March 14th, 2014
Current Medicare reform policy proposals mainly focus on lowering annual cost or cost increase per capita, but they fail to recognize Medicare as a lifetime plan that covers each beneficiary from age 65 to death. I propose a Lifetime Value-Based Payment Plan (LVBPP) for Medicare reform. LVBPP aims to achieve efficient use of the government contribution to Medicare for each beneficiary from age 65 to death and features shared responsibility among beneficiaries, providers, and federal government.
LVBPP includes six major components to create incentives for chronic disease prevention and efficient use of medical care resources by promoting market-based competition on quality of care and innovations in medical technology and care delivery models. Preliminary results indicate that LVBPP could lead to better health in terms of longer longevity and lower disability rate, save up to $70 billion over 10 years, and save up to $164 billion for the federal government over the lifetime of the cohort of upcoming beneficiaries age 55 to 59, as of the 2010 census. (The bases for these savings estimates, as well as suggested values for the expenditure thresholds and copayment rates involved in LVBPP, are provided in the Simulation Appendix below.)
The challenge. There is wide consensus that chronic disease is the leading cause of mortality and rapidly increasing health care costs in the US. Lifestyle choices have been found to be a major factor behind the increasing prevalence of chronic disease. It is estimated that 60 percent of deaths and 70 percent of health care spending in the US are related to lifestyle choices. However, despite volumes of science-based clinical trial results demonstrating positive effects of behavioral change on patients’ long-term well-being, and continuous public media campaigns promoting lifestyle change, there is no sign of reduction in the prevalence rate of chronic disease in the US population.
On the contrary, chronic disease prevalence continues to rise. For example, the United Health Group projects that the prevalence rate of diabetes and pre-diabetes among US adults will increase from 37 percent in 2007 to 52 percent in 2020.
We must confront the challenge of the rising epidemic of chronic disease or face a massive economic loss as related health care costs, and in particular publicly financed Medicare costs, continue to rise. For example, research by economists Kevin Murphy and Robert Topel from University of Chicago showed that, despite multi-trillion dollar welfare gains in extended longevity and better quality of life since the 1970s due to the development and diffusion of medical technology, there is substantial societal economic loss among older adults (especially females age 65 or higher) when factoring in health care expenditures for chronic disease.
Economists suggest two reasons rooted in the current health care payment system for disappointing results in the control of chronic disease and health care costs. First, the price distortion created by annuity, third-party payer, and Medicare fee-for-service (FFS) mechanisms discounts the values individuals place on their own longevity and quality of life. The current health care financing system does not provide enough economic incentives for working and elderly adults to pay attention to the big picture of longevity and quality of life and to make wise lifestyle choices that maximize their long-term well-being. Today, working and elderly adults pay only a small fraction of the cost of procedures required to treat the complications of their chronic diseases out of pocket. From the beneficiary’s perspective, the opportunity cost of current lifestyle change is drastically higher than the out-of-pocket cost of current and/or future medical care.
Second, private health insurance plans and the government are not incentivized to reimburse and promote prevention or chronic disease management interventions. Because the financial benefit of prevention is often realized in the long-term, there is no incentive for private insurance plans covering working adults to pay for prevention; any future benefit generated from prevention may benefit other private insurers if enrollees switch plans to seek more generous coverage or treatment. More importantly, because Medicare covers each beneficiary from age 65 until death, whether increased longevity from chronic disease prevention will lead to higher or lower lifetime Medicare expenditures is still subject to debate. Providers are financially motivated by the FFS system, which rewards high volumes of expensive procedures; when beneficiaries live longer, more procedures can be carried out on them over their lifetime.
I would argue that the focal points of policy debate on Medicare reform, therefore, should be two issues:
- patient cost-sharing mechanisms that create shared responsibility in the adult population and confront them with the marginal cost of their lifestyle choices, creating incentives for behavioral change, and
- health care reimbursement mechanisms that create incentives for providers to proactively offer appropriate preventive care and chronic disease management, but without limiting access to the most advanced life-saving medical technology and the most skillful medical care providers.
LVBPP is fundamentally different from current Medicare reform proposals or demonstrations, which all focus on controlling annual Medicare costs or annual cost increase per capita. Such proposals, represented by Accountable Care Organizations (ACO) etc., mainly focus on regulating physicians’ practices by controlling price, procedure choice, and volume of medical serves. Such approaches leave the patients out of the decision making picture, and therefore often create trade-offs between cost control and reduced access, lower quality, and jeopardizing patient safety.
In contrast, LVBPP focuses on Medicare’s unique feature of covering each individual from age 65 to the end of life to promote the highest value for government contributions to seniors’ health care.
LVBPP includes six key elements as introduced below.
I. Free choice between traditional government “defined benefit” plan and private insurance carriers.
Beneficiaries will be allowed to choose to stay with the traditional “defined benefit” government plan or join a private carrier for their Medicare services at any time over their lifespan from age 65 to death.
- If the beneficiary chooses to stay within the traditional Medicare FFS program, Medicare will continue to reimburse health care providers under the fee-for-service (FFS) system for Part A, Part B, and the “defined contribution” plans of Part D unless new reimbursement regulations are passed.
- If the beneficiary chooses to join a private carrier, Medicare will either pay the private carrier a negotiated annual premium that is similar to the current Medicare Advantage (MA) payment, or let the private carrier manage the patient under a procedure-based FFS payment method similar to the current Part D plan but without an annual expenditure cap. The scope of services covered by the private carriers will be regulated by the federal government, but with flexibility in a variety of “service package” combinations.
II. A lifetime expenditure threshold that triggers an additional copayment charge based on means testing.
Medicare will establish a lifetime expenditure threshold above which a higher copayment rate will be applied to additional FFS payments and/or additional contributions from Medicare to private carriers. Such a threshold will serve as a benchmark to promote efficient use of medical care and avoid waste or abuse of expensive services. The threshold, for example, could be defined at between one and one and a half (1.5) times the average lifetime payroll tax contribution to Medicare per capita (currently around $77,000).
The copayment rate should be means-tested based on income and assets at age 65. However, I suggest all seniors share at least a 5 percent copayment despite their income level to promote responsibility and positive behavior change among all individuals.
I expect that sicker, poorer and older patients are more likely to stay in, or switch back to, traditional FFS Medicare, leading to a cost transfer from the private to the public sector. Therefore, this threshold will be applied not only to beneficiaries who choose to stay in traditional Medicare, but also to those who choose to join private carriers. Specifically:
- For beneficiaries who choose to stay in the FFS Medicare program, actual Medicare expenditures will be summed up to the lifetime expenditure threshold calculation. When the lifetime Medicare expenditures for a beneficiary exceed the threshold, a higher copayment rate of an additional 20 percent or higher on average will be charged for each procedure. Medicare will reduce its payment by an equivalent amount.
- For beneficiaries who choose to join private insurance carriers, Medicare contributions to private carriers will be summed up and compared to the lifetime expenditure threshold calculation. When the lifetime Medicare contribution exceeds the threshold, a higher copayment rate of an additional 20 percent or more will be charged for the premium contributions or specific medical procedure reimbursement, and Medicare will reduce its contribution by an equivalent amount.
III. A Health Promotion Reward to encourage behavioral change and competition on preventive care.
To accompany the lifetime expenditure threshold that triggers increased copayment to control costs, I propose a Health Promotion Reward mechanism to promote lifestyle changes and encourage market competition to provide higher quality preventive care. Specifically:
- Beneficiaries who stay in the traditional Medicare FFS program and engage in preventive measures as advised by medical experts and meet age-adjusted wellness goals to reduce chronic disease risks will be rewarded financially by Medicare. The reward rate, for example, could be set at 20 percent or higher of the average per-member-per-year (PMPY) Medicare expenditures.
- For beneficiaries who choose private carriers, if the private carrier can provide better quality of care that leads to the individual beneficiaries exceeding their wellness goals provided by Medicare FFS, the beneficiaries will share a higher rate of reward than the Medicare beneficiaries who choose FFS.
- The reward will be based on both the participation and the clinical outcomes of preventive services, with at least 50 percent based on participation. The participation reward should include elements such as an annual physical, routine diabetes and hypertension management, and fall prevention education. The outcome reward should include the risk of cardiovascular diseases adjusted by age, prevention resources at the community level, and disability. Considering that sicker beneficiaries will likely reach the lifetime expenditure threshold earlier, the reward rate could be set higher for sicker beneficiaries with special care needs (e.g., cancer or Alzheimer’s disease patients) based on their baseline health status — especially preexisting chronic diseases — at age 65 or when they start using Medicare benefits.
- The reward will be deemed personal credit in a variety of forms, e.g., to count against previously accumulated costs or to pay for future copayments. However, it will not be returned as cash to beneficiaries or providers, nor may it be collected by the beneficiary’s family if he or she dies before reaching the lifetime expenditure threshold.
IV. Increased reimbursement rate for preventive care and innovative chronic disease management models within the threshold.
- For the beneficiaries who choose to stay within the traditional Medicare FFS plan, I propose a higher reimbursement rate for primary, secondary, and tertiary prevention interventions, and innovative integrated care models for people with multiple chronic diseases. The payment rate should be negotiated between health care providers and the federal government and determined by the cost of labor, technology upgrades, and supporting services at the community level.
- For the beneficiaries who choose private carriers, the reimbursement rate for preventive and integrated care will be determined between private insurance companies and health care providers, and patients will have free choice of alternative plans based on competitive bidding.
V. Catastrophic coverage protection.
In cases with continual need for high-cost medical care over a long period of time, the additional copayment rate above the lifetime expenditure threshold could cause significant financial burden beyond the financial abilities of beneficiaries and their families. Such cases might include, for example, the oldest beneficiaries (age 85 or older), End Stage Renal Disease (ESRD) patients, cancer patients in need of experimental treatments, and those with genetic disorders. Therefore, I propose to set an upper cap on lifetime expenditures to provide catastrophic protection. Medicare will cover all costs for services without copayment when lifetime expenditures exceed the upper cap, including both those who chose the FFS plan or a private carrier.
VI. Flexibility in Medicare initiation age.
Because LVBPP is lifetime based, I suggest allowing seniors to initiate Medicare coverage at an age older than 65. If a relatively healthy senior retires later than age 65, and can stay covered by her employer-provided or privately purchased health insurance policy, she can postpone her Medicare benefit initiation age without any reduction in the lifetime benefit “package”. Furthermore, Medicare can reward the later retirees financially at 20 percent of the PMPY cost, to be added to the lifetime expenditure cap. Therefore, healthier and more productive seniors won’t lose benefits from their lifetime contributions to Medicare as a result of an eligibility age mandate, and late retirees won’t be incentivized to substitute Medicare for private insurance. Instead, LVBPP encourages late retirees to maintain their employer-based or other forms of private insurance.
In summary, LVBPP creates shared responsibility among beneficiaries, health care providers, and the federal government to promote more efficient use of health care resources. Unlike ACO and other annual proposals that place annual financial pressure on hospitals and physicians exclusively, LVBPP creates incentives for beneficiaries to pay attention to the big picture of lifetime health costs and quality of life, and to work together with payers and providers as active participants to maintain their health, rather than clients that passively consume health care services.
LVBPP offers free choice between government and private carriers to protect beneficiaries’ options for obtaining higher quality of care in the market. In addition, it gives physicians the flexibility to execute medically necessary procedures at any time without an annual limit. On the other hand, however, it keeps both beneficiaries and health care providers responsible and accountable for patients’ long-term well-being. The higher copayment charged beyond the lifetime expenditure threshold will work together with the Health Promotion Reward as a “stick and carrot” strategy to motivate chronic disease prevention and, more importantly, to reward health promotion and cost-saving behavior.
The catastrophic coverage protection, on the other hand, ensures that access to necessary services and technology with be protected for the most vulnerable and frail beneficiaries. Lastly, the flexibility of allowing Medicare initiation above age 65, along with the lifetime expenditure threshold design and Health Promotion Reward, creates incentives for later retirees to substitute other forms of private insurance for Medicare for longer, since the Medicare lifetime benefit package is still guaranteed in older age.
LVBPP has several advantages compared to numerous annually-focused Medicare financing reform models:
Fairness. One major policy proposal to reduce Medicare cost is to charge the wealthy elderly a higher copayment or premium for inpatient and/or outpatient services. However, without a known expectation of longevity, it is not financially effective to charge higher-income elderly a higher copayment immediately when they become eligible for Medicare. More importantly, to charge the wealthy elderly a higher copayment will not lead to meaningful behavioral change at the population level if the policy is purely designed for achieving economic equality.
Under LVBPP, to guarantee fairness, contributions up to the lifetime expenditure threshold create basic coverage for all seniors to ensure they do not pay more than others for the guaranteed benefit that they deserve because of their hard work and contributions to Medicare in their younger age.
Shared responsibility and more incentives for better health. The lifetime expenditure threshold that triggers a higher copayment, the means-tested copayment, and the Health Promotion Reward create shared responsibility among the government, providers, and beneficiaries to maintain beneficiaries’ health and modify their behavior.
Predictable out-of-pocket cost with lifetime catastrophic coverage. Many Medicare reform models with annual premiums or annual catastrophic coverage make patients face unpredictable out-of-pocket expenditures amidst decreasing wealth due to aging and a higher demand for health care in their older age. In contrast, LVBPP guarantees a predictable out-of-pocket maximum expenditure. Therefore, Baby Boomers and later generations could plan for medical expenditures during retirement earlier and more easily with saving or retirement plans.
More incentives for medical technology and care delivery innovation. LVBPP does not establish an annual contribution or a payment limit, so it avoids cuts in physician fees or limits in access to modern technology that would harm both patient safety and medical technology innovation. Because LVBPP doesn’t impose an annual limit on expenditures, patients can, with the help of predictable out-of-pocket expenditure and catastrophic coverage, seek the best care provided by the most skilled physicians, with cutting edge technology, when needed. Such a payment model will likely promote innovations in both medical technology and delivery models.
Multiple negotiation points for policy adjustment that will increase the chances for success in the legislative process and leave flexibility for modification in implementation.
The data used in the simulation in the technical appendix below is not the most recent and lacks Medicare Part D information. In addition, the model doesn’t include Home Health (HH) care costs. Therefore, the estimate of the total Medicare expenditure is on the lower end. I plan to expand the model with updated data through 2009 (CMS has a 4-year lag to release MCBS data), and modify the model to add Part D and HH care costs.
There is a lack of references about how people will react to such a payment model, as a similar model does not exist to date. However, based on the success of defined-contribution Medicare Part D, economic literature confirms the practicality of “consumer-directed health care.” Researchers have concluded that by confronting consumers with the full or partial marginal cost of the services they use and the marginal costs and benefits of the behavioral choices they make, it is possible to achieve efficient allocation of health care resources in a competitive market-oriented structure.
Besides the formidable effort needed to design a sophisticated lifetime Medicare financing model, the implementation of such a plan is a separate and possibly even more challenging task. One major issue is the asymmetric information that patients face when choosing between private carriers and the traditional “defined benefit” Medicare package, including their expectations of longevity, future health and health care needs, etc. Successful implementation of such a plan would have to be based on transparency and sincere collaboration between the federal government and local communities.
Although the market-oriented Medicare Part D plan was successful, it is an annually based plan to manage the use of prescription medications alone. To incorporate the dynamic demand for inpatient care, outpatient care, home health care, prescription medication, and end-of-life care into a market-oriented model with free annual enrollment choice and lifetime budget restrictions requires a tremendous amount of effort from the federal government, industry, and academia to find the most efficient solutions. The federal government plays and would continue to play a key role in protecting patients from making unwise choices by regulating service packages, market entry, and the reimbursement strategies of the private carriers offering the Medicare benefit.
Despite the pain of the annual negotiation, and repeated call from the physician groups for meaningful reforms to abandon the Sustainable Growth Rate (SGR) and establish a long-term physician payment mechanism, there is a gap in policy literature about the alternatives to SGR and how the federal government will regulate both physician and team-based clinical services under a “defined contribution” framework. Although I have confidence in the practicality of LVBPP as a financing framework, there is still a lot of work that needs to be done to reform SGR and basic physician and team-based care payment mechanisms.
Lastly, I am concerned that the current Congressional Budget Office (CBO) scoring system lacks the ability to measure the dynamics of market competition and behavioral response to policy changes. The current CBO scoring system has forced the “defined contribution” advocates to include annual Medicare budgets or spending caps to cope with this demand, which makes the budgeting task easier but could hurt the efficiency of the allocation of medical care resources. This approach should be updated to cope with the demands of a dynamic lifetime payment model for the highest quality evidence-based policy design.
Appendix: Simulating LVBPP
I conducted a preliminary policy simulation of LVBPP using nationally representative Medicare data. The methods and results are presented below. I used Medicare Current Beneficiary Survey (MCBS) data from 1997 to 2005 to estimate the lifetime expenditures for Part A (inpatient and SNF), Part B (outpatient), and outpatient prescription drugs under the current FFS payment system. MCBS is collected by the Centers for Medicare and Medicaid Services (CMS) as a nationally representative sample of current Medicare beneficiaries.
The simulation employed a Dynamic Aging Process (DAP) model to approximate the aging process from age 65 to death, including the development of chronic diseases, disability, and death, and the demand for medical and nursing care reimbursed by Medicare and Medicaid. Using the parameters obtained from the estimation, I simulated the baseline lifetime expenditures among current beneficiaries from age 65 to death, as well as the magnitudes of savings under a variety of scenarios assuming changes in reimbursement and behavior under LVBPP. For the technical details of the estimation and simulation strategy, please refer to previous publications (Yang and Hall 2008; Yang, Gilleskie and Norton 2009; Atherly and Yang 2011; Thorpe and Yang 2011).
Payment reform design outline:
Based on the six key elements as introduced above, outlined below are suggested specifications for the key payment caps, copayment rate, and Health Promotion Rewards rate. These values were used in the simulation. All values are in 2013 dollars.
- A lifetime expenditure threshold that triggers an additional copayment charge at $120,000. This is about 1.5 times the per capita lifetime contribution to Medicare (approximately $80,000), to guarantee the basic Medicare contribution in health care for seniors. Based on preliminary analysis among the current Medicare beneficiaries under the FFS system, the median lifetime expenditure is also approximately $120,000. Guaranteeing such coverage benefits both seniors and governmental fiscal solvency.
- An annual Health Promotion Reward at $2,000 (approximately 20 percent of the PMPY cost) for 15 years. Medicare will stop the reward after the beneficiaries reach age 80. The reward will be based on prevention and disease management of diabetes and cardiovascular disease, including heart diseases and stroke. The reward will be redeemed if patients do not show higher than age-adjusted risk for diabetes and cardiovascular disease as measured by the management of body weight, HbA1c level, cholesterol level, and blood pressure. The specific measures could be based on quality metrics from the Medicare Physician Group Practice (PGP) demonstration. Due to data limitations, and to simplify the simulation for preliminary results, I didn’t include the scenario of reward for participation as introduced above, but only focused on the outcome measures of CVD and diabetes for the Health Promotion Reward.
- 30 percent copayment rate above $120,000 without health promotion rewards, or above $120,000 plus the amount of health promotion rewards, with a maximum of $150,000 ($120,000 + $2,000*15). The lifetime additional copayment amount will be capped at $30,000 maximum, and at a minimum of $0. The copayment rate should be adjusted by means test, so that the poor elderly pay lower rates, but will require all beneficiaries to share at least 5 percent of the cost of the treatment reimbursed by Medicare.
- Catastrophic coverage above $220,000 in lifetime expenditures without copayment.
I first simulated a baseline scenario under the current FFS system. This was focused on three categories of services covered by Medicare under current law, including inpatient care (Part A), outpatient physician care (Part B), and outpatient prescription drugs. As my data represents the population between 1997 and 2005, when Medicare Part D was not yet enacted, I used the total outpatient prescription drugs expenditures as a proxy. I simulated the lifetime and 10-year total Medicare expenditures per capita, as well as average beneficiary longevity per capita.
On average, the baseline lifetime Medicare expenditures per capita were $147,316 (2013 dollar value), the 10-year Medicare expenditures per capita were $60,395, and the life expectancy per capita was about 17.9 years. This estimate of life expectancy is very close to the 17.7 years published by the National Center for Health Statistics (Centers for Disease Control and Prevention, 2011), which again demonstrates the validity and consistency of the simulation model.
In addition to the baseline scenario, I simulated five scenarios under the LVBPP design and compared the average lifetime expenditures and longevity per capita with the baseline scenario under the current FFS system. In addition, I calculated the population-level savings for a 5-year cohort of future Medicare beneficiaries who are currently age 55 to 59 based on the 2010 Census.
The five scenarios simulated were:
- Most cautious scenario without any behavioral change or improvement in health.
- Moderate behavioral change with 10 percent reduction in acute cardiovascular disease events (both new and recurring).
- Significant behavioral change with 20 percent reduction in acute cardiovascular disease events (both new and recurring) and 8 percent reduction in BMI among the overweight and obese at baseline.
- “Spillover” effect with a 5 percent drop in cardiovascular disease and diabetes prevalence when the seniors turn 65 and become eligible for Medicare.
- Most optimistic scenario with the combination of savings from scenario 3 and 4.
The results as depicted in Table 1 (below) indicate that LVBPP could lead to 4.8 percent or higher savings in 10 years, and 6.8 percent or higher savings over the lifetime per capita per cohort. The highest savings per cohort could reach 11.7 percent at $17,252 over the lifetime per capita, and 12.3 percent at $7,408 over 10 years per capita. Greater reductions in acute CVD events and baseline weight at age 65 are associated with increased longevity and lower cost to the federal government. At the population level, LVBPP could result in $70 billion in government savings over ten years, and $164 billion in savings over the lifetime of the cohort of future beneficiaries who are currently between age 55 and 59, based on 2010 census.
Table 1, Summary of Magnitude of Savings from LVBPP among Current 55-59 Cohort
These savings estimates are likely conservative for two reasons. First, they assume that the pattern of medical decision making in the year of death does not change in response to LVBPP’s financing changes. In other words, the estimates assume that patients will seek, and physicians will provide, equally intensive care under LVBPP as under the FFS payment system. In fact, however, patients or their families might be more likely to abandon or replace intensive care with hospice or other less expensive end-stage care due to the higher co-payment structure beyond the lifetime expenditure threshold.
Second, the estimates do not consider the possible savings from cutting waste. Because LVBPP holds both beneficiaries and their health care providers responsible for not only the beneficiaries’ health status, but also their financial well-being in the long run, it could motivate both parties to cut waste and redundant procedures.
However, the savings estimates for LVBPP also assume the 30 percent copayment amount is strictly implemented across the board without a means test, which could lead to over-estimated savings. Nevertheless, based our previous estimation, the mean of lifetime Medicare expenditures is about $150,000 (in 2013 dollars), and the median is around $120,000. Therefore, even under the current FFS system, if the lifetime cap is applied, about 50 percent of the beneficiaries won’t need to pay additional out-of-pocket costs.
On the contrary, LVBPP creates more incentives for the lower 50 percent to change behaviors and live longer and healthier lives. Because of the positive relationship between education, wealth, and longevity, I assume that the longer the life expectancy, the higher the beneficiaries’ ability to afford the copayment. Thus, the upper 50 percent of the beneficiaries are more likely to have the financial ability to afford the copayment if implemented at a $30,000 lifetime maximum.
Because the simulation is based on retrospective Medicare data under the current FFS system, patients still faced out-of-pocket payment under Part A, B, and D, and payment from Medigap plans. The copayment after the threshold of LVBPP should be considered additional out-of-pocket expenditure. However, with the LVBPP design as a general concept, my future work will provide more detailed discussion about the specific payment reforms under the LVBPP framework to reduce annual out-of-pocket payments and eliminate annual financial risk of out-of-pocket payments for the patients. Therefore, Medicare beneficiaries will face less financial uncertainty under LVBPP than the current FFS plan.
In summary, I believe that the estimate of the savings from the copayment is cautiously optimistic with a slight overestimation. Combining these four factors together, I expect that the overall savings estimate is still conservative, and that actual savings could be higher.Email This Post Print This Post
Don't miss the insightful policy recommendations and thought-provoking research findings published in Health Affairs.