, public and private payers - including Medicare - have implemented multiple managerial tools to constrain health care cost growth with varying levels of success. The most common approach to impact consumer behavior is cost shifting: requiring beneficiaries to pay more in the form of increased premiums and increased cost-sharing for clinician visits, diagnostic tests, and prescription drugs. I can tell you with great confidence that the typical Medicare beneficiary does not worry about the total amount that the U.S. spends on health care, but they do care deeply about what it costs them. In 2016, more than 25% of Medicare beneficiaries spent 20% or more of their income on out-of-pocket (OOP) health care costs.

A significantly growing share of out-of-pocket spending is devoted to high cost medications, many of which have profound positive impact on beneficiary health. Most Medicare beneficiaries taking a specialty drug will spend more than $2,000 over the course of one year. Out-of-pocket costs for common, life-changing treatments for rheumatoid arthritis, Hepatitis C, and multiple myeloma frequently surpass $4,500, $6,500, and $11,500 respectively. To meet the growing burden, charitable foundations collectively provide Medicare members hundreds of millions of dollars each year. As health care costs escalate, most suggest that member OOP will continue to grow.

Dangers of a Blunt Approach to Beneficiary Cost-sharing - the Importance of "Clinical Nuance"

With some notable exceptions, MA plans implement cost-sharing in a 'one-size-fits-all' way, in that beneficiaries are charged the same amount for every doctor visit, diagnostic test, and prescription drug [within a specified formulary tier]. As Medicare beneficiaries are required to pay more to visit their clinicians and fill their prescriptions, a growing body of evidence demonstrates that increases in patient cost-sharing lead to decreases in the use of both non-essential and essential care across the entire continuum of clinical care. A systematic review of the published literature revealed that the rise in cost-sharing for Medicare beneficiaries resulted in lower adherence with recommended preventive screenings and prescription drugs to manage common chronic conditions, as well as reduced outpatient visits, leading to a rise in hospitalizations. Cost-related non-adherence (CRN) was shown to negatively impact the most vulnerable patient populations, especially those with lower socioeconomic status and multiple chronic conditions.

A noteworthy example is a New England Journal of Medicine study that examined the effects of increases in copayments for doctor visits in Medicare Advantage plans [Trivedi A. N Engl J Med. 2010;362(4):320-8]. As expected, individuals who were charged more to see their physician went less often; however, these patients were hospitalized more frequently, and their total medical costs increased. While this blunt approach may reduce expenditures in the short-term, higher rates of noncompliance may lead to inferior health outcomes and higher overall costs in certain clinical circumstances. This seemingly counterintuitive effect simply demonstrates that the age-old aphorism "penny wise and pound foolish" applies to health care. The lack of robust consumer incentives to improve their own health, coupled with illness burden, intense medication needs, and high out-of-pocket costs, often lead to undesired clinical and financial outcomes.

Since the decreased use of essential clinical services leads to reductions in quality, suboptimal patient-centered outcomes, and - in certain instances -increases in aggregate health care spending, solutions to this growing problem are urgently needed. To efficiently reallocate medical spending and optimize population health, the basic tenets of clinical nuance must be considered. These tenets recognize that: 1) medical services differ in the benefit provided; and 2) the clinical benefit derived from a specific service depends on the patient using it, as well as when, where, and by whom the service is provided.

Does it make sense to you, Mr. Chairman, that my Medicare patients pay the same copayment to see a cardiologist after a heart attack, as they do to see a dermatologist for mild acne? Or that their copayment is the same for a drug that could save their life from cancer, diabetes, or heart disease, as it is for toenail fungus treatment? On the generic drug tier available to most Americans, there are drugs so valuable that I have often reached into my own pocket to help patients fill these prescriptions; while for the same price, there are also drugs of such dubious safety and efficacy, I honestly would not give them to my dog. In the specialty drug tier, Medicare patients pay the same co-insurance for a 'precision' drug targeted to a specific genetic marker that cures cancer 90% of the time, as they do for a conventional therapy that rarely cures a single case.

Our current 'one- size- fits- all' system lacks clinical nuance, and frankly, to me, makes no sense. MA beneficiaries use too little high-value care and too much low-value care. We need benefit designs and other programs that support consumers in obtaining evidence-based services such as diabetic retinal exams and life-saving drugs through lower cost-sharing (when clinically indicated) and discourage individuals through higher cost-sharing from using dangerous or low-value services such as those identified by professional medical societies in the Choosing Wisely initiative. By incorporating greater clinical nuance into benefit design, payers, purchasers, beneficiaries and taxpayers can attain more health for every dollar spent.

VALUE-BASED INSURANCE DESIGN [V-BID]

Over the past two decades, public and private payers have implemented clinically nuanced plans, referred to as Value-Based Insurance Design, or V-BID. The basic V-BID premise calls for reducing financial barriers to evidence-based services and high-performing providers and imposing disincentives to discourage use of low-value care. A V-BID approach to benefit design recognizes that different health services have different levels of value. It's common sense -when barriers to high-value treatments are reduced and access to low-value treatments is discouraged, these plans result in better health at any level of care expenditure.

Let me be clear, Mr. Chairman, I am not asserting that V-BID is a panacea to the challenges facing MA plans. But, if we are serious about "bending the health care cost curve" and improving health outcomes, we must change the incentives for consumers, as well as those for providers. Cost containment through blunt changes to Medicare benefit design must not produce avoidable reductions in quality of care.

Your Subcommittee is examining many of the bright spots in Medicare

Advantage aimed to better integrate and coordinate care. If these initiatives provide incentives to clinicians to recommend the right care, it is of equal importance that incentives for the patients are aligned with these goals as well. As a physician practicing in an alternative payment model, it is incomprehensible to realize that my patients' coverage often does not offer easy access for those exact services for which I am benchmarked. Does it make sense that I am offered a financial bonus to get my patients' diabetes under control when the benefit design makes it prohibitively expensive to fill their insulin prescription or provide the copayment for their eye examination?

I'm pleased to tell you that the intuitiveness of clinically nuanced design is driving momentum at a rapid pace, and we are truly at a "tipping point" in its adoption. Hundreds of public purchasers, private self-insured employers, non-profits, and insurance plans have designed and tested value-based programs. Just a few examples include the State Employee Plans in Oregon, Connecticut, and Kentucky, each of which provide incentives for individuals with chronic diseases to seek the right care, at the right time, from the right provider. In January 2018, the TRICARE program will launch a V-BID demonstration to improve health outcomes and enhance the experience of care for U.S. Armed Forces military personnel, military retirees, and their dependents.

INFUSING 'CLINICAL NUANCE' INTO MEDICARE ADVANTAGE

In theory, Medicare Advantage can implement innovative programs designed to improve value by applying techniques successfully implemented in the commercial health insurance market. In reality, the tools available to Medicare Advantage are limited, and include network formation, performance bonuses, and utilization management programs. The use of these blunt instruments often does not align economic incentives with clinical value, thereby hindering a plan's ability to design benefits to promote quality and efficiency. This lack of flexibility is problematic, in that it fails to recognize the well-accepted notion that health care services differ in the clinical benefit achieved. Moreover, it does not align with the exciting advances in personalized or 'precision' medicine that are tailored to specific clinical characteristics. Additional flexibility in benefit design would allow Medicare Advantage plans to achieve greater efficiency and encourage personal responsibility among members.

There are two major restrictions within the Medicare Advantage program that prevent clinical nuance and the promotion of high-value services and providers: (1) a lack of flexibility to steer patients to high-value providers, and (2) a rigid, outdated benefit design. The standards for provider networks and non-discriminatory benefit designs were established in an effort to protect consumers from unfavorable practices such as predatory risk steering. While some of these provisions successfully improve consumer protection, they also severely limit innovation within the Medicare Advantage program and perpetuate a 'one-size-fits-all' approach to care delivery. Since these consumer protection standards prevent seniors from receiving the highest possible clinical benefits of care, they may be construed as undermining their original intent.

I. Flexibility in Imposing Differential Cost-Sharing for Use of Different Providers or Settings

Since the value of a clinical service may depend on the specific provider or the site of care delivery, Medicare Advantage plans should have the flexibility to vary cost-sharing for a particular outpatient service in accordance with who provides the service and /or where the service is delivered. The Commonwealth Fund Commission on a High Performance Health System estimated that $189 billion in savings would accrue to Medicare over 10 years if we were to "develop a value-based design that encourages beneficiaries to obtain care from high-performing care systems." This flexibility is increasingly feasible, as quality metrics and risk-adjustment tools become better able to identify high-performing health care providers and/or care settings that consistently deliver superior quality. For example, a Medicare Advantage plan might wish to impose a $50 copayment for an out-of-network office visit, a $25 copayment for an in-network office visit, and a $0 copayment for an in-network office visit that takes place at a recognized patient-centered medical home (PCMH), that has demonstrated better performance on key quality measures. Existing rules prohibit this level of variance in beneficiary cost-sharing, as Medicare Advantage plans are allowed to create a provider network, but are limited in how they vary copays within that network. Strict standardization in the cost-sharing structures within a network severely hinders the ability of Medicare Advantage plans to promote high quality care and take steps to reduce waste and inefficiency.

The provider network requirements also create challenges for care coordination among providers. The inability to use incentives to encourage beneficiaries to access care across a specified provider group hinders the ability for practitioners to track progress, encourage proper follow-up, and prevent the need for costly services due to lack of medical adherence. This is particularly important as we seek a return from a multi-billion dollar investment in health information technology. While the long-term intent of electronic medical records is to seamlessly share data across all providers, currently the most common use is among providers in a designated group.

Improving provider choice is an essential tool that will allow plans to incorporate clinical nuance, enhance consumer engagement, and drive higher quality of care in Medicare Advantage products. Network adequacy standards must allow issuers to create multi-tier cost-sharing structures by encouraging and requiring different tiers of co-pays for services and providers that have proven high- and low-value outcomes. Many stakeholders recognize the merits of permitting plans greater flexibility to incentivize beneficiaries to select high performing providers; the Medicare Payment Advisory Committee submitted these recommendations in several recent Reports to Congress.

II. Flexibility in Imposing Differential Cost-Sharing for Use of Different Services

To date, most clinically nuanced designs have focused on lowering patient out-of-pocket costs for high-value services. These are the services I beg my patients to do - for which there is no question of their clinical value - such as immunizations, preventive screenings, and critical medications and treatments for individuals with chronic diseases such as asthma, diabetes and mental illness (e.g. as recommended by National Committee for Quality Assurance, National Quality Forum, professional society guidelines). Despite unequivocal evidence of clinical benefit, there is substantial underutilization of these high-value services in the MA program across the spectrum of care. Multiple peer-reviewed studies show that when patient barriers are reduced, compliance goes up, and, depending on the intervention or service, total costs go down.

Yet, from the payer's perspective, the cost of incentive-only based V-BID programs depends on whether the added spending on high-value services is offset by a decrease in adverse events, such as hospitalizations and visits to the emergency department. While these high-value services are cost-effective and improve quality, many are not cost saving - particularly in the short term. However, research suggests that non-medical economic effects - such as impact on caregiver burden - can substantially impact the financial results of V-BID programs.

While significant cost-savings are unlikely with incentive-only programs in the short term, a V-BID program that combines reductions in cost-sharing for high-value services and increases in cost-sharing for low-value services can both improve quality and achieve net cost savings. Removing harmful/unnecessary care from the system is essential to reducing costs, while creating an opportunity to improve quality and patient safety. Evidence suggests significant opportunities exist to save money without sacrificing high-quality care. Though less common, some V-BID programs are designed to discourage use of low-value services and poorly performing providers.

Low-value services result in either harm or no net benefit, such as services labeled with a D rating by the U.S. Preventive Services Task Force. Many services that are identified as high quality in certain clinical scenarios are considered low-value when used in other patient populations, clinical diagnoses, or delivery settings. For example, cardiac catheterization, imaging for back pain, and colonoscopy can each be classified as a high- or low-value service depending on the clinical characteristics of the person, when in the course of the disease it is provided, and the where it is delivered.

Fortunately, there is a growing movement to both identify and discourage the use of low-value services. The ABIM Foundation, in association with Consumers Union, has launched Choosing Wisely, an initiative where medical specialty societies identify commonly used tests or procedures whose necessity should be questioned and discussed. Thus far, more than 40 clinical specialty societies have identified at least five low-value services within their respective fields. Immediate and substantial cost savings are achievable through the reduction of low-value care. Thus, programs that include both carrots and sticks may be particularly desirable in the setting of budget shortfalls.

III. Flexibility in Imposing Differential Cost-Sharing for Certain Services For Specific Enrollees

Since a critical aspect of clinical nuance is that the value of a medical service depends on the person receiving it, we recommend that Medicare Advantage plans be granted the flexibility to impose differential cost-sharing for specific groups of enrollees. The flexibility to target enrollee cost-sharing based on clinical information (e.g., diagnosis, clinical risk factors, etc.) is a crucial element to the safe and efficient allocation of Medicare Advantage expenditures. Under such a scenario, a plan may choose to exempt certain enrollees from cost-sharing for a specific service on the basis of a specific clinical indicator, while imposing cost-sharing on other enrollees for which the same service is not clinically indicated. Under such a clinically nuanced approach, plans can recognize that many outpatient services are of particularly high-value for beneficiaries with conditions such as diabetes, hypertension, asthma, and mental illness, while of low-value to others. For example, annual retinal eye examinations are recommended in evidence-based guidelines for enrollees with diabetes, but not recommended for those without the diagnosis. Without easy access to high-value secondary preventive services, previously diagnosed individuals may be at greater risk for poor health outcomes and avoidable, expensive, acute-care utilizations. Conversely, keeping cost-sharing low for these services for all enrollees, regardless of clinical indicators, can result in overuse or misuse of services leading to wasteful spending and potential for harm.

Currently, Medicare Advantage plans - with the exception of those participating in the CMS MA V-BID model test (discussed in detail below) - are constrained by non-discrimination rules that prohibit plans from tailoring benefits to particular subgroups of patients, for which a given service may be of particularly high-value. If MA plans were to encourage the use of a certain service by lowering copays, they must lower copays for everyone in the plan, even though clinical appropriateness may vary. In order to allow plans to incorporate the principles of clinical nuance in their MA products, the standards regarding targeting intervention by clinical circumstance should be updated.

Although the 'one-size-fits-all' approach to Medicare copayments dates back to its inception in the 1960s, support for the incorporation of V-BID principles into Medicare Advantage (MA) plans has garnered longstanding multi-stakeholder and bipartisan political support. In 2009, Senators Hutchison and Stabenow introduced a bipartisan bill, S.1040: Seniors' Medication Copayment Reduction Act of 2009, to allow a demonstration of V-BID in the Medicare Advantage program. The Seniors' Medication Copayment Reduction Act (2009, S. 1040), the Better Care, Lower Cost Act of 2014 (S. 1932), and The Strengthening Medicare Advantage through Innovation and Transparency for Seniors Act of 2015 (H.R. 2570) all proposed incorporating V-BID principles into MA.

To assess the fiscal impact of the first year of MA V-BID programs, an actuarial analysis from the patient, plan, and societal perspectives was undertaken for diabetes mellitus (DM), chronic obstructive pulmonary disease (COPD), and congestive heart failure (CHF). After the first year, V-BID programs reduced consumer out-of-pocket costs in all three conditions. Plan costs increased slightly for DM and COPD, and the plan realized cost savings for CHF. From the societal perspective, the DM program was close to cost neutral; net societal savings resulted in the COPD and CHF programs.

CMS MEDICARE ADVANTAGE V-BID MODEL TEST

In the fall of 2015, the Centers for Medicare and Medicaid Services (CMS) announced the Medicare Advantage V-BID model test to assess the utility of structuring consumer cost-sharing and health plan elements to encourage the use of high-value clinical services and providers. MA plans in in Arizona, Indiana, Iowa, Massachusetts, Oregon, Pennsylvania, and Tennessee were eligible to implement programs for seven CMS specified chronic conditions. Changes to benefit design made through this model may only reduce cost-sharing and/or offer additional services to targeted enrollees. Under no circumstances can targeted enrollees receive fewer benefits or have to pay higher cost-sharing than other enrollees as a result of the model. Four approaches to benefit design are permitted in the model:

1. Reduced Cost-Sharing for High-Value Services

Plans can choose to reduce or eliminate cost-sharing for items or services, including covered Part D drugs, that they have identified as high-value for a given target population. Participating plans have flexibility to choose which items or services are eligible for cost-sharing reductions; however, these services must be clearly identified and defined in advance, and cost-sharing reductions must be available to all enrollees within the target population. Examples of interventions within this category include eliminating co-pays for eye exams for members with diabetes and eliminating co-pays for angiotensin converting enzyme inhibitors for enrollees who have previously experienced an acute myocardial infarction.

2. Reduced Cost-Sharing for High-Value Providers

Plans can choose to reduce or eliminate cost-sharing when providers that the plan has identified as high-value treat targeted enrollees. Plans may identify high-value providers based on their quality and not solely based on cost, across all Medicare provider types, including physicians/practices, hospitals, skilled-nursing facilities, home health agencies, ambulatory surgical centers, etc. Examples of interventions within this category include reducing cost-sharing for members with diabetes who see a physician who has historically achieved strong results in controlling patients' HbA1c levels and eliminating cost-sharing for heart disease patients who elect to receive non-emergency surgeries at high-performing cardiac centers.

3. Reduced Cost-Sharing for Enrollees Participating in Disease Management or Related Programs

Participating plans can reduce cost-sharing for an item or service, including covered Part D drugs, for enrollees who choose to participate in a plan-sponsored disease management or similar program. This could include an enhanced disease management program, offered by the plan as a supplemental benefit, or it could refer to specific activities that are offered or recommended as part of a plan's basic care coordination activities. Plans using this approach can condition enrollee eligibility for cost-sharing reductions on meeting certain participation milestones. For instance, a plan may require that enrollees meet with a case manager at regular intervals in order to qualify. However, plans cannot make cost-sharing reductions conditional on achieving any specific clinical goals (e.g., a plan cannot set cost-sharing reductions on enrollees achieving certain thresholds in HbA1c levels). Examples of interventions within this category include elimination of primary care co-pays for diabetes patients who meet regularly with a case manager and reduction of drug co-pays for patients with heart disease who regularly monitor and report their blood pressure.

4. Coverage of Additional Supplemental Benefits

Under this approach, participating plans can make coverage for specific supplemental benefits available only to targeted populations. Such benefits may include any service currently permitted under existing Medicare Advantage rules for supplemental benefits.

Nine MA plans started the model test in January 2017. Aetna's "Healthy Heart Partnership," Geisinger's "COPD Support" and UPMC's "Spark Your Health" are excellent examples of how enhanced benefits for members with a complex chronic condition can be coupled with care management programs to better engage patients and improve clinical outcomes. Responding to interest from MA plans in states not included in the demonstration, CMS announced that the model will expand to 10 (from 7) states and add two clinical conditions for 2018.

BIPARTISAN SUPPORT TO EXPAND MA V-BID MODEL TO ALL 50 STATES

Due to V-BID's success in the public and private sector, the TRICARE V-BID pilot, and early enthusiasm for the MA V-BID demonstration, the U.S. Senate Finance Committee introduced S.870: Creating High-Quality Results and Outcomes Necessary to Improve Chronic Care Act (CHRONIC) of 2017, a bipartisan bill that specifically calls for the expansion of the V-BID MA demonstration to all 50 states. Recently, Representative Diane Black (R-TN), along with co-sponsors Earl Blumenauer (D-OR), Cathy McMorris Rodgers (R-WA), and Debbie Dingell (D-MI), introduced the V-BID for Better Care Act of 2017 (H.R. 1995), which seeks to provide national testing of the Medicare Advantage V-BID Model. The national implementation of clinically nuanced benefit designs presents an enormous opportunity for the Medicare Advantage program.

Although there is urgency to bend the health care cost curve, cost containment efforts should not produce avoidable reductions in quality of care, particularly for the most vulnerable among us. It is my hope that as your Subcommittee considers changes to the Medicare Advantage program, you will take the important step of providing MA plans in all 50 states the flexibility to set cost-sharing levels based on whether an intervention is high-value or low-value. Encouraging the use of high-value services and providers, and discouraging those with low value, will decrease cost-related non-adherence, reduce health care disparities, and improve the efficiency of health care spending without compromising quality. This approach - working in concert with other exciting integrated care models discussed today - would result in a healthier population, and contain the growth of Medicare expenditures, thus serving the best interests of American taxpayers and future beneficiaries.

Thank you.

Read this original document at: https://waysandmeans.house.gov/wp-content/uploads/2017/06/2017.06.07.-HL-Testimony-Fendrick.pdf