Dr. Melinda Buntin, PhD, testified on June 27, 2018 before the Senate HELP Committee. Below is a copy of her prepared testimony for the record, with footnotes.
--
Dr. BUNTIN: I would like to thank Chairman Alexander and Ranking Member Murray for giving me the opportunity to speak today about how we can address the high costs of health care in the United States. My name is Melinda Buntin, and I am the Mike Curb Professor of Health Policy in the Department of Health Policy at the Vanderbilt University School of Medicine. This testimony is derived in part from recent academic work with colleagues at Vanderbilt and from earlier work done while I was at the Congressional Budget Office and RAND.
Problem Statement:
The amount that we spend on health care in the United States is high -- $3.3 trillion dollars per year. That works out to more than $10,000 for every man, woman, and child in the country. As a result, health care accounts for a large fraction of our total national output, or GDP. We currently devote 18 percent of our GDP to health care – almost one dollar out of
every five spent in our economy is spent on some form of health care. Many households devote an even greater share to health care. Consider, for example, the Milliman Medical Index, which captures the average costs of a typical employer-sponsored plan for a family of four. It was over $28,000 in 2017, which is roughly equivalent to the wages of two full-time workers at the federal minimum wage.1
This level of expenditure is a major reason that the wages of American workers have stagnated. It is also a reason why employers have been slow to hire full-time workers as we have grown out of the recession.2 This level of spending also puts a high burden on our working population to support benefits for older, disabled, and poor citizens who depend on Medicare
and Medicaid.
Yet despite these high spending levels, in this testimony I will argue that it is not overall dollar amounts, or that proportion of GDP per se, that is a problem. Instead, I will argue that it is per capita cost growth that is the most important factor to watch. Per capita growth gives the clearest indicator of the growing cost of care delivery and the changes in our health care system.3 Trends in per capita costs also underscore that policy choices we have made and can make in the future do drive changes in health care delivery. In fact, per capita cost growth in the Medicare program has been low over the past decade and examining those trends provides some concrete examples of how cost growth might be kept in check.
What drives what we spend on health care?
To think about how policy choices might affect the levels of and growth in health care spending, it is important to understand the components of health care costs and what drive them. These vary by payer, whether that payer is Medicare, Medicaid, or private insurance.
First, there is the number of people covered. Overall Medicare and Medicaid spending have grown rapidly over the past decade primarily because of growth in the number of people covered. Medicare has grown because of the aging of the baby boomer generation and the increases in life expectancy for Americans at older ages. Medicaid rolls grew during the
recession, by design, and grew due to expansions in coverage under the Affordable Care Act.
Whether or not the current rules for Medicaid eligibility are too lenient or too stringent is the subject of debate but can be separated from debates about the costs of insurance. Private insurance coverage levels vary with conditions in the labor market and have been climbing slightly in recent years.4
Total spending is the number of people covered times the cost per person of coverage. The cost per person for coverage is determined by the numbers of health care products and services used and the prices paid for those services – plus insurer costs for administration and profit.
Those in turn are determined by factors that economists group into supply-side and demand-side factors. The numbers of products and services used are dependent on the supply of those services and how accessible they are. We have millions of people employed by the health care industry and thousands of hospitals, medical offices, and pharmacies across the country. This infrastructure of professionals and providers is built around a health care financing system in which, by and large, providing more services brings in more revenue. To counter incentives to deliver more services, managed care plans put prior authorization requirements in place. Increasingly, however, insurers are using payment methods and quality measurement to encourage the delivery of high-value care and discourage overutilization of low-value care.
On the demand side, there are also clearly interactions between the prices of health care products and services, and how many products and services people use. People with insurance are insulated from the full prices of care but do face deductibles and cost-sharing requirements. The approximately 10 percent of the population who are uninsured also use health care services, financed largely through patient out-of-pocket payments and federal and state programs that support hospitals, community health centers, and other providers. Overall, the uninsured use fewer services than those with insurance,5 and when they do use health care services some providers charge on a sliding scale and sometimes pay higher prices because they pay the “list price” rather than a price negotiated by an insurer. Changes in the prices faced by patients, either because of what is charged or how generous their insurance is, affect demand for insurance and for care. Demand-side factors also include how healthy or sick people and populations are, and how much they can afford to spend on health care.
Growth in health care spending is thus fueled by growth in numbers of people served, numbers of products and services on offer, the prices paid for those services, and how much demand there is for them. It is also fueled by expectations about all of those factors, because those expectations drive investments in facilities and in research and development of new technologies. Indeed, health economists generally attribute about half of growth in health care spending in the United States to the growth of new technologies.6
It is also important to mention the commonly accepted figure that about 30% of what we spend on health care is waste – or expenditure that brings little or no benefit to patients.7 While this is an enormous sum, there is little consensus on how to define waste in practice and even less on how to substantially reduce it. What seems more fruitful is to focus on the health care system features that give rise to such a wasteful set of structures for delivering health care.
Accordingly, health economists, including my colleagues at the Congressional Budget Office, often focus on “excess cost growth” rather than spending levels when talking about the sustainability of health care spending. Excess cost growth is growth in per capita health care costs above growth in per capita GDP. In other words, it is growth in health care costs that outpaces the ability of our society to pay for it. Arguably, as high as spending is, our society is paying for the health care system we have now and recent projections of GDP growth for next year are strong. Of concern is whether the lower rates of health care cost growth in recent years can be sustained while the economy grows overall.
Low Spending Growth in Medicare
As mentioned above, the recent decade of low per capita cost growth in the Medicare program is an instructive example. From 2007 to 2015, total Medicare expenditures increased 50%, but much of this growth was due to the number of Medicare beneficiaries covered. Indeed, on a per capita basis, Medicare spending has been lower than per-capita GDP growth from 2010-2016. The figure below puts this in context: Medicare per capita spending growth has been low both in relation to prior decades and to national health spending overall.
The figure shows that average annual growth in Medicare per capita spending was 1.4 percent between 2010 and 2016, down from 7.1 percent between 2000 and 2010, due in part to reductions in payments to providers and plans and to an influx of younger beneficiaries from the baby boom generation aging on to Medicare (who have lower per capita health care costs.)8 According to the 2018 Medicare Trustees Report, Medicare per capita spending is projected to grow at an average annual rate of 4.6 percent over the next 10 years.9 The trustees project this level of growth due to their forecasts of increased use of services, intensity of care, and rising health care prices – but those factors are affected by policy choices. If choices are made that keep Medicare per capita spending growth below the rate of GDP growth, that will relieve spending pressure on the federal government and have implications for private payers as well.
The volume – or number – of health care services delivered to Medicare beneficiaries has also been relatively flat. Indeed, according to figures from MedPAC, the volume of inpatient hospital services has declined: inpatient discharges per beneficiary declined almost 20 percent between 2006 and 2015.10 Some of that decline was due to a shift from inpatient to outpatient care settings, but the decline still represents a decline in the amount paid for such services.
And, unlike in prior periods of time, new inpatient technologies didn’t replace those moving to the outpatient sector. Indeed, when my colleagues and I adjusted per capita Medicare costs for payment rate increases we found that they have been relatively flat since 2007, indicating that use of care per beneficiary has been similarly stable. These low rates of growth in the use of health care services, and especially inpatient services, have been found in the private sector as well.11
In work I did while at the Congressional Budget and have since updated, my colleagues and I also found that slow growth in Medicare payment rates contributed to the slowdown in per capita cost growth.12 We found that prices have increased more rapidly in private insurance than in Medicare for inpatient, outpatient, and physician services. The exception to this pattern has been drugs – both Medicare Parts B and D have seen years of dramatic growth in spending and spending per prescription – for example around the introduction of the very expensive drugs for hepatitis C. The amount allocated to Medicare Advantage (MA) has also increased, but that reflects the increased enrollment in Medicare Advantage plans: prices paid to providers by MA plans track Medicare prices because of a rule that allows MA insurers to pay out-of-network providers at the Medicare rate.13
What accounts for the slow growth in Medicare? While academics debate the portion of the credit that should go to each factor and how they should be grouped, there is general agreement that the following factors collectively explain the slowdown in spending growth.
• Changes to Medicare payment rates: Sequestration, the slow growth in physician payments and the temporary freezing of physician payment rates under MACRA, and the Medicare hospital productivity adjustments have all contributed directly to lower spending and slower spending growth. Prices paid by private insurers have risen more.
• Changes to Medicare’s payment methods: The Congress and CMS have both signaled a strong interest in moving towards more risk-based and value-based payments for providers under Medicare. While individual payment demonstration projects have yielded only modest savings, if any, the orientation towards value has driven changes in alignment and investment by providers.
• Changes on the consumer/demand side: Although cost-sharing requirements have been very stable in Medicare, Americans have become more exposed to health care costs overall and new retirees may have had experience with high-deductible plans that leads them to be more cautious users of health care services. In addition, while the levels are still high and growing, the rate at which Americans are developing chronic diseases appears to have attenuated somewhat, which helps to keep health cost growth down.
These lessons learned from Medicare’s experience can inform policy choices that might prolong the cost growth slowdown and be applied to other sectors.
Addressing the Drivers of Health Spending
What then are policy options that the Congress can and should consider to ensure that we get the most from what we spend on health care -- and that health spending does not crowd out more valuable goods and services? There are three areas, looking forward, that merit attention from policymakers.
1. Seek innovative ways to make sure that drugs are affordable and appropriately utilized.
The Committee has had numerous hearings on the issues of drug development and pricing and has heard ideas from experts in these areas. As new specialty drugs are projected to be a major driver of cost increases in the future, it is important to ensure they are accessible to patients but targeted only to those likely to benefit from them.
Signals about future prices and value standards will influence the drugs developed and the prices at which they are brought to market. The utilization of medicines for managing chronic diseases is also important and could contribute to offsetting lower medical service costs.14
2. Continue vigilance on payment rates.
Congress, with the help of MedPAC, has a rigorous system for evaluating Medicare payment rate increases. MedPAC has begun to focus on the costs of efficient providers – rather than the average provider – in making recommendations about payment rates. This is an important development, as payment rates benchmarked to standards of efficiency should create incentives to invest in cost-saving technologies and operational procedures. It is also important as most private payments are benchmarked to Medicare rates.
In addition, research has consistently shown that provider consolidation in the health care industry raises prices. Congress should monitor merger and consolidation trends in the health care industry and support more research to better understand how to mitigate those effects.
3. Continue to advance value-based payment methods including episodes/bundles and more comprehensive risk-bearing models.
It is important that the federal government continue to pilot new payment models and to expand the models found to save money without compromising quality. This sends a strong signal to the health care industry that it should invest in information systems, care coordination initiatives, and a population health orientation. The federal government should also support multi-payer payment reforms because they are more likely to reduce spending over the long term than reforms implemented by one sector and it should continue to develop better methods of measuring quality of care.
On the consumer demand side, I am less optimistic about opportunities to contain cost growth without doing harm. While work I conducted with colleagues at RAND suggests that highdeductible health plans can reduce health care spending, the effects are attenuated by accounts like HSAs. We also found evidence that consumers cut back on investments in preventive care when faced with high deductibles (even when preventive care is exempt from deductibles and cost-sharing.)15 Subsequent work has confirmed these findings, and found that price transparency tools did not improve the care choices of high-deductible plan enrollees.16 Given this, and the high levels of health care expense already borne by Americans, efforts focused on the suppliers of health care are more likely to attenuate cost growth without adversely affecting health outcomes.
For all of the concern about health care costs, we do have one of the most advanced health care systems in the world, albeit one that does not serve all citizens equally well. We have gleaming hospitals that employ thousands of people in communities across the country, and nearly every day brings stories of medical breakthroughs like immunotherapy. In other words, our costs are also cures, jobs, and incomes – and thus stemming their growth is not without challenges and costs of its own.
1 Note that this figure does not include the cost of administration or insurer profits.
http://www.milliman.com/uploadedFiles/insight/Periodicals/mmi/2018-mill…
2 David I. Auerbach, Arthur L. Kellermann. “A Decade of Health Care Cost Growth Has Wiped Out Real Income
Gains for an Average US Family.” Health Affairs, v. 30, no. 9, Sep. 2011, p. 1620-1636.
3 Per capita costs would ideally be adjusted for changes in the age and health status distribution of the population
in questions as well. See https://www.healthaffairs.org/do/10.1377/hblog20150728.049597/full/ .
4 https://www.cdc.gov/nchs/data/nhis/earlyrelease/TrendHealthInsurance196…
5 Katherine Baicker, Ph.D., Sarah L. Taubman, Sc.D., Heidi L. Allen, Ph.D., Mira Bernstein, Ph.D., Jonathan H.
Gruber, Ph.D., Joseph P. Newhouse, Ph.D., Eric C. Schneider, M.D., Bill J. Wright, Ph.D., Alan M. Zaslavsky, Ph.D.,
and Amy N. Finkelstein, Ph.D. for the Oregon Health Study Group. “The Oregon Experiment — Effects of Medicaid
on Clinical Outcomes.” N Engl J Med 2013; 368:1713-1722.
6 Newhouse JP. An iconoclastic view of health cost containment. Health Aff (Millwood). 1993; 12 Suppl:152-71.
7 Smith M, Saunders R, Stuckhardt L, McGinnis JM, eds. Committee on the Learning Health Care System in America,
Institute of Medicine. Washington, DC: National Academies Press; 2012. ISBN: 9780309260732.
8 Juliette Cubanski and Tricia Neuman. The Facts on Medicare Spending and Financing. KFF, June 2018.
9 2018 Annual Report of the Boards of Trustees of the Federal Hospital Insurance and Federal Supplementary
Medicare Insurance Trust Funds accessible at https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-
Trends-and-Reports/ReportsTrustFunds/Downloads/TR2018.pdf .
10 MedPAC 2017 Data Book. http://www.medpac.gov/docs/default-source/databook/
jun17_databookentirereport_sec.pdf?sfvrsn=0.
11 HCCI 2016 Cost and Utilization Report accessible at
https://drive.google.com/file/d/1vi3S2pjThLFVwB7OtYwFmOiLVPTFl_wk/view.
12 Michael Levine and Melinda Buntin. “Why Has Growth in Spending for Fee-for-Service Medicare Slowed?“
Congressional Budget Office Working Paper. August 2013.
13 Jared Lane K. Maeda PhD, MPH, Lyle Nelson, PhD. How Do the Hospital Prices Paid by Medicare Advantage
Plans and Commercial Plans Compare With Medicare Fee-for-Service Prices? Inquiry. First Published June 11, 2018.
14 Melinda Buntin and Tamara Hayford. Offsetting Effects of Prescription Drug Use on Medicare’s Spending for
Medical Services. CBO Report. November 2012.
15 Melinda B. Buntin, PhD; Amelia M. Haviland, PhD; Roland McDevitt, PhD; and Neeraj Sood, PhD. “Healthcare
Spending and Preventive Care in High-Deductible and Consumer-Directed Health Plans.” Am J Manag Care.
2011;17(3):222-230.
16 Brot-Goldberg, Zarek, Amitabh Chandra, Benjamin R. Handel, Jonathan T. Kolstad, 2017. "What does a
Deductible Do? The Impact of Cost-Sharing on Health Care Prices, Quantities, and Spending Dynamics*," The
Quarterly Journal of Economics, vol 132(3), pages 1261-1318.