Senate Finance Subcommittee Issues Testimony From University of Chicago Dean

WASHINGTON, Feb. 18 -- The Senate Finance Subcommittee on Fiscal Responsibility and Economic Growth, issued the following testimony by Katherine Baicker, dean of the Harris School of Public Policy at the University of Chicago, involving a hearing on Feb. 2, 2022, entitled "The Hospital Insurance Trust Fund and the Future of Medicare Financing".

Baicker is also a health economics researcher.

* * *

SUSTAINABILITY AND VALUE IN THE MEDICARE PROGRAM

My name is Katherine Baicker, and I am Dean of the Harris School of Public Policy at the University of Chicago and a health economics researcher. I would like to thank Senator Warren, Senator Cassidy, and the Distinguished Members of the Committee for giving me the opportunity to speak today about sustainability and value in the Medicare program. I serve on a number of boards and advisory panels, but am presenting my own views. This statement draws on several pieces I have written in this area, as well as research conducted by many others.

The importance of access to health care and the financial protections that insurance should provide have never been more salient, and Medicare is a vital part of our health care system for millions. Ensuring that Medicare maximizes health benefits within a financially sustainable system requires careful attention to insurance design and balancing tradeoffs across multiple dimensions of coverage and payment structure.

Payment models to promote value

Moving towards models of paying for value, rather than volume, is a crucial step in ensuring that health care resources are well spent. As Michael Chernew and I have written,/1 there are promising strategies for improving quality and reducing ineffective spending that rely on reforming the way we pay for health care to align providers' incentives to improve value. Providers' judgment is crucial in finding ways to reduce waste and help patients chose the most efficient sites and types of care based on their health care needs. Giving providers a financial stake in driving value can be much more effective than reforms that focus solely on patients' incentives or rely on inflexible government coverage rules.

* * *

1 Baicker, Katherine and Michael Chernew, "Alternative Alternative Payment Models," JAMA Internal Medicine 177, no. 2 (2017 Feb 1): 222-223.

* * *

There have been a number of experiments with promising mechanisms, but these have been implemented in limited ways and with limited financial consequences - and thus with limited effects. Some payment models focus on episodes of care, using bundled payments to incentivize providers to limit spending during the episode while achieving quality benchmarks. The savings usually accrue to the hospital or specialist responsible for the episode. Evidence on the effectiveness of such models is mixed. Some studies suggest that bundled payments for joint replacement may reduce spending - both for the patients covered by that payment model and for others treated by the same providers./2,3

Much of the savings may derive from reductions in post-acute care utilization, highlighting the importance of how broadly bundles are defined./4

Other studies find much smaller changes in spending./5

Attention needs to be paid to incentives to select healthier patients or shift costs of care downstream, as well as to the "extensive margin" of the number of bundles, not just the "intensive margin" of the cost per bundle - the risk of offsetting volume increases. Furthermore, bundled payments as currently constructed cover only a small fraction of Medicare spending, and savings to date have been driven by a small subset of episode types./6,7

Other approaches focus on total population spending, such as Accountable Care Organizations. These models provide incentives for provider groups to reduce per-capita spending and improve quality. The savings generally accrue to the organization that employs the primary care provider. Population-based payments have the potential to cover a greater share of spending and thereby have a bigger impact system-wide, although savings to date have been modest./8

It is important to note that the Medicare program doesn't capture all of the savings in either model - a large share of savings are "shared" with providers. Potential savings to the Medicare program depend on how the benchmark payments are set. In the episode-based models, benchmarks are often set a bit below estimated spending, guaranteeing that Medicare will reap some savings, but any greater savings would go to providers. The share of savings providers get to keep - and their risk of loss - drives the incentives to improve efficiency. Over time, savings to Medicare could grow if benchmarks rose more slowly than they otherwise would.

* * *

2 Finkelstein, Amy, Einav, Liran, Ji, Yunan, and Mahoney, Neale, "Randomized Trial Shows Healthcare Payment Reform Has Equal-Sized Spillover Effects on Patients Not Targeted By Reform," PNAS 117, no. 32 (2020 Aug 11): 18939-18947.

3 Finkelstein, Amy, Ji, Yunan, Mahoney, Neale, and Skinner, Jonathan, "Mandatory Medicare Bundled Payment Program for Lower Extremity Joint Replacement and Discharge to Institutional Post-acute Care Interim Analysis of the First Year of a 5-Year Randomized Trial," JAMA 320, no. 9 (2018): 892-900.

4 Navathe, A.S., Troxel, A.B., Liao, J.M., et al, "Cost of Joint Replacement Using Bundled Payment Models," JAMA Intern Med 177, no. 2 (2017): 214-222.

5 Dummit, L.A., Kahvecioglu, D., Marrufo, G., et al, "Association Between Hospital Participation in a Medicare Bundled Payment Initiative and Payments and Quality Outcomes for Lower Extremity Joint Replacement Episodes," JAMA 316, no. 12 (2016): 1267-1278.

6 Urdapilleta, O., Weinberg, D., Pedersen, S., Kim, G., Cannon-Jones, S., Woodward, J., "Evaluation of the Medicare Acute Care Episode (ACE) Demonstration: Final Evaluation Report," (2013): http://downloads.cms.gov/files/cmmi/ACE-EvaluationReport-Final-5-2-14.pdf.

7 Dummit, L., Marrufo, J., Marshall, J., et al, "The Bundled Payments for Care Improvement Initiative Models 2-4: Year 2 Evaluation & Monitoring Report," (2016): https://innovation.cms.gov/Files/reports/bpci-models2-4yr2evalrpt.pdf.

8 McWilliams, J.M., "Savings From ACOs--Building on Early Success," Ann Intern Med (2016): http://annals.org/aim/article/2566329/savings-from-acos-building-early-success.

* * *

As discussed below, Medicare Advantage is an increasingly attractive option for beneficiaries, giving beneficiaries a choice among private plans that often come with expanded benefits, more active management, and limitations on provider networks. None of these alternatives seems to lower the quality of care, but both payment structure and risk adjustment should be constructed with an eye to promoting quality and access./9,10,11

Alternative coverage and payment models could thus be substantial improvements over the fee-for-service system that dominates Medicare now, but the effectiveness of these tools will depend on having broad scope and real financial stakes; and the way that payments are calibrated will drive the share of any savings that accrues to the Medicare program (taxpayers), providers, and patients.

Patient choices and financial protection

Another driver of access, utilization, and value is the set of cost-sharing and coverage parameters faced by patients - what services are covered, which providers are included, and how much patients are expected to pay out-of-pocket. Patient cost-sharing is often perceived as merely a mechanism to shift costs to patients that results in restricted access to needed care, but more nuanced use of patient cost-sharing can be a powerful way to promote better use of health care resources without creating barriers to needed care./12

The traditional economics model is based on patients having detailed information about the value of care options and the ability to implement choices that fully incorporate their preferences and priorities. Health insurance protects patients against the financial risk of needing expensive care,/13 but it also generates use of care that is of lower value by insulating patients from having to pay the true cost of care (which they would only do if it was worth it to them in terms of improved health) - described as "moral hazard" by economists. There is ample evidence that higher copayments do reduce the use of health care./14,15

* * *

9 McWilliams, J. Michael, Hatfield, Laura, Chernew, Michael., et al, "Early Performance of Accountable Care Organizations in Medicare," New England Journal of Medicine 374 (2016): 2357-2366.

10 Baicker, Katherine and Jacob Robbins, "Medicare Payment Policy and System-Level Health-Care Use: The Spillover Effects of Medicare Managed Care," American Journal of Health Economics 1, no. 4 (2015 Fall): 399431.

11 Baicker, Katherine, Chernew, Michael, and Robbins, Jacob, "The Spillover Effects of Medicare Managed Care: Medicare Advantage and Hospital Utilization," Journal of Health Economics 32, no. 6 (2013 Dec): 1289-1300.

12 Baicker, Katherine, "Rethinking Health Insurance Design," JAMA Health Forum 2, no. 5 (2021 May 13): e211440.

13 Baicker, Katherine and Levy, Helen, "Cost Sharing as a Tool to Drive Higher-Value Care," JAMA Internal Medicine 175, no. 3 (2015 March): 399-400.

14 Newhouse, J.P., Insurance Experiment Group, Free for All? Lessons From the RAND Health Insurance Experiment (Cambridge, MA: Harvard University Press, 1993).

15 Baicker, Katherine and Goldman, Dana, "Patient Cost-Sharing and Healthcare Spending Growth," Journal of Economic Perspectives 25, no. 2 (2011): 47-68.

* * *

Patient cost-sharing would ideally balance the positive effect of risk protection and the negative effect of excess use. In this simple world, if a $10 copay deters use of a service, that indicates that the patient valued the health benefit of the service at less than $10 - with the copayment thus deterring use of low-value care with little health benefit. Copayments should be highest for care where patients are most price-sensitive, since that is the care of least value to them. Higher copayments wouldn't increase patients' total costs in this simple case, since health insurance premiums would be commensurately lower.

But, of course, that simple model does not capture the complex reality of the difficult choices patients have to make, often in fraught circumstances, under time pressure, and with incomplete information. There is strong evidence from behavioral economics, medicine, and psychology that higher copayments reduce use of high-value as well as low-value care. This is of particular importance for low-income patients, but is also seen in higher income populations for whom the copay is not unaffordable. The health costs of reductions in care in response to even modest copays, which we've called "behavioral hazard," can be severe, reflecting the real-world limitations in decision-making that all patients face, such as limited information, limited time, challenges in follow-through, and misperceptions of risk./16,17

These findings can inform the design of nuanced cost-sharing that is a positive force for higher value care./18

Copays could be higher for care that is of questionable health benefit, and lower (sometimes free, or even negative) for care that is of high health benefit. Such insurance design could simultaneously improve the important financial protection that health insurance offers for enrollees and substantially improve health outcomes./19

While such insurance design improvements would not necessarily generate savings for the Medicare program itself, they could amplify the effectiveness of provider payment reforms, indirectly benefiting the program's finances as well as enrollees. Reforms to Medigap in particular could be helpful in this regard.

Competition to foster innovation, affordability, and value Competition among insurance plans can be a powerful driver of innovation that both improves health outcomes and reduces prices. The Medicare Advantage program provides an example of giving enrollees a choice among publicly-funded insurance options.

One advantage of having multiple plans available is that different enrollees have different preferences and priorities - both for the total share of resources they would like devoted to health care and for the types of features in health insurance that they value the most, such as tradeoffs between lower copayments, more expansive networks, lower premiums, and more comprehensive coverage./20

* * *

16 Chandra, Amitabh, Flack, Evan, and Obermeyer, Ziad, "The Health Costs of Cost-Sharing," NBER Working Paper No. 28439 (2021 Feb).

17 Baicker, Katherine, Mullainathan, Sendhil, and Schwartzstein, Joshua, "Behavioral Hazard in Health Insurance," Quarterly Journal of Economics 130, no. 4 (2015 Nov): 1623-1667.

18 Baicker, Katherine and Levy, Helen, "Cost Sharing as a Tool to Drive Higher-Value Care," JAMA Internal Medicine 175, no. 3 (2015 March): 399-400.

19 Chernew, Michael E., Rosen, Allison B., and Fendrick, A. Mark, "Value-Based Insurance Design," Health Affairs 26 (2007): w195-w203.

20 Baicker, Katherine and Chandra, Amitabh, "What Values and Priorities Mean for Health Reform," New England Journal of Medicine 383, (2020 Oct 8): e89.

* * *

Since Medicare's inception, health care has gotten much more complex and expensive, income disparities have widened, and the cost to taxpayers has increased dramatically./21 Adding flexibility along multiple dimensions - along with subsidies to ensure that robust insurance is affordable across the income distribution - can leave everyone better off.

Another advantage of plan competition is that it has the potential to drive down costs and accelerate valued innovation./22 This requires true competition within the insurer market (as well as among clinicians, hospitals, innovators, and other health care institutions), however, which is not the case in many parts of the country.

The Medicare program provides vital access to care and financial protection for millions of Americans. Ensuring that it provides the best health outcomes possible while maintaining financial sustainability and affordability - both for individual beneficiaries and for the taxpayers of today and tomorrow who must fund the benefits - should be a policy priority. Evidence points to opportunities to reform provider payments and benefit design to focus health care resources where they will do the most to improve health and wellbeing.

I thank you again for this opportunity, and look forward to answering any questions you may have.

* * *

21 Shepard, Mark, Baicker, Katherine, and Skinner, Jonathan, "Does One Medicare Fit All? The Economics of Uniform Health Insurance Benefits," Tax Policy and the Economy 34 (2020): 1-41.

22 Dafny, Leemore and Lee, Thomas, "Health Care Needs Real Competition," Harvard Business Review (2016 Dec).