THE ROLE OF HEALTH POLICY IN IMPROVING HEALTH OUTCOMES AND HEALTH EQUITY

ABSTRACT

Despite higher per-capita health care spending than any other country, the United States lags far behind in health outcomes. Additionally, there are significant health inequities by race, ethnicity, socioeconomic position, and rurality. One set of potential solutions to improve these outcomes and reduce inequities is through health policy. Policy focused on improving access to care through insurance coverage, such as the Affordable Care Act’s Medicaid expansion, has led to better health and reduced mortality. Policy aimed at improving health care delivery, including value-based payment and alternative payment models, has improved quality of care but has had little impact on population health outcomes. Policies that influence broader issues of economic opportunity likely have a strong influence on health, but lack the evidence base of more targeted interventions. To advance health outcomes and equity, further policy change is crucial.

INTRODUCTION

There are persistent and pervasive inequities in health outcomes along racial, ethnic, socioeconomic, and geographic lines in the United States. Given that many of these disparities have their roots in structural inequities, addressing them requires attention to the broader policies that can impact health outcomes. These include policies that influence access to care, policies that influence quality of care, and policies that influence broader social and economic context such as community development, education, and job opportunities.

Health Inequities in the United States

It is well documented that the United States, despite spending roughly twice as much per capita on health care than other high-income nations, has worse health outcomes and lower life expectancy (1). Further, underlying these statistics is evidence of deep, persistent inequities in health outcomes based on several socioeconomic factors. For example, Black individuals have a 20%–30% higher population-level burden of cardiovascular death than White individuals, a difference that narrowed between 1999 and roughly 2010 but has been stagnant since then (2). There are similar health inequities based on poverty, with data showing that poverty is associated with as much as a two-fold increase in annual mortality rates among older adults (3). Geography is also an important source of health inequities, with both neighborhood deprivation (4-6) and rurality (7-15) independently associated with worse health outcomes across a range of medical and surgical conditions. Many of these risk factors are also intersectional, meaning that people can be affected by many at once; the worst health outcomes in the country across the lifespan, from infant mortality to maternal mortality to heart failure mortality, are generally in areas that are poor, rural, and heavily minority, particularly in the southern United States. (10, 16, 17).

When we consider the root causes of these inequities, we can think about factors that influence health broadly. These include genetic factors, such as diseases demonstrated to be under either monogenic or polygenic influence, although typically the risk conferred in polygenic situations is relatively minor. We can consider “lifestyle” factors, such as diet, exercise, sleep, and smoking, which are known to influence a wide range of health outcomes. We can consider emerging risk factors such as mental health, stress, inflammation, and environmental exposures that are increasingly recognized to contribute to heart disease, cancer, and other conditions. We can consider social factors such as poverty, lack of education, or lack of employment. And finally, we can consider health care itself, either via access to care (primary, specialty, tertiary) or quality of care. But what each of these shares, with the possible exception of genetics, is that they are unequally distributed at birth. Due to deep structural inequities associated with racism, poverty, and geography, children are born into contextual environments that powerfully and profoundly impact their eventual health outcomes (18-20). The presence of these structural inequities is thus a crucial part of the creation of health inequities in this country and an issue with which we must reckon as a society if we are ultimately to address them.

Health policy provides the tools to begin to address these structural determinants of health, across three main areas: access to care, quality of care, and broader socioeconomic context. The remainder of this manuscript will focus on policy efforts in these three areas and examine how they impact health outcomes and health inequities.

Health Policy Related to Access to Care

The most powerful determinant of access to care in the United States is insurance coverage. The United States is alone among industrialized nations in not offering its citizens universal health coverage, and, as a result, major gaps remain that impact individuals’ ability to pay for the health care they need. Prior to passage of the Affordable Care Act (ACA) in 2010, the uninsurance rate in the United States was roughly 15%–20%. The ACA had four major features aimed at increasing insurance coverage: the “individual mandate,” a requirement that individuals purchase health insurance or face federal fines; a series of reforms aimed at insurance companies, such as requiring that companies cover individuals without discrimination based on preexisting conditions and eliminating lifetime caps on benefits; the expansion of the Medicaid program to all individuals earning less than 138% of the federal poverty level; and the creation of insurance exchanges, government-run marketplaces where individuals could purchase health insurance, often subsidized, on the individual market.

Medicaid expansion was initially intended to be implemented by all 50 states, but following a Supreme Court case in which the constitutionality of expansion was challenged, it was declared optional. Consequently, while 25 states (including DC) expanded Medicaid in 2014, the remainder either expanded later (an additional 16 states have expanded since then) or declined to expand, leaving 10 states without Medicaid expansion as of January 2024. Coverage is thus different by state. While all states use Medicaid dollars to cover pregnant women and most children up to 200%–300% of the federal poverty limit and people with disabilities, eligibility for people who do not fit into these categories ranges broadly. Adults with dependent children face income eligibility thresholds that vary, from less than 20% of the federal poverty limit in Alabama and Texas to 100% of the federal poverty limit in Tennessee and Wisconsin. In non-expansion states (with the exception of Wisconsin), adults without dependent children have no way to get Medicaid, regardless of their income.

Despite the uneven implementation of Medicaid, as a result of these reforms, the U.S. uninsurance rate has dropped to 8%–10%. In 2021, 19% percent of the U.S. population was covered by Medicaid, and an additional 10% of the population purchased private insurance on the individual market, meaning directly from an insurance company; both of these increased significantly during the 2010s due to the ACA. The largest proportion of individuals in the United States (54%) had private insurance through their employer in 2021, while 18% were covered by Medicare and 3.5% by other sources such as the Department of Defense or Veterans Administration (note that since a small proportion of people have more than one source of coverage, numbers add up to slightly more than 100%) (21).

The reason these coverage numbers matter is that insurance coverage is a powerful predictor of access to care and ultimately of achieving optimal health. A growing body of evidence suggests that insurance coverage, and specifically Medicaid expansion, which largely impacts people who otherwise could not buy insurance on their own, has positive effects on financial security as well as on broader health and well-being (22). Medicaid expansion has led to better access to primary, preventative, and specialist care; higher rates of cancer screening; and higher rates of treatment for chronic conditions such as diabetes, hypertension, and dyslipidemia (23-26). Medicaid expansion is also associated with better access to behavioral health services, which is particularly important given high rates of mental illness among people chronically living in poverty (27), and to prenatal and perinatal care (28). Overall, Medicaid expansion has had a positive impact on health outcomes (22). Expansion is associated with fewer preventable hospitalizations (29) and reductions in cardiovascular and cancer mortality as well as reductions in disparities in infant mortality (26, 28, 30, 31).

Of course, access to health care is about more than health insurance, and there are major remaining disparities that speak to the need for focused policy efforts to address them. For example, the growth in high-deductible health plans and other forms of cost-sharing has contributed to increases in the degree to which people forego needed health care due to costs (32, 33). Geographic access issues also persist, with rural areas having less physical access to specialist care, as well as high-tech medical care provided in the hospital setting, such as targeted temperature management for cardiac arrest and mechanical thrombectomy for acute ischemic stroke (8, 34, 35). Policies to address affordability and to expand access to specialized services through the expansion of telehealth are sorely needed to close remaining access gaps, even as insurance expansion remains a policy priority.

Health Policy Related to Quality of Care

There are also policy efforts aimed at improving the quality of care delivered; these policies impact equity both directly and indirectly. A large consistent body of research has demonstrated that the quality of care delivered in the United States is uneven (36), with many missed opportunities for optimal treatment, and that suboptimal care is more often delivered to people from racial or ethnic minority backgrounds who live in poverty, lack insurance, face other socioeconomic challenges, or live in rural areas (37-42). As discussed earlier, some of these inequities are related to access issues, but a great deal are also related to the infrastructure available to deliver care in historically disinvested areas and at struggling hospitals. For example, safety net hospitals have fewer resources, may find it harder to attract and retain staff, have tougher working conditions, receive less ancillary support, and care for sicker, more complex patients. Hospital segregation, and care segregation more broadly, are the result of structural and interpersonal racism and their attendant policies and have had profound implications for health outcomes (18-20). Critical access rural hospitals often lack any specialist staff and may operate with extremely limited personnel at times. The economic decline of many rural communities has led to what are referred to as “care deserts,” or large geographic areas with no medical services, further worsening disparities in rural versus urban areas.

The federal government’s approach to quality improvement in the last decade has largely been via value-based purchasing (VBP), the idea that payment models should pay more for high quality and better outcomes rather than simply paying for the volume of care delivered. To that end, the ACA created a number of mandatory payment programs, including the Hospital Readmissions Reduction Program (HRRP, measures hospital readmissions), the Hospital Value-Based Purchasing Program (HVBP, measures mortality, safety, patient experience, and costs), the Hospital-Acquired Condition Reduction Program (HACRP, measures adverse inpatient events like infections and blood clots), and similar programs in the skilled nursing, home health, dialysis, and general outpatient settings. Collectively, these programs have transformed the payment for care, with very few remaining services unlinked to measures of quality or outcomes.

Unfortunately, these programs have not proven to be very effective in improving outcomes. While the HRRP reduced readmissions when it was first announced (43), gains have stagnated in the 13 years since, and overall hospital revisits have, if anything, increased (44). The HVBP was not associated with reductions in mortality (45), and the HACRP was not associated with improvements in safety (46). Further, in part because the risk models used to account for differences in patient populations do not take social risk into account, all of these programs have disproportionately penalized hospitals that serve poor, minority populations, potentially worsening rather than improving health equity (47).

Alternative payment models (APMs), which are programs that take an even more disruptive approach to changing health care payment and hold entities accountable for patient outcomes for a 90-day or one-year time period, have also been initiated over the last decade by Medicare as well as Medicaid and private payers. Here, the evidence is somewhat more mixed; accountable care organizations (ACOs) and bundled payment programs have failed to have much effect on patient outcomes (48-55), but they have improved emphasis on prevention and population health and are broadly considered to be an important mechanism by which to advance further goals related to quality and outcomes.

There is clearly a need in the quality domain to rethink how policy efforts can meaningfully improve quality, while also explicitly addressing and improving health equity. The focus on inpatient care has taken attention away from sorely needed investments in prevention and population health in the outpatient setting. The lack of inclusion of equity as a core value and goal of these programs has led to profound unintended consequences that should be unacceptable going forward. Quality policy that centered on equity would prioritize and reward access, population health, and risk reduction among the highest-risk groups. It would provide resources to historically marginalized communities instead of taking them away. And it would leverage clinician and patient engagement instead of thwarting it. Truly, the lack of inclusion of clinicians and patients as co-creators in these programs has led to a broad paucity of buy-in from these two crucial groups. Very few clinicians and even fewer patients would line up for a program that was aimed at saving money as many of the programs, particularly alternative payment models, were presented. Focusing programs around patient outcomes would have been a more popular approach.

Health Policy Related to Social and Economic Development

Most broadly, the factors that influence health are found in the social and economic development space, rather than in health care per se. Economic opportunity, with its roots in education, criminal justice, and job creation, is a powerful determinant of health that cannot be addressed by the health system alone. Housing, an incredibly powerful determinant of health outcomes, cannot be addressed by the health system alone. Similarly, food insecurity, adverse environmental exposures, and transportation cannot be addressed by the health system alone. In some sense, all social policy is health policy, in that all social and contextual circumstances help to create health. As such, the health policy community would be well-served to collaborate and cooperate in cross-sectoral efforts to improve social welfare more broadly. While specific policy recommendations in this space are beyond the scope of this paper, the recognition that health is created by a broad range of policies is a crucial step in research and advocacy aimed at finding ways to maximize health more broadly.

CONCLUSIONS

The poor health outcomes and striking health inequities we see in the United States today are not accidents. Rather, they are the result of decades, or even centuries, of a lack of investment in collective public and population health and well-being, policies rooted in racism, classism, and an underappreciation of health as a public good. As we aim to address these inequities and improve health outcomes more broadly, greater attention to the effects of policy on health outcomes is crucial. Research and advocacy efforts aimed at improving health through more effective and equitable policy should be priorities.

DISCUSSION

Ross, St. Louis: Karen, wonderful presentation. I must go back and complete that letter of recommendation for you to be named secretary of Health and Human Services so we can get on with this policy initiative debate. We’ve talked a lot about policy, and I am concerned that we overemphasize health policy without acknowledging the different levels of payment reform, which is important but ultimately may be insufficient, to really ameliorate the disparities that you talked about and that we just heard about regarding strokes. I wonder if you know about the health-in-all policy initiative that was started almost a decade ago, which argues that we need to have a greater focus on transportation policies, housing policies, and food policies. The Centers for Medicare & Medicaid Services (CMS) has placed a greater emphasis and greater funding on such initiatives as part of payment reform. What have you been exposed to, and what do you think about the health-in-all policies initiative?

K. Maddox, St. Louis: Thank you. That’s a terrific comment. Part of the trick right now is that many of the government agencies and groups that would be charged to do this are underfunded and underpaid. If we look to the housing industry or the Department of Housing and Urban Development (HUD), the transportation industry, or other government agencies that you would expect to play this role, they really don’t have the infrastructure or the funding. Because of the crazy way that we pay for health care in the United States, we’ve got all this private money in health care, which is very different from education or housing. It’s sort of an interesting set of questions and I think represents a real tension. Should hospitals be providing housing? Or should HUD be providing housing? But if you look, the government isn’t doing it well enough, right? So, hospitals have been stepping in, in places where they see there’s a connection to health. I think there’s both a need for us to address these needs as part of what we do, but also to very intentionally try to bring those other people to the table. There’s a lot of political power in health care, particularly in the insurance industry frankly, but also the hospital industry. I think there’s an opportunity for us to very intentionally engage with those groups. Transportation and housing agencies would probably like to be at the table but are not necessarily funded well. Some interesting experiments are going on in North Carolina Medicaid for example, and other states, where they’re letting Medicaid pay for social services. As long as that goes through and helps to fund and sustain community-based organizations, I think that’s probably the right kind of investment. I worry a little bit about running everything through hospitals; if you want to make something more expensive, have the hospital sell it to you, like your Tylenol, right? There is something to be said for making sure that we keep some of that money actually out of hospitals’ pockets, even if it flows through hospitals, and push for its investment directly into those spaces and communities. So I 100% agree with you. Thank you for the question.

Blumenthal, Baltimore: Karen, that was a great presentation. What are your thoughts about the impact of the Affordable Care Act, and do you see any reasonable chances that there could be bipartisan work to further improve health care and address some of the great questions that you posed?

K. Maddox, St. Louis: Maybe. I often think about the Affordable Care Act in terms of its effect on access, and then separately, its effect on care delivery or quality. I think it’s been a fairly unmitigated success on access. We’re at the highest levels of insurance coverage ever. Obviously, we’re falling back off now due to Medicaid disenrollment, which is another space where there’s a great need for policy intervention. I think in general we are moving toward better coverage. I think there are bipartisan efforts to figure out how to make that better. There will always be the small group that wants everyone to go back to just paying out of pocket and let people die if they can’t afford care. That small group is never going to go away, but there’s a real centrist, bipartisan group that wants to keep moving in the direction of broader, fairer coverage. The other part of the ACA, aimed at improving quality through payment policies, hasn’t been very successful. Under the current administration, there’s been a lot of movement toward trying to rethink that as well—a lot of movement into accountable care organizations, into population health, into bringing social needs into health care, which has been very bipartisan historically. Some centrist right leaders have helped along our progress in health care. Romney Care was the predecessor to Obama Care so there really is a history of bipartisan work in that space. Unfortunately, COVID stopped a lot of the work that was happening, and some very unfortunate wars have also stopped progress. There has also been a lot of political in-fighting; so there’s been progress, but it’s largely been at the agency level, not at the political level. When things calm down, I do think there’s bipartisan potential.

Stapleton, Iowa: I enjoyed your talk very much. I’m an HIV doc and a molecular biologist. I don’t think in financing terms as directly as you discussed. However, I’m struck by how federal funding for support of HIV that my clinic receives allows patients to obtain the kind of care that everyone deserves and very few people get. Have people compared HIV clinics and outcomes and been able to show different outcomes? In our case, we are able to support patients in emergency housing through Housing Opportunities for People with AIDS (HOPWA) and provide medical and social support care through the Health Resources & Service Administration Ryan White Care Legislation (HSRA) and the AIDS Drug Assistance Program (ADAP). These funding streams allow us to provide so much compared to other diseases, since there are very few disease-specific funding programs. I wonder if people have compared HIV clinics that have these resources with those that don’t, and if so, can that be utilized to encourage people to think more broadly?

K. Maddox, St. Louis: That’s a great question. I’ve not seen that research. For people who may not know, Ryan White is the program that pays for HIV care, and it provides quite generous coverage to people with HIV. It was named after a young White boy with hemophilia who contracted HIV from a transfusion. The Ryan White program is in place for the same reason that end-stage renal disease patients, or all patients who are on dialysis, have Medicare and ALS patients have Medicare. People with these diseases would have been uninsurable through a private market. These programs exist because a person who someone in Congress either was related to or knew well had these diseases, and they decided to create a program to make sure people with this disease could get health care. I’m sure they were thinking how could we let this person who we know with this terrible illness not have health care? We have these strange little pockets where we think that people somehow deserve health care due to some fairly interesting personal histories and decisions. Every person deserves health care, regardless of whether they fit in one of those buckets. But right now, for most diseases, it is a patchwork. It is a great research idea actually, so I’ll have to look into that.

Rounds, Providence: As a physician working in the Department of Veterans Affairs (VA), I have to remind everyone that the VA has done a spectacular job providing the type of uniform and standardized health care and access that you’re describing in the last 25 or 30 years. However, there’s an unfortunate new development called Community Care. I fear that this may turn the VA into an insurance agency. Are there any data yet on the outcomes of veterans being outsourced to community institutions as compared to continuing to have their care within the VA system? I’m not referring to highly specialized care, which, of course, should only be done in referral centers.

K. Maddox, St. Louis: Thank you. That’s a great point, and I was remiss in not mentioning the VA. The VA thinks very differently about access and about people deserving access than much of the rest of the U.S. health care system. Community Care is a little difficult to study because it is a mix, as you point out, of people seeking specialized care that they could not receive within the VA and people who for whatever reason are motivated to find other care outside the VA. My understanding is that so far, the outside-the-VA care is a little bit more patchy or more variable. In general, the providers of Community Care are not higher quality than VA providers so it looks like VA patients are not accessing a particularly high-quality group of providers. The disruption in continuity between the two groups also may create some higher costs and more holes. It’s a tricky thing to study because of the data issues that occur when people are going between systems—it’s almost like sorting out layers of Swiss cheese. We don’t know the net effect on different groups of patients yet.

Rounds, Providence: Thank you.