Kosali Simon, the Herman B. Wells Endowed Professor in health economics and associate vice provost for health sciences at the Paul H. O’Neill School of Public and Environmental Affairs at Indiana University, opened the session by highlighting the unprecedented pressures the COVID-19 pandemic has placed on health care spending as well as on public health delivery and infrastructure. Spending on health care accounts for one-sixth of all spending in the U.S. economy (CMS, 2020). Furthermore, prior to the COVID-19 pandemic, projections indicated that health care spending would increase at a rate of 1.1 percent faster than the U.S. gross domestic product (GDP) from 2019 to 2028, increasing health care’s overall share of the economy from 17.7 percent in 2018 to 19.7 percent by 2028.1 In the context of the pandemic, despite a decrease in health care spending in the early part of 2020, spending still grew faster than in 2019, Simon said. Panelists Mac McCullough, Katherine Baicker, and Monica Bharel then described and discussed trends in health care spending, the opportunity costs associated with health expenditures that result in little to no health benefit, and the question of reallocating some portion of health spending to support other social determinants of health.
A HEALTH DIVIDEND FOR AMERICA: U.S. HEALTH CARE EXPENDITURE AND OPPORTUNITY COSTS
J. Mac McCullough, an associate professor at the Arizona State University College of Health Solutions and a health economist at Maricopa Public Health Department, emphasized that per capita health spending in the United States surpasses that of all other nations (McCullough et al., 2020). However, U.S. life expectancy lags behind that of other member countries in the Organisation for Economic Co-operation and Development (OECD) (Kindig et al., 2018). McCullough posited two potential sources of this disconnect between health care spending and health outcomes: (1) not all drivers of health outcomes—including social and physical environments—are touched by health care spending, and (2) not all health care spending affects health outcomes.
Disconnect Between Health Care Spending and Health Outcomes
Although some have argued that the United States spends close to the OECD mean for social spending, McCullough pointed out that is only the case when spending on pensions and on health care are combined
1 For U. S. spending on health, see CMS (2020), and for an analysis of the pandemic’s effects of health care spending, see https://www.healthsystemtracker.org/chart-collection/u-s-spending-healthcare-changed-time (accessed August 31, 2021).
(Papanicolas et al., 2018). When looking solely at family benefits spending (e.g., cash benefits for children, paid parental leave), which is more closely related to spending on the core social determinants of health, the United States ranks second to last among OECD nations.2 The United States spends less on underlying social and physical environments than on clinical care, and not all health care spending has an effect on health outcomes, he added.
Health spending that does not improve health is referred to as “wasteful spending,” and systems-level research indicates that removing these expenditures from the system would have no negative effects on health, McCullough said (Berwick and Hackbarth, 2012; Fredell et al., 2019; Kelley, 2009; PricewaterhouseCoopers’ Health Research Institute, 2008; Shrank et al., 2019). Studies have come up with different figures for the total aggregate magnitude of this waste, with estimates for 2019 ranging from $601 billion to $1,905 billion, with a median of $879 billion (Speer et al., 2020). Areas of wasteful health spending include clinical inefficiencies, missed prevention opportunities, overuse, administrative waste, excessive prices, and fraud and abuse. McCullough noted that there are interdependencies and overlap among these categories. Furthermore, the impacts of wasteful medical spending extend beyond a lack of health value to the creation of opportunity costs.
“Opportunity cost is the loss of potential gain from other alternatives when one alternative is chosen” McCullough said, and dollars spent on wasteful medical care are unavailable for use in other areas. For instance, each dollar spent on an unnecessary test or scan in a private health insurance scenario is a dollar less that an employer has available for paying employees, remaining globally competitive, or satisfying shareholders. Funds spent on overpriced services are no longer available to the federal government to apply toward reducing debt or other public priorities. Similarly, resources used for administrative overhead cannot be allocated toward social or infrastructure priorities that add benefit to health outcomes. McCullough suggested that when medical care spending improves health, allowing opportunity costs may be reasonable. However, when these expenses do not result in improved health outcomes, those funds may be better allocated toward health-promoting strategies, to non-health priorities, or to profit, rent, or wages. McCullough noted some caveats to this argument, however. First, the “wasted” funds spent on unnecessary medical
2 Data on public spending by OECD nations can be found at https://data.oecd.org/socialexp/family-benefits-public-spending.htm (accessed May 14, 2021).
care represent waste to the individual or entity paying for the care. Second, while an intervention may not result in improved health, there may be some theoretical value to the reassurance of having laboratory work performed or by seeing a specialist. Third, while “wasteful spending” may not be of benefit to the individual or entity paying for it, this medical care may represent revenue to providers and suppliers.
The “Health Dividend” of the Opportunity Cost
Not only does wasteful spending result in a lack of return on the expense, but the associated opportunity cost also represents the loss of potential gain were the money to be spent in other areas, McCullough explained. Given the estimated $879 billion spent annually on low- or no-value health care, the priorities that could be addressed with this sum are virtually unlimited (Speer et al., 2020). Furthermore, spending could be shifted to areas that have a positive impact on health, from specific evidence-informed health care services to non-clinical programs addressing social needs, quality of life, housing, infrastructure, and climate. McCullough outlined a scenario in which the $879 billion of public and private funds spent on non-value-added health care were reallocated; the resulting public share would be $396 billion, and if half were applied toward deficit reduction, $198 billion—or $596 per capita—would remain available to invest in alternative public uses.
McCullough explained that in this public reinvestment scenario, $123 billion in funding is designated for social programs (McCullough et al., 2012). This could include $3.3 billion to provide regular home visits by nurses for 500,000 pregnant smokers and pregnant teenagers, an intervention that has been causally linked with a reduction in emergency department (ED) visits in the first two years of life and in the incidence of low birth weight. Nine billion dollars would enable 1.1 million students to participate in an evidence-based social development program of the sort that has been causally linked with decreases in risky sexual behavior and drug use as well as improved work, social, and emotional functioning as adults. With an additional $14.7 billion, Head Start—which currently serves only
3 McCullough references data based on updating the research first published in McCullough et al. (2012) and informed by sources provided in Appendix A (ASCE, 2020; Agosto and Hughes-Hassell, 2009; County of Los Angeles, 2019; Farrelly et al., 2005; U.S. GAO, 2019a; Garces et al., 2000; Gomby et al., 1995; House Committee on Transportation, 2020; Larimer et al., 2009; Lonczak et al., 2002; Ludwig and Phillips, 2007; Muennig and Woolf, 2007; Nord and Prell, 2009; Olds et al., 1986a,b, 2004; Schweinhart et al., 2005; SEED, 2020; Statista, 2021; Thrush, 2018).
half of eligible children—could be expanded to serve all eligible youth. Head Start interventions have been linked with a reduction in childhood obesity and decreased smoking prevalence later in life, McCullough noted. Universal pre-kindergarten, which has been causally linked to long-term gains in cognitive ability and socialization, could be provided to all non-Head Start participants for $53.6 billion. Class sizes could be decreased to a maximum of 17 students per classroom for all students in grades 1–6 for $57.7 billion. Decreased class sizes have been causally linked with increased graduation rates and gains in life expectancy.
In addition to social programs, this scenario includes $49 billion for quality-of-life programs, McCullough said. This allocation could enable investment into green spaces in the built environment and provide a community facility and rural economic development grant to every small town in America over 10 years. Furthermore, every K–12 school in the United States could receive a safe routes to school grant. The Job Corps program could be doubled in size. Universal basic income of $500 per month could be provided to individuals in low-income neighborhoods. McCullough noted that a recent randomized trial found that providing universal income at this level is associated with a host of positive impacts (Kornfield, 2021).4 Additionally, all public libraries in the United States could be expanded, investment in water infrastructure and clean drinking water could be doubled, and the Supplemental Nutrition Assistance Program (SNAP) could be extended to all individuals experiencing food insecurity.
Housing, Infrastructure, and Climate Programs
McCullough described how this scenario would allocate $613 million toward housing, offering the Housing First interventions to all chronically homeless individuals with severe alcohol problems. This program has been linked to improved health outcomes, and it yields cost savings when considering all societal costs, such as reduced ED visits (Larimer et al., 2009). Infrastructure spending of $17.6 billion could fulfill the public transit backlog (i.e., unaddressed costs for transit agencies around the country) over 10 years (ASCE, 2020). Allocating $1.3 billion toward climate spending could enable the implementation of a disaster mitigation and adaptation fund, a potential investment that was identified by the U.S. Government Accountability Office (U.S. GAO, 2019a).
4https://www.stocktondemonstration.org/#summary-of-key-findings (accessed November 15, 2021).
Health Dividend Considerations
The array of investments described previously is only one of countless combinations of funding priorities that could be met through the reallocation of wasted health care spending, McCullough said. Among the myriad possibilities are the creation of behavioral crisis intervention teams, full funding of the nation’s public health system, the implementation of equity initiatives such as social impact bonds, and the resolution of Medicare insolvency. Moreover, the programs outlined in this scenario only account for $198 billion of the total $879 billion that stands to be reallocated into debt reduction or the private sector. McCullough noted that a fundamental principle underlying the idea of the health dividend is that the current prioritization of non-value-added services over evidence-based programs carries an opportunity cost. Reprioritizing evidence-based social, quality-of-life, housing, infrastructure, or climate programs may yield a health dividend above and beyond the health outcomes generated by current health care spending. Even without increasing expenditure, reallocating spending toward non-medical initiatives known to improve health could generate a health dividend, defined as “a sizeable stream of resources that would come at no net cost to people’s health and that could be invested in achieving two important objectives: stabilizing the nation’s fıscal health and improving well-being” (McCullough et al., 2012, p. 651). The net budgetary effect of the initiatives discussed would be to reduce governmental spending on health and well-being, yet it would still be expected to result in a positive net health effect, he said.
The existence of wasteful health care spending and underfunding of initiatives that could generate improved health points to the need to reduce health spending, McCullough said, but entrenched interests and politics present challenges. Prior to the COVID-19 pandemic, the United States consistently saw year-over-year growth in health services spending (Peterson-KFF Health System Tracker, 2020). Reaching the OECD median in health care spending would require health care spending to decline by 7 percent per year for the next decade. Paradoxically, the scale of this issue is both a challenge and an opportunity, he remarked. The magnitude of $879 billion is so large that it can depersonalize the impact that such an enormous figure can have. Moreover, this figure only considers the portion of health spending deemed to add zero benefit and does not take into account low-value care that may result in additional opportunity costs. Equity issues should also be considered, McCullough said, given that current resource allocation results in some people having no access to the health system while others have access to excess services that do not result in medical benefit.
EXPANDING HEALTH ACCESS WHILE DECREASING WASTEFUL HEALTH SPENDING
Katherine Baicker, the dean and Emmet Dedmon Professor at the Harris School of Public Policy at the University of Chicago, outlined two potential goals of health system reform: (1) to expand access to care for populations with insufficient access to valuable treatments, and (2) to improve the value of the health care delivered, maximizing the health benefit of every dollar spent. Ensuring that every health care dollar is spent on services that produce substantial health improvement does not necessarily mean spending less, Baicker said, but it likely means allocating health care resources differently.
Although research indicates that some portion of health care spending is wasted, estimates of the amount of that waste vary greatly. Furthermore, it can be difficult to pinpoint the expenditures that are wasteful. For instance, an MRI may not result in a benefit for one patient, but for another it may result in a life-saving diagnosis. Evidence indicates that there is substantial low-value spending in the U.S. health care system. The challenge is to create a health system that reduces the use of low-value care while maintaining access to high-value care and driving innovation that extends new treatments to as many people as possible. While it is clear that waste, fraud, abuse, and care that produce zero or negative health benefit should be eliminated, a substantial portion of care provides positive benefits to patients but at a very high cost, and making determinations about whether or not that small benefit justifies the expense is complex, Baicker said.
Baicker suggested that it is important to improve the incentives offered to providers and to patients to expand access and encourage the use of high-value care. Baicker emphasized the need for a more sophisticated insurance design that adjusts the amounts patients pay for care based on their income and on the value of the treatment. For example, statin drugs prescribed to lower cholesterol in patients with diabetes are often very cost-effective, and such patients could have no copays or even be subsidized to take the medication. In contrast, statins may be a wasteful expenditure for people with elevated cholesterol but no other risk factors, and these patients could have higher copayments for the drugs. Similarly, providers could be paid based on the value of care and on patient population outcomes, rather than just for the quantity of services delivered.
Baicker said that price and quantity are not independent levers in the health care system. Policies that change price are likely to affect the quantity of care available and access to that care. Policies should consider root problems as well as the desired outcomes. In some cases, the goal may be to increase usage of underused services and in other cases to decrease usage of overused services. Increasing value does not necessarily mean spending
less. Many preventive services, for example, do not result in net savings, but they do improve health outcomes at a moderate price. More nuanced insurance policies for individuals and payment policies for providers could drive higher use of underused services and limit use of low-value services.
Non–health care inputs play an important role in health, Baicker noted, but incorporating such inputs into health insurance design requires careful targeting. For example, for some patients health outcomes may improve and health spending go down when healthy food is provided as a way of improving disease management or if air conditioners are provided to avert asthma-driven ED visits. However, adding these to health insurance policies as standard benefits could increase overall spending substantially while improving health outcomes for only a few (though providing other non-health benefits). A novel mechanism for adding flexibility coupled with accountability for improving health outcomes would be a pay-for-success contract; for such arrangements, randomized controlled evaluations are a key mechanism for assessing the success of targeting resources to improve health outcomes in a cost-effective way.
HEALTH EQUITY INITIATIVES
Monica Bharel, a commissioner of the Massachusetts Department of Public Health (Mass DPH), emphasized that the United States is spending too much on health care costs without good and equitable health outcomes to match the investment. Increasing health care access and value can be addressed by focusing on equity and accountability in health care spending while decreasing waste. Bharel spoke about the unnecessary waste she witnessed during the 20 years she practiced primary care internal medicine in resource-limited settings. For example, because housing is a key social determinant of health, she asked a homeless patient whether his case manager was helping him obtain housing. He replied that he had five case managers across various programs, and none of them were truly helping him. Additionally, she noted the common practice of discharging a patient after a long, expensive stay in the intensive care unit for addiction-related medical complications and providing the patient with a phone number for a substance abuse provider, as opposed to offering a warm handoff by directly connecting the patient with the provider.
Creating a health system with a true focus on overall health—versus merely treating disease—will require upstream community engagement and policy change on multiple levels, Bharel said. At the individual level, medical providers can improve the quality of their interactions with patients. At the community level, efforts could include population health approaches, such as health systems considering how to engage with asthma patients who may not be receiving care. At the structural and institutional level, Bharel
continued, policy change and bias identification are needed to drive health opportunity for all people.
Bharel said that health equity is embedded in the mission of Mass DPH and that it works to expand equity, opportunities, and choice in health care to all. In 2017 the department opened its Office of Population Health in an effort to use abundant Mass DPH data to better understand the social determinants of health and disparities. Using what it terms a precision public health approach,5 Mass DPH targets communities and populations most in need of attention and resources. In practice, this has involved shifting the organization’s focus from specific conditions and diseases—such as substance abuse disorder, heart disease, and diabetes—to a focus on the populations most burdened by these conditions. Bharel added that, regardless of the particular medical condition, the same populations carry excess burden of disease, often due to social and environmental factors. These populations include: communities of color; individuals with a history of incarceration, homelessness, substance use disorder, or mental illness; and pregnant and postpartum people. Bharel said that precision public health lends itself to addressing disparities with a racial equity lens; the racial disparities observed in COVID-19 outcomes in the United States underscore the need for this approach.
Highlighting state-level efforts to address overall health and health equity, Bharel pointed to the Moving Massachusetts Upstream (MassUP) Investment program, which is aimed at supporting partnerships between health care providers and community organizations to address social determinants of health. MassUP awards state-funded grants to grassroots community innovations intended to achieve food security and economic stability and mobility. For example, a three-way community collaboration among a local medical center, food pantry, and food security council works with food retailers and residents to expand the number of locations offering affordable, healthy food and engages with the community to increase access to existing food benefits and entitlements. Additionally, Bharel said, Massachusetts offers the Community Health Initiative to address the social determinants of health with the goal of eradicating health inequities. Using a data-driven approach, the program looks beyond healthy behaviors to address health priorities such as homelessness and housing instability, substance use disorder, mental health and wellness, and preventable chronic disease. The program is funded by state taxes levied on all health
5 Precision public health is defined in the context of Mass DPH efforts as the work of using big data to allow better understanding of and action on the social determinants of health; see remarks from Dr. Bharel https://catalyst.nejm.org/doi/full/10.1056/CAT.20.0432 (accessed September 21, 2021). Elsewhere, or more broadly, precision public health extends beyond treating illness to “providing the right intervention to the right population at the right time” and includes prevention and early detection efforts (Khoury et al., 2016, p. 1).
system renovation or expansion projects. These projects go through a determination-of-need application process which encourages health systems to explicitly identify the role of structural racism in creating inequitable community conditions by requiring hospitals to describe their strategies to address inequities and the structural causes of inequalities. This initiative has funded community investment projects addressing housing access and grocery store development within food deserts.
Bharel spoke about COVID-19-related community inequities and the public health steps that Massachusetts has taken to mitigate these disparities. The state’s pandemic response involves three pillars: (1) preserving life; (2) preserving the health care system; and (3) equity. In early April 2020, Massachusetts began acquiring demographic data for COVID-19 cases. The following month, the COVID-19 Health Equity Advisory Group convened to understand and address how the social determinants of health create unequal disease burdens between communities. The efforts, Bharel said, included a 2020 statewide community survey to better understand the immediate and long-term health impacts of the pandemic, including access to testing, health care delays, and access to housing, food, safe working conditions, and other factors known to contribute to the COVID-19 burden. By engaging with a network of community partners to intentionally reach populations that are often underrepresented in surveys, the group collected responses from more than 33,000 people. An initial analysis revealed that many groups were disproportionately affected both by COVID-19 and by indirect consequences of the pandemic. These groups included: the lesbian, gay, bisexual, transgender, and queer community; people with disabilities; individuals experiencing homelessness or housing instability; communities of color; individuals with low income and educational attainment; and people who speak languages other than English. Bharel added that the survey results are being used to set research priorities and to address inequities exacerbated by the pandemic. A recently launched health equity initiative focuses on 20 communities in Massachusetts that have been most affected by COVID-19, as indicated by case burden and the Social Vulnerability Index created by the Centers for Disease Control and Prevention (CDC). The initiative also factored in the presence of historical trauma in communities of color and its impact on trust. By providing enhanced assistance, support, and resources, the program seeks to increase community access to COVID-19 vaccines while building trust and confidence in the vaccine. Data are used to inform the allocation of resources that have been determined by community input.
Bharel encouraged the use of disparity data to direct spending toward specific populations that are overrepresented in terms of disease burden. Grant funding can be a mechanism for funding innovation to address the social determinants of health that lead to a disproportionate impact of
disease on certain populations. A population health approach is needed to broaden the definition of health beyond health care and to move away from the false dichotomy separating health care and public health, Bharel said. Taking advantage of data and community voices has the power to create real and lasting change. Moreover, the COVID-19 pandemic serves as a reminder of the immediate need for policy changes aimed at creating health equity.
U.S. Health Policy Needs
Given that the 2010 Affordable Care Act (ACA) is intended to expand health coverage while improving quality and reducing costs and that the American Rescue Plan (ARP) Act of 20216 temporarily increases ACA subsidies, Simon asked how health care policy should be built from this point forward. Baicker said that a key accomplishment of the ACA was expanding access to health care insurance coverage through two mechanisms: Medicaid expansion and the creation of health insurance exchanges. Not all states opted to expand Medicaid, while health exchanges’ insurance premiums have been somewhat volatile. The COVID-19 pandemic has exposed the costs of a patchwork system that lacks continuity of insurance coverage and access to care, Baicker said. Because most privately insured Americans receive coverage through their employers, the pandemic and associated economic recession put people at risk of losing employer-provided health insurance in the midst of a public health crisis. In response, some states opened new enrollment periods for their state-run health insurance exchanges, and on January 28, 2021, newly elected President Joseph R. Biden signed an executive order to open a special enrollment period in the federally run health insurance exchanges. The ARP provides incentives to states that have not expanded Medicaid to do so; it also increases ACA health insurance subsidies to extend to more people unable to afford plans offered on the exchange.
Baicker added that the ACA brought some experimentation with payment reforms that moved toward paying for quality and outcomes rather than just the quantity of care delivered. Accountable care organizations—in which primary care physicians share in the savings associated with higher-value care for their patients—are another tool that could contribute to system change, though incentives and reach are modest to date. She noted that expanded coverage through the ACA creates a more robust safety net and valuable access, but it does not address the cost of health care. As
6https://www.congress.gov/117/plaws/publ2/PLAW-117publ2.pdf (accessed September 17, 2021).
Medicare is one of the largest single providers of insurance and drives much health care use, the effects of moving toward higher-value care in Medicare would be felt systemwide, Baicker said. Many private insurers explicitly or implicitly reference Medicare pricing and coverage, and therefore moving Medicare toward higher value and better outcomes could have outsized system-level effects.
Addressing Pushback to System Change
Noting the entrenched interests and political ramifications of the current system, Simon asked how to approach the complicated issues surrounding any shift of health care dollars toward initiatives that may be more beneficial for advancing health. McCullough suggested focusing on the enormous scope of both the current opportunity cost and the potential payoff of reducing waste while facing the challenges to instigating change. Top-down and bottom-up approaches can be used simultaneously through policy proposals—such as the real-time assignment of beneficiaries into Medicare Choice—and building the political will to build on the ACA and Medicaid expansion to create a paradigm shift in terms of how care is paid for and delivered in the United States. Baicker noted pushback from health care sector employees and from local policy makers who argue that health care employment is an engine of job growth. She remarked that the health care system should not be viewed as a jobs program—one that is very expensive and drives up health insurance costs. Rather, innovative programs should focus on improving health as much as possible for every dollar spent, which might ultimately increase or decrease employment in health care. Bharel emphasized that data are a valuable tool for highlighting the unsustainable nature of current health care spending. She added that another mechanism for change is to examine existing community funding and ensure that it is directed toward community initiatives that truly lead to better health.
Simon asked how the needed health care changes and investments in mitigating the negative effects of the social determinants of health can be implemented in rural settings. Baicker replied that telemedicine—which was underused and not always reimbursed prior to the COVID-19 pandemic—is an example of a potentially cost-effective method of expanding access to primary care, behavioral and mental health, and specialist services not just in rural areas, but also to underserved urban areas. McCullough remarked that opportunity costs are present in rural settings, so examining the money spent on health and the benefit received from those expenditures remains
worthwhile. Bharel commented that each community has unique needs; amplifying the community voice to better understand those needs is important in developing policy.
Market Forces in Care Options
Given that technology is one of the major drivers of health care costs, Simon asked how technology, innovation, market forces, and competition factor into current health care economics. McCullough shared a personal anecdote from graduate school, in which he noticed that smoking cessation was not included in his student insurance coverage. He posited that the student insurance did not cover smoking cessation because most people covered by the plan were not yet old enough to have developed lung cancer. This is an example of a market failure; examining where the market encourages desired health outcomes and identifying areas where it does not encourage those outcomes can inform the development of interventions, McCullough said. Baicker added that many long-term health consequences appear when people are old enough to qualify for Medicare, which limits the incentives for private insurance companies to address health care needs with long-term consequences. While traditional economics might suggest that consumers will incorporate these long-term consequences when choosing an insurance plan, there is ample evidence that such decisions are limited by individuals’ medical expertise and the complexity of offerings. Baicker continued that there is a strong role for both private competition and public policy. Social insurance can ensure broad access and a social safety net, while private sector competition can drive innovation and tailor offerings to meet individual needs. Partnerships between the government and the private sector can be highly effective, as was the case with the development of COVID-19 vaccines.
Bharel remarked that among the many lessons learned during the COVID-19 pandemic are the extent to which humans are interconnected and how an individual’s health decisions can affect other people. Therefore, health can be approached as a public good that all people should feel invested in. To that end, the historically underfunded public health infrastructure needs to be strengthened using an equity-focused, data-driven approach, she added.
Baicker said that many well-intentioned programs are not effective in practice. Evaluation coupled with experimentation can lead to a system that delivers greater value and better health outcomes. The public and private sectors both have roles in driving innovation and addressing the current gaping disparities in access, disease burdens, and health outcomes.
McCullough pointed out that in the United States, innovation tends to focus on creating the best technology, such as a magnetic resonance imaging (MRI) technology with the highest possible resolution. While this has benefits, innovation that focused instead on developing less expensive MRIs might enable greater access to this tool. Because public dollars are being spent on health care, McCullough said, the opportunity costs of various routes to producing health outcomes should be considered.