2

Connection Between Health Equity and History, Federal Policy, and Data

INTRODUCTION

This chapter begins by outlining U.S. governmental structures and power and how policies play out today, followed by a description of the historical and contemporary factors that have contributed to health inequities and the state of these health inequities. The chapter also discusses important factors that intersect with race and ethnicity, including sex, gender, sexuality, age, disability status, geographical considerations, and citizenship, that have implications for health equity. Finally, the committee reviews the state of data sources for racial, ethnic, and tribal populations and outlines critical gaps and data needs.

GOVERNMENTAL STRUCTURES AND POWER

This section provides a brief overview of federalism and available policy levers and the unique governing aspects for American Indian and Alaska Native (AIAN) people. Policy has four main levers. (1) Laws/statutes are passed by Congress and signed by the president. (2) Rules/regulations—when Congress enacts legislation, it frequently delegates rulemaking authority to federal agencies, which establish specific requirements (this generally requires a public notice and comment period). (3) Guidance is supplemental materials that provide clarity on laws or rules. Guidance with broad applicability may be published in the Federal Register, but some may only appear on agency websites, and it generally does not require notice and public comment. The Administrative Procedure Act (APA) guides agency rulemaking; agencies have

to establish an administrative record that lays out the legal, scientific, technical, economic, and policy basis for the rule. Rules go through an extensive “clearance” process within each agency and at the White House before they are released. According to APA and Supreme Court precedent, agencies must provide a reasoned explanation when they want to change established policy. (4) Finally, on the implementation side, enforcement is through a combination of public and private efforts, such as the enforcement of civil rights statutes.

One additional lever is the Executive Order (EO), which is issued by the president, acting in their capacity as head of the executive branch, directing a federal official or administrative agency to engage in or refrain from a course of action. It is an often-used approach for swift policy implementation. It is appealing in that it lays out priorities for a given administration, takes immediate effect, and does not require input or review from Congress or other constituents. It is an effective tool to demonstrate action on promises made to constituents, create momentum, and bring focus to an effort across one or many agencies. Staff are compelled to take action, so attention and resources are quickly redirected to carrying out these orders. They are also precarious because they become viewed as the darling of a given administration and can fall prey to shifts in political leadership. As a tool, they are a less enduring form of policy implementation and risk a subsequent administration easily drafting a new EO ending or undoing the authority of the original. The limitations need to be carefully considered when trying to address long-held systemic barriers to equity. A review of active EOs may reveal progress toward equity goals that would benefit from actions to stabilize and make permanent the changes beyond a given administration. EOs can advance and contribute to health equity. For example, EO 13985¹ Advancing Racial Equity and Support for Underserved Communities Through the Federal Government is focused on advancing equity. However, EO 13950² Combatting Race and Sex Stereotyping is an important example of how federal policy could negatively impact efforts aimed at improving health equity (it was revoked in January 2021), as it prohibited federal contractors and subcontractors from providing certain workplace diversity training and programs.

Federalism

The committee has been charged with reviewing the effects of federal policies on health equity, but federal policies cannot be viewed in a vacuum because they interact with or act on other levels of government—state, tribal, territorial, and local. In addition, power is shared between states and the federal government, with states generally taking the lead on most issues.

___________________

¹ Exec. Order No. 13985, 86 FR 7009 (January 2021).

² Exec. Order No. 13950, 85 FR 60683 (September 2020).

Federalism “refers to the division and sharing of power between the national and state governments” and is a fundamental principle upon which the U.S. government was established (Congress, n.d.). The Constitution divides governmental power between the federal and state governments, limiting federal authority to issues of national interest. Through the 10th Amendment to the Constitution, “[t]he federal government is delegated certain enumerated powers while all other powers not otherwise prohibited by the Constitution are reserved to the states” (Cornell Law School, n.d.-b; NGA, 2018). The enumerated powers include the power to tax and spend, regulate interstate commerce, standardize immigration/naturalization, establish and maintain a military, and declare war. All other powers, broadly defined as “police powers,” which include the authority to protect public health and safety, are left to the states. It is through the principle of federalism that the concept of states as laboratories of democracy was established. In New State Ice Co v. Liebmann (1932),³ Supreme Court Justice Louis Brandeis described how a single “courageous State may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country.”

Theoretically, federalism establishes a unified national government while preserving localized governing and decision making. However, the interplay between the federal and state, and state and local, authority is often contentious and used as a tactic to voice opposition. States cite federalism when pushing against federal actions aimed at constraining state autonomy. Federalism provides both protections and challenges on the path to furthering health equity; it created and, in many ways, continues to perpetuate racial inequities. For example, it informed Supreme Court attitudes toward slavery as a state right. In Dred Scott v. Sanford (1857),⁴ which involved moving an enslaved person from a state that allowed slavery to one that outlawed it, the court’s views of federalism led it to find against the enslaved person (Maltz, 1992).

In the 21st century, federalism played a key role in the passage and evolution of the Patient Protection and Affordable Care Act (ACA). It expanded health care coverage, with the federal government providing financing and establishing a floor for regulating the insurance markets; states enacted regulations that shaped access and other aspects of implementation (Collins and Lambrew, 2019). A 2019 Commonwealth Fund analysis found that ACA’s combination of federal standards and subsidies and state regulatory authority significantly improved coverage and access nationally and narrowed regional differences. “However, the law’s federalist structure, established in statute and altered through regulations and court decisions,

___________________

³ 285 U.S. 262 (1932).

⁴ 60 U.S. 393 (1857).

resulted in disparities in coverage and access across states” (Collins and Lambrew, 2019). One important change that led to this outcome is the 2012 Supreme Court’s decision in NFIB v. Sebelius,⁵ which made optional the requirement that states extend Medicaid to all adults with incomes below 138 percent of the federal poverty level. The court found that this individual mandate⁶ was not valid under the Commerce Clause because Congress cannot use that power to require someone to purchase health insurance. However, the court upheld the mandate as a valid use of congressional taxing power (treating the penalty for not purchasing insurance as a tax) and upheld Medicaid expansion by judicially prohibiting the Secretary from withdrawing existing Medicaid funds from states that refuse compliance. “More broadly, policy decisions about the allocation of state versus federal governing responsibility in health care have implications for the relative performance of states as well as the overall health of the U.S. population” (Collins and Lambrew, 2019).

States have what is generally called “police powers” to protect the public’s health and sometimes delegate it to local governments. Federal authority to govern health resides primarily in the ability to tax and spend and regulate interstate and foreign commerce, through congressional power via the Commerce Clause of the Constitution; it and the Public Health Service Act give federal health agencies broad authority.

Government responsibilities for public health reside largely at the state level, with the federal government providing resources, generally paired with program requirements, and technical assistance to states and to local public health entities. In other domains, such as education and transportation, the relationship between the federal department and agency and the state-level equivalents reflects a similar division of authority.

Federal-level policies aimed at furthering equity, such as the Civil Rights Act, and especially Title VI, place specific requirements on state governments as a condition of receiving federal support. Examples include Department of Education funding to the states (DOE, 2020) and Federal Highway Administration funding to the state departments of transportation (FHWA, n.d.). These requirements may help further the goal of health equity because they are intended to prevent discrimination, such as by prioritizing environmental justice in transportation planning (AASHTO, n.d.). Federal civil rights law provides a set of minimum standards on which states can build by enacting further protections (Pepin and Weber, 2019). One example is

___________________

⁵ National Federation of Independent Business v. Sebelius, 567 U.S. 519 (2012).

⁶ The individual mandate of ACA imposed a “penalty” or “tax” on certain individuals who failed to obtain health insurance through their employer, the government, or a private company and expanded Medicaid to specified individuals below the poverty line. Federal funding to states’ Medicaid programs was conditioned on acceptance of these terms.

state application of standards to protect individuals with a criminal record from discrimination that may make it difficult, if not impossible, to find employment or housing. Beyond civil rights law, federal agencies may use a range of strategies to encourage state innovation and experimentation to identify the most effective solutions. This includes the Centers for Medicare & Medicaid Services (CMS) State Innovation Models and Medicaid waiver programs (Kissam et al., 2019; Vleet and Paradise, 2014) (see Chapter 5). On a negative note, in the context of the COVID-19 pandemic, federal government action under the Commerce Clause has been complicated by a range of factors, including judicial challenges. Federal action in the domains of housing, transportation, and public health has been heavily contested by both the federal judiciary and states. For example, some states banned sharing infectious disease data with the federal government (Rai, 2021) or criticized the CDC-imposed moratorium on evictions (Stennett, 2021). Federalism may also have hampered an effective COVID-19 response because of minimal coordination among states to agree on standard public health approaches to address the threat they all faced (Kettl, 2021). A federal judge vacated the federal mask requirement on public transportation in April 2022 based on a narrow interpretation of U.S. Code “Regulations to Control Communicable Diseases”⁷ (Jost, 2022).

Role of the Judiciary in Determining the Responsibility of the Federal Government

Given the separation of powers among the executive, legislative, and judicial branches, these dynamics also unfold in a complex judiciary context. For example, the Supreme Court decision to jettison decades-old precedent on abortion in Dobbs v. Jackson Women’s Health Organization (2022)⁸ was made based on arguments that the decision should be returned to the domain of the states. A similar decision can be found in voting rights, with the Supreme Court’s ending of preclearance requirements for specific states under the Voting Rights Act in Shelby County v. Holder (2013).⁹ For ACA, the judicial interpretation of federalism found the Commerce Clause was insufficient to support federal ability to compel individuals to carry insurance, but federal tax and spend power prevailed, as the tax penalty in the individual mandate was upheld (Aaron, 2012). ACA’s Medicaid expansion was also challenged in court, and the Supreme Court found that although the federal government could withhold new funds for that expansion, it could not take away existing Medicaid funds (Cornell Law School, n.d.-a).

___________________

⁷ 42 U.S.C. 264.

⁸ Dobbs v. Jackson Women’s Health Organization, 597 U.S. ___ (2022).

⁹ Shelby County v. Holder, 570 U.S. 529 (2013).

The role of the federal government could be viewed in a top-down frame, as a central government imposing its will, or in a collaborative frame, which could have created a pandemic response that brought states and their federal partners together to confront a shared threat.

American Indian and Alaska Native People: Federal Trust Responsibility

The relationship between American Indians and the federal government is termed the federal trust responsibility: the government is responsible for supporting those and their descendants who ceded land to the United States by force, coercion, or mutual treaty-making. Treaties between tribes and the United States confirm each nation’s rights and privileges. This includes protection for AIAN from attacks upon their lands; health care; education; sovereignty and religious freedom; confirmation and protection of certain rights, including self-government and jurisdiction over their own lands. Tribes ceded title to vast amounts of land in exchange for these protections and services but reserved certain lands and rights. However, the slate of federal policy periods that followed this doctrine undermined it, leading to the inequities AIAN people face today.

Tribes are neither states nor foreign countries. When AIAN people are enrolled members of the 574 federally recognized tribes, which are sovereign nations, they have a formal nation-to-nation relationship with the U.S. government (Schwartz, 2023). “Sovereignty” is a legal term meaning the authority to self-govern. “Hundreds of treaties, along with the Supreme Court, president, and Congress, have repeatedly affirmed that Tribal Nations retain their inherent powers of self-government and created a fundamental contract between tribes and the United States” (NCAI, n.d.).

Tribes are the only group to be dual citizens originating from within what is now the United States, which sets them apart from other groups who are minoritized. They share similar adversities with the additional, unique challenges of having been described historically as “Wards of the Nation,” been tied currently or ancestrally to reservation bases, and suffered under federal policy periods aimed at solving the “Indian problem” (Moss, 2019). For example, they are the only population listed in U.S. Code, as Title 25-INDIANS (25 U.S.C).¹⁰ However, obstacles to self-governance are based on three areas that have been identified in recent years: (1) outmoded bureaucratic processes, (2) lack of federal agency coordination, and (3) regulations and laws that prevent tribal governments from equitable access to federal programs on par with state and local governments (NCAI, n.d.).

___________________

¹⁰ The Code of Federal Regulations Title 25 contains the codified federal laws and regulations that are in effect as of the date of the publication pertaining to American Indians (Native Americans), including gaming/casinos, arts and crafts, education, and health (U.S. Code, 2011 Edition).

U.S. federal Indian policies/eras that were implemented from the late 1700s through 1978 to solve the “Indian problem” and allow for Euro-American expansion included exterminating, removing, assimilating, and relocating AIAN people (Moss, 2019). These policy periods and their offshoot policies can be mapped directly onto the definition of genocide from the United Nations in Article 2 of the Convention on the Prevention and Punishment of the Crime of Genocide (“. . . a crime committed with the intent to destroy a national, ethnic, racial or religious group, in whole or in part”) (UN, 1948). There are five parts:

1. Killing members of the group;
2. Causing serious bodily or mental harm to members of the group;
3. Deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part;
4. Imposing measures intended to prevent births within the group; and
5. Forcibly transferring children of the group to another group (UN, 1948).

By this definition, the United States perpetrated genocide within its borders specifically on this population. This is important to understand as policy makers contemplate the contribution of federal policies to health inequities.

“Invasion is a structure not an event” (Wolfe, 1999, p. 2). This quote, well known in Indian Country, indicates that the structures left behind by this history make it extremely hard to move toward health equity, as AIAN people are often blamed for their circumstances, including changing their health and wellness circumstances and abilities. The three most problematic policy periods creating and continuing colonial structures were specifically meant to either “get rid of the Indian problem” or “kill the Indian, save the man/or child.” They were (1) Removal and Reservations (1830–1886), (2) Assimilation (1887–1932), and (3) Termination (1946–1960).

Termination is when the federal government recognized tribes’ sovereignty, trusteeship over reservations, and the exclusion of state law’s applicability. This era saw a reversal in policies from the self-government era, when the federal government resolved to end the special trustee relationship (House Resolution No. 108, 83rd Congress. August 1, 1953). In addition to eliminating the right to be sovereign nations, the policy terminated federal support of most health care and education programs. On its face, the resolution seemed well intended—to liberate tribes from federal control. However, it became another means of controlling and erasing AIAN rights. The government-to-government relationship was unilaterally severed by Congress—over 100 tribes were cut from the list of those with federal recognition. Those now “unrecognized” tribes were no longer considered Indian in the eyes of Congress and the federal government.

Numerous examples illustrate how federal policies have impacted the health and well-being of the AIAN population. For example, President Washington ordered the outright killing of AIAN people, and President Lincoln ordered the largest mass execution by the U.S. government on U.S. soil on Dakhóta Nations in Minnesota 6 days before the Emancipation Proclamation in 1862 (University of Minnesota, n.d.). The United States withheld citizenship for AIAN people until 1924 (the last racial group to gain citizenship) and, until the 1950s, forced exile to reservations and outlawed AIAN religious and cultural practices. The Indian Health Service (IHS) undertook coerced or nonconsensual sterilization of AIAN women into the 1970s (Torpy, 2000). For decades, many Indian children were removed to boarding schools or adoption and/or foster to White families. Historical trauma (trauma persisting from previous generations) coupled with some of the highest contemporaneous traumas for AIAN people have resulted in substantial health inequities (see Chapter 7). The traumas that have unfolded over generations, and done so “invisibly,” resulted in untold cumulative harm, the effects of which are still felt today.

However, “the sovereignty that the Indian tribes retain is of a unique and limited character. It exists only at the sufferance of Congress and is subject to complete defeasance.”¹¹ Tribal Nations would argue that they retain inherent sovereignty for being on this land for 1,000s of years before contact. Another way this has been described legally is that a tribe is only a tribe if Congress confirms it. By extension, an Indian is only an Indian if Congress confirms it. Congress designates tribes, and tribes designate membership. These tribes are then called “federally recognized.”

Hawaii, Pacific Islands, and Territories

People from Hawaii, the Pacific Islands, and U.S. territories have varied histories, cultures, and membership in or ties to the United States. These relationships are briefly described here (see Chapter 7 for additional historical information).

Until January 17, 1893, Native Hawaiians, Kānaka Maoli, were a self-governing people; between 1893 and 1898, a provisional government comprised mainly of Americans was in place. Aboriginal people have lived on Hawaii for more than 1,000 years, but the U.S. overthrow of the Kingdom of Hawaii in 1893 and its annexation in 1898 without the consent of the people means no federally recognized native governing body exists (Kana‘iaupuni and Malone, 2006). Native Hawaiians do not have a government-to-government relationship (which is based on treaties) comparable to AIAN people (Davis, 2021), and residents of Hawaii are subject to Hawaiian state and U.S. federal laws. Hawaii has sent congressional delegations to the Senate and House of Representatives since it became a state in 1959. Although Native Hawaiians

___________________

¹¹ United States v. Wheeler, 435 U.S. 313, 98 S. Ct. 1079 (1978).

never relinquished their sovereignty (see Chapter 7 for additional details) and the United States provided a formal Apology Resolution (U.S. Public Law 103–150), Native Hawaiians are not seen as sovereign under U.S. federal law and do not have self-governance rights. Some have engaged in a movement to have their status as Native people revived (though not all Native Hawaiians agree this is the best path—some argue for complete independence from the United States, federal recognition and Indigenous status, or more control over Native Hawaiian assets, such as crown and ceded lands) (Greene, 2021; Nation of Hawai‘i, n.d.; Pacheco, 2005). “However, the Department of the Interior has a special political and trust relationship with the Native Hawaiian Community that exists even without a formal government-to-government relationship (43 CFR 50)” (DOI, 2022). Other federal statutes, regulations, and reports¹² outline certain responsibilities of the federal government to Native Hawaiians, such as the Hawaiian Homes Commission Act (see Chapter 6) and its land trust obligation and the Native Hawaiian Health Care Improvement Act¹³ (see Chapter 5).

Puerto Rico and the U.S. Virgin Islands are unincorporated territories (meaning that only select parts of the Constitution apply to their residents); Puerto Rico is also commonwealth (it has a political union with the United States). Individuals born in both places are considered U.S. citizens, but residents cannot vote in federal elections; the federal government also does not collect income tax from residents (except for federal employees). Puerto Rico residents are subject to most federal laws.

Guam is also an unincorporated territory. Individuals born in Guam are considered U.S. citizens. Residents elect a delegate to the House of Representatives, who serves for a term of 2 years and has limited voting abilities, but cannot vote in federal elections. Residents of Guam also caucus for presidential primary candidates.

American Samoa is an unincorporated, unorganized territory. Congress has not established a system of government for it. Individuals born in American Samoa are U.S. nationals rather than citizens (they may apply for citizenship). Residents cannot vote in federal elections.

The Northern Mariana Islands are a commonwealth. The U.S. president is head of state of the Northern Mariana Islands, which receives funds from the U.S. government. Individuals born in the Northern Mariana Islands are considered U.S. citizens, but residents cannot vote in federal elections.

The Compacts of Free Association govern the relationships between the United States and the Republic of the Marshall Islands (RMI), Federated States of Micronesia (FSM), and Republic of Palau. The compacts grant the United States the choice to operate military bases in Freely Associated States (FAS—the RMI, FSM, and Republic of Palau) and make decisions

___________________

¹² For examples, see https://www.doi.gov/hawaiian/lawreport (accessed May 31, 2023).

¹³ 42 U.S.C. 11701-11714.

related to their security. Citizens of the FAS have the right to reside and work in the United States and its territories as lawful nonimmigrants or “habitual residents,” meaning they can freely travel, live, and work in the United States without a visa and with no time restraints. They do not have their own militaries and are eligible to join the U.S. military (CRS, 2022).

HISTORICAL AND CURRENT STRUCTURAL RACISM: FUNDAMENTAL CONTRIBUTORS TO HEALTH INEQUITY

As discussed in Chapter 1, federal policies can prevent, address, and contribute to health equity. Not all policies are designed to address health inequities—a policy or program could be simply neutral. In addition, most policies do not intend to increase inequities; it is an unanticipated outcome. However, historical policies have shaped the current state of health inequities.

The United States has a deep history that has contributed significantly to the health inequities seen today. Settler colonialism, when European settlers infected Native Americans (and made infection rates worse due to lack of quarantine) with new epidemic diseases (Patterson and Runge, 2002), was followed by violence from warfare, forced displacement, and enslavement to control Native populations (Fisher, 2017). Hundreds of thousands of people were captured from Africa, forced into slavery in U.S. and other colonies, and exploited to work in the production of crops, textiles, and a host of other industries (Library of Congress, n.d.). During Reconstruction after the Civil War, the Freedmen’s Bureau, which provided services to newly freed people, was dissolved (United States Senate, n.d.); the bureau was not able to carry out its initiatives and could not provide long-term protection for African American people (Maltz, 1992). Rather than address this, the government framed African American people as innately diseased due to presumed biological inferiority (Hogarth, 2019). Racism was built into the structure of society, not only influencing how future policy and practices are designed but also shaping culture, ideology, norms, and values in ways that serve to maintain racial order and hierarchies in which Black, AIAN, and other minoritized people are deemed inherently inferior (Downs, 2012).

Although the 13th Amendment abolished slavery and hundreds of treaties were signed to protect Native Americans, racial violence took on different forms through new policies and practices, affecting the health and well-being of all racially and ethnically minoritized populations (e.g., Jim Crow and Japanese internment camps) (Downs, 2012; U.S. Commission on Civil Rights, 2018). Even today, with federal recognition of past racial harm and efforts to create policies and laws to ameliorate and prevent harm, policies are still created that negatively affect the health and well-being of communities that have historically been disenfranchised and underserved, as discussed in Chapters 3–7. See Chapter 7 for more on how historical and current policies have led to both trauma and healing for racialized populations.

Structural Racism, Laws and Policies, and Racial and Ethnic Health Inequities

Racism is a key determinant of racial and ethnic health inequities and operates at all levels of society (Williams et al., 1997, 2019). Racism systematically excludes, Black, AIAN, Latino/a, and other targeted people from power, status, and other social, economic, and political resources and opportunities, which contributes to health inequities (Baah et al., 2019; Williams et al., 2019). Racism has long been studied in the social science, law, and the humanities, but in recent years, population health researchers have increasingly examined the role of not only interpersonal but also structural racism in shaping population health outcomes across and within racial and ethnic groups (Bailey et al., 2017; Gee and Ford, 2011; Williams et al., 2019). “Specifically, investigators have found that structural racism—defined as the historically contingent and persistent ways in which social systems and institutions generate and reinforce inequities in access to power, privilege, and other resources” (Agénor et al., 2021, p. 428) among racial and ethnic groups deemed to be superior and inferior (Bailey et al., 2017; Gee and Ford, 2011; Williams et al., 2019)—negatively affects health outcomes, including breast cancer, premature mortality, birth outcomes, and cardiovascular disease (CVD), among Black people (Bailey et al., 2017; Krieger et al., 2013, 2014; Krieger et al., 2017b; Lukachko et al., 2014; Williams et al., 2019). Additionally, scholars have shown that structural racism, including genocide and immigration policies, undermines the physical and mental health of AIAN and Latino/a people, respectively (Bailey et al., 2017; Gee and Ford, 2011; Viruell-Fuentes et al., 2012).

Agénor and colleagues (2021, p. 429) explain that

Policies that explicitly or implicitly favor or disadvantage certain racial and ethnic groups (e.g., historical and contemporary housing and banking laws, and practices such as redlining, mortgage lending, zoning) result in minoritized racial and ethnic groups being more likely than White people to live in systematically underserved neighborhoods (Agénor et al., 2021).

As a result, they live in neighborhoods that lack access to high-quality infrastructure, social and health services, and educational and employment opportunities (Agénor et al., 2021; Badger, 2013; Bailey et al., 2017; Gee and Ford, 2011; Pager and Shepherd, 2008; Rugh and Massey, 2010; Williams et al., 2019).

Criminal legal system laws, policies, and practices also disproportionately affect minoritized racial and ethnic groups (see Chapter 7). This leaves Black, AIAN, and Latino/a people (who tend to be the target of overpolicing, arrest, and incarceration negative racial stereotypes) disproportionately represented in prisons and jails (Alexander et al., 2012; Bailey et al., 2017; NAACP, 2014; Pager and Shepherd, 2008). Consequently, voting rights laws that disenfranchise people with convictions disproportionately affect Black and other minoritized groups, which has implications for the composition of the electorate that make laws and set policy agendas (Alexander et al., 2012). “Furthermore, given pervasive racist stereotypes that erroneously depict Black people as dangerous criminals, stand-your-ground laws systematically threaten their rights and lives by selectively allowing. . . . [someone] to harm or kill a Black person by claiming self-defense, even in the absence of an actual threat (Morris, 2016)” (Agénor et al., 2021, p. 429).

Minimum-wage, income-related housing, and predatory lending laws disproportionately affect minoritized groups (Agénor et al., 2021). This leads to Black, AIAN, and Latino/a people having lower levels of wealth and income than their White counterparts (Bailey et al., 2017; Bell, 2017; Delgado and Stefancic, 2017; Pager and Shepherd, 2008; Williams, 1999). This is the result of laws and policies that prevent the intergenerational transfer of wealth among Black and other underserved communities, as well as structural and interpersonal racism in the education system and employment sector. Because of biases—conscious or unconscious—that inaccurately depict Black people as threatening or violent, discipline that uses force is more likely to be directed at Black students than their White counterparts in both elementary and high schools (Agénor et al., 2021; Gregory, 1995; Morris, 2016; Riddle and Sinclair, 2019). Immigration laws also disproportionately target Black and Latino/a people and lead to the social exclusion of minoritized groups by “prohibiting entry, facilitating deportation, and limiting access to social, economic, and political opportunities and resources as well as legal rights” (Agénor et al., 2021, p. 429) for groups considered to be undesirable (Gee and Ford, 2011; Viruell-Fuentes et al., 2012).

Theory and empirical research suggest that these historical and contemporary federal and state laws disadvantage people from minoritized groups shape racial and ethnic health inequities through various social, economic, physical, and psychological mechanisms into low-quality neighborhoods, schools, and jobs and other circumstances, including chronic and acute psychosocial stressors (Bailey et al., 2017;

Gee and Ford, 2011; Komro et al., 2012; Williams, 1999; Williams and Collins, 2001; Williams et al., 1997, 2016, 2019). “For example, research shows that racial residential segregation, which is linked to housing and banking laws, policies, and practices that disproportionately disadvantage Black people and people from other marginalized racial and ethnic groups, is negatively associated with low-birthweight and preterm birth among Black women and higher rates of breast and lung cancer mortality among Black people compared with White people (Gee and Ford, 2011; Williams and Collins, 2001; Williams et al., 2019)” (Agénor et al., 2021). Living in a neighborhood with high levels of stop and frisk—which occurs more often in Black and Latino/a neighborhoods and affects AIAN people disproportionately (ACLU, 2020; NAACP, 2014; NYCLU, 2019), is correlated with several negative mental and physical health outcomes, including psychological distress, diabetes, and high blood pressure (Sewell and Jefferson, 2016).

These examples illustrate not only how structural determinants of health are embedded within institutions across society but also how different manifestations and dimensions of racism interact, reinforce, and even replace each other to maintain inequalities and make them seem natural and even quotidian rather than a consequence of historical and contemporary processes (Gee and Hicken, 2021).

STATE OF HEALTH INEQUITIES IN THE UNITED STATES

Despite data gaps that prevent accurate and full documentation of health and well-being for many racially and ethnically minoritized populations, data that are available suggest persistent and sometimes growing health inequities. The inequities discussed in this report reflect historical and contemporary structural disadvantage that have led to different social and economic opportunities (see Figure 2-1)—for a comprehensive, in-depth assessment of the root causes of health inequities, see Communities in Action: Pathways to Health Equity (NASEM, 2017). The report identified root causes that can be organized into two clusters (p. 7):

Furthermore, the report concluded that “health inequities are the result of more than individual choice or random occurrence. They are the result of the historical and ongoing interplay of inequitable structures, policies, and norms that shape lives” (NASEM, 2017, p. 8) and that interventions targeting the above factors hold the greatest promise for promoting health equity. Although family circumstances and personal choice play a role in health outcomes, the evidence suggests that improving the social and environmental circumstances where people, families, and communities live, work, play, age, and pray—so that all people have the opportunity to make the healthy choice—is essential. This includes supporting parents, families, and communities via federal policy.

Racial, ethnic, and tribal health inequities take on many forms, such as higher rates of chronic disease and premature death. Despite progress to narrow the gaps in some racial, ethnic, and tribal health outcomes, there are still ongoing and persistent inequities tied to the social and structural determinants of health. For example, in the AIAN population, non-Hispanic AIAN people are 1.8 times more likely to be killed by the police compared to non-Hispanic White people and diabetes prevalence remains high; in addition, the COVID pandemic put this population further behind compared to other racial and ethnic groups, with an estimated loss in life expectancy at birth of 4.5 years in 2020 and 6.4 years in 2021 relative to 2019 (ASPE, 2019; GBD 2019 Police Violence U.S. Subnational Collaborators, 2021; Goldman and Andrasfay, 2022). The following section provides an overview of a sample of these health inequities.

Racial and ethnic groups are also not homogenous. Take, for example, the experience of the Hispanic and Latino/a population, specifically when examining data in disaggregated form (Fernandez et al., 2023). In general, data

show that people from Puerto Rico have worse health outcomes than every other Hispanic and Latino/a group in the United States (Aguirre-Molina et al., 2001; Díaz et al., 2020). An analysis of childhood adversity and mental health measures from the National Heart, Lung, and Blood Institute Hispanic Community Health Study/Study of Latino/a people demonstrates this trend as well (Cooper et al., 2021). People from Puerto Rico have the highest age-adjusted mortality rate, at 605.7/100,000, followed by people from Mexico (523.7), Cuba (489.1), Central American (393.2), and South American (315.5) (Fernandez et al., 2023). Health outcomes for Hispanic people vary based on birthplace, place of ancestry, immigrant status, length of time in the United States, where they live, age, gender, and socioeconomic compositions (Fernandez et al., 2023). “Broad statements about Hispanics as a group often do not translate into better comparable health outcomes among all Hispanics” (Fernandez et al., 2023, p. 163).

Another example is Asian American people, who are the fastest-growing U.S. immigrant group and projected to become the largest by 2065 (Kim et al., 2021). However, research on this population is lacking and has not kept pace with the changing demographics (Ðoàn et al., 2019; Kim et al., 2021). Kim and colleagues (2021) explain that this may result in part from the myth of the model minority stereotype for this population. However, the Asian population is one of the most diverse groups in the United States, with widely varying backgrounds across more than 37 ethnocultural groups. The Asian population also has the largest within-group wealth gap; the lowest income levels show the least growth 1970–2016 compared to other racial and ethnic groups (Kim et al., 2021).

This heterogeneity applies to other population groups as well (e.g., the AIAN, Black, and Native Hawaiian and Pacific Islander [NHPI] populations) and impacts all measures of health. This report uses available data, which typically do not sufficiently break out the heterogenous places of origin within race and ethnicity. These masked differences have large health care implications, and disaggregating data would allow for more individualized or tailored approaches to addressing health equity (see Data Gaps and Opportunities later in this chapter).

Life Expectancy

From 2000 to 2019, overall U.S. life expectancy increased by 2.3 years, but this gain was not consistent across racial and ethnic groups or by geography. In 2019, the life expectancy was 85.7 years for the Asian, 82.2 for the Latino, 78.9 for the White, 75.3 for the Black, and 73.1 for the AIAN populations (NIH, 2022). AIAN people experienced the sharpest decline of any racial and ethnic group for 2019–2021 (Goldman and Andrasfay, 2022). Researchers have noted that estimates that do not disaggregate the Asian

population from the NHPI populations likely mask important differences in life expectancy. “Regional studies generally show worse outcomes for NHPI populations, further underscoring the need to study these groups individually” (NIH, 2022). A study measuring population- and race-specific age-standardized mortality rates found, on average, 74,402 excess deaths (all-cause mortality) for Black people compared to White people each year for 2016–2018 (Benjamins et al., 2021). Annual excess deaths varied considerably by location—at the city level, this ranged from 6 in El Paso to 3,804 in Chicago. These higher rates of death translate into years of life lost. A study by Amiri and colleagues (2022) examined years of potential life lost by race and ethnicity in Washington State and found that relative to non-Hispanic White people, non-Hispanic Black, AIAN, Asian or other Pacific Islander, multiracial, and Hispanic decedents had significantly higher rates. Disparities were reduced, but not eliminated, when controlling for sociodemographic factors, and controlling for place-based risk factors did not further lessen differences. A 2023 study by Caraballo and colleagues calculated excess age-adjusted all-cause mortality, cause-specific mortality, age-specific mortality, and years of potential life lost rates (per 100,000 individuals) for Black people in comparison to White people in a cross-sectional study using 1999–2020 Centers for Disease Control and Prevention (CDC) data; the Black population experienced more than 1.63 million excess deaths and more than 80 million excess years of life lost. After a period of improvement for the gap in the age-adjusted mortality rate, it plateaued and then worsened in 2020. The analysis showed similar trends in rates of excess years of potential life lost (Caraballo et al., 2023).

Maternal Health and Infant Mortality

Maternal Health

Gunja et al. (2023) found that the United States had the highest rate of maternal mortality of 13 high-income countries in its analysis, even though over 80 percent of these deaths are preventable (Trost et al., 2022). Severe maternal morbidity rates have also been increasing in recent years, with higher rates than other high-income countries (Ahn et al., 2020; Hoyert, 2022), with stark racial and ethnic inequities in maternal health (see Figure 2-2). Black, AIAN, and NHPI people have higher rates of inadequate prenatal care compared to White people (Hill et al., 2022; March of Dimes, 2022). Black and AIAN pregnancy-related mortality rates are over three and two times higher respectively compared to White rates, with racial disparities increasing by age (Hoyert, 2022; Petersen et al., 2019). Several factors may contribute to these inequities, including access to care, quality of care, prevalence of chronic diseases, structural racism, and implicit biases (CDC, 2023a) manifested in a complex and multifactorial manner (see Figure 2-1).

For example, lack of federally mandated paid family leave,¹⁴ lack of insurance coverage, geographic inequities in access to care, lack of a culturally congruent workforce, and a lack of trust in the patient–provider relationship and health care system are shown to contribute to these inequities (IOM, 2003; NASEM, 2019). The National Academies convened a 2021 workshop, “Advancing Maternal Health Equity and Reducing Maternal Morbidity and Mortality,” to address this urgent issue (NASEM, 2021a). Although recent federal attention to these inequities is promising (see Chapter 5), they both reflect and result from the complex, intersectional, and cross-sectoral issues discussed in this report.

Infant Mortality

Infant mortality is the death of an infant before their first birthday. Stark racial inequities have persisted for decades, even with continued advancements in medical care (Hill et al., 2022). Racially and ethnically minoritized people are more likely to experience certain birth risks and adverse birth outcomes compared to White people. Black, Hispanic, AIAN, and NHPI people had higher shares of preterm or low-birthweight infants and received late or no prenatal care compared to White people in 2020 (Hill et al., 2022). Asian people were also more likely to have low-birthweight babies than White people. Similarly, Black, NHPI, and AIAN infants were two times as likely to die as White infants (Ely and Driscoll, 2022) (see Figure 2-3). Asian infants had the lowest mortality rate. “Notably, disparities in maternal and infant health persist even when controlling for certain underlying social and economic factors, such as education and income, pointing to the roles racism and discrimination play in driving disparities” (Hill et al., 2022).

Chronic Disease

Chronic diseases¹⁵ and illnesses can cause financial stress, disability, and even death. According to CDC, chronic conditions—such as heart disease, stroke, cancer, type 2 diabetes, obesity, and arthritis—“are among the most common, costly, and preventable of all health problems and leading drivers

___________________

¹⁴ The United States is one of only six countries, and the only Organisation for Economic Co-operation and Development member, that does not mandate paid family leave at the federal level (BPC, 2022). The Family and Medical Leave Act provides only unpaid leave and job protection, only covers employers with 50 or more employees, and does not cover part-time employees (DOL, n.d.). This disproportionately harms underserved racial and ethnic low-income families, as they are less likely to be able to afford to take leave.

¹⁵ Chronic diseases are defined broadly as conditions that last 1 year or more and require ongoing medical attention, limit activities of daily living, or both.

of the nation’s $4.1 trillion in annual health care costs” (CDC, 2022). A recent study, using National Health Interview Survey data (1999–2018), found persistent racial and ethnic inequities in multimorbidity (defined as the presence of two or more concurrent chronic health conditions) in the United States over a 20-year period (Caraballo et al., 2022). Annual prevalence of important chronic conditions was persistently higher among Black Americans compared with Asian, Latino/Hispanic, and White people. The following paragraphs discuss two common U.S. chronic diseases—obesity and cardiovascular disease (CVD).

Obesity

Obesity can be thought of as a collection of diseases with multiple causes, contributors, and clinical expressions (in contrast to the common belief that it is typically caused by eating too much and moving too little) (NASEM, 2022a). Examples include dysregulation of hormones involved in thyroid function, hunger, and appetite satiety. This alternate view points out that contributors are extrinsic—outside of a person—and can lead to obesity in the presence of causes that are preventable, modifiable, and treatable. Examples include poor sleep quality, food insecurity, and exposure to energy-dense

foods (NASEM, 2022a). A 2021 study found that socioeconomic status, health behaviors, neighborhood environments, and early childhood health factors explained substantial racial and ethnic differences in obesity (Min et al., 2021). This framing is important because it affects relevant interventions. A CDC analysis of 2017–2018 U.S. data found that the age-adjusted prevalence was highest for non-Hispanic Black adults (49.6 percent) overall, followed by Hispanic (44.8 percent), non-Hispanic White (42.2 percent), and non-Hispanic Asian (17.4 percent) adults (Hales et al., 2020). Racial and ethnic disparities in obesity exist by the age of two (Pan et al., 2016). A study found that African American children had the highest prevalence of risk factors, whereas Asian children had the lowest (the sample size for Pacific Islander children was too small, so their data were not included in this study) (Isong et al., 2018). The rate of infant weight gain during the first 9 months was a major contributor to the gap, which accounted for 14.9–70.5 percent of explained disparities between White children and racially and ethnically minoritized peers. Another important contributor to these inequities—especially between White and Hispanic children—was socioeconomic status. A higher proportion of African American (56.7 percent), Hispanic (51.3 percent), and American Indian (50.1 percent) children belonged to households earning less than $25,000 per year compared with White (19.8 percent) and Asian (17.7 percent) children. The study found that early childhood risk factors, such as lack of fruit and vegetable consumption and amount of television viewing, were less important in explaining racial and ethnic differences (Isong et al., 2018).

Cardiovascular Disease

CVD is the leading cause of morbidity and mortality globally. Racial and ethnic inequities have been extensively documented, and although management of CVD has led to improved mortality, striking inequities in outcomes have persisted and widened among racial and ethnic groups (Glynn et al., 2019; Mazimba and Peterson, 2021; Singh et al., 2015). Black Americans experience the highest mortality rates attributable to CVD and stroke, with almost 30 and 45 percent higher mortality, respectively, than non-Hispanic White Americans (Churchwell et al., 2020). Recent data for NHPI people are limited; however, in 2014, they were 10 percent more likely to be diagnosed with coronary heart disease than non-Hispanic White people (OMH, 2021b). CVD is the leading cause of death among AIAN people, with over one-third of CVD deaths occurring before the age of 65 (Breathett et al., 2020). American Indian people are 1.5 times as likely to be diagnosed with coronary heart disease compared with the White population (Javed et al., 2022). Furthermore, hypertension prevalence rates reported among AIAN women (age 18–44) are significantly higher compared to non-Hispanic White women

(12 percent versus 8.2 percent) (Hutchinson and Shin, 2014). Several studies and papers find that these inequities are not fully explained by socioeconomic status, and several studies point to structural racism as a contributor (Churchwell et al., 2020; Javed et al., 2022; Mazimba and Peterson, 2021).

Intersections and Considerations

Although the committee was asked to focus on racial, ethnic, and tribal health inequity specifically, these populations have intersectional identities along multiple other axes that are also an important part of the picture when addressing policy needs (Cooper, 2015), including sex, gender, sexuality, age, disability status, geographical considerations, and citizenship. Inequalities exist in these dimensions for all racial and ethnic groups, and each could be reviewed separately; however, the committee considers only at the intersection of these factors with race and ethnicity (for example, the intersection of being Black and being a woman) in alignment with its statement of task. Experiences and health outcomes cannot be considered in a vacuum with regard to race and ethnicity but are impacted by these other aspects (Ritzer and Stepnisky, 2022). These identities combine in complex and interwoven ways to create novel interaction effects (Crenshaw, 1991). The committee provides several examples to illustrate the complex intersectional factors that need to be considered when developing or changing policy to advance racial, ethnic, and tribal health equity. One cross cutting consideration is the adverse consequences of limited English proficiency on health equity—see Chapter 5 for more information.

Disability

The intersection of race and disability represents a complex and interconnected history of oppression, stigma, and discrimination, resulting in disproportionate rates of disability among racially and ethnically minoritized people and inequities in health care access, health outcomes, and quality of life (Morgan et al., 2022; Varadaraj et al., 2021). Furthermore, racial inequities exist among people with disabilities, even those with the same medical condition or disability. For example, Black, AIAN, and other minoritized populations with autism are diagnosed later and more likely to be misdiagnosed (Morgan et al., 2022). Native American and Black people have the highest rates of disability among working-age adults, followed by White, NHPI and Hispanic, and Asian people (CDC, 2020; Courtney-Long et al., 2017; Siordia et al., 2017). Due to long-standing structural determinants of health, both racially and ethnically minoritized people and people with disabilities carry disproportionate burdens of ill health, face more significant barriers to accessing quality health care, are more likely to be

unemployed and have lower incomes, and experience discrimination from health care professionals (Gulley et al., 2014; Lagu et al., 2022; National Disability Institute, 2020; Peterson-Besse et al., 2014).

Law and federal policy have played a substantial role in advancing and institutionalizing racism and ableism. For example, forced and coerced abortion and sterilization are recurring themes for the disability community. In 1927, the Supreme Court case Buck v. Bell¹⁶ set a legal precedent that states may sterilize residents of public institutions. The court argued that imbecility, epilepsy, and feeblemindedness are hereditary and that residents should be prevented from passing these to the next generation. Buck v. Bell has not been overturned, but its reasoning has been discredited. In another example, a landmark 1999 Supreme Court case, Olmstead v. L.C.¹⁷ found that the unjustified segregation of people with disabilities is a form of unlawful discrimination under the Americans with Disabilities Act and led to developing new opportunities for individuals to live and work in their communities instead of institutionalizing them.

One current federal policy example is Supplemental Security Income (SSI). It provides essential economic assistance to people with disabilities, including 1.1 million children and their families (SSA, 2022). In addition to important financial support for low-income families, in many states, qualifying for SSI makes recipients eligible for Medicaid and other social supports (Musumeci and Orgera, 2021). Black, Hispanic, and AIAN people make up disproportionate shares of SSI enrollees compared to White people (Musumeci and Orgera, 2021). The delivery of these vital programs can be hampered by racial and ethnic inequities in Social Security Administration (SSA) policies and processes, creating inequities in individuals’ and families’ ability to access and maintain SSI benefits. Administrative, literacy, citizenship, and financial barriers to applying for and maintaining SSI benefits disproportionately affect racially and ethnically underserved individuals and families. Eligibility is limited to U.S. citizens except in very limited circumstances (SSA, 2023).¹⁸ SSA’s complex rules for financial eligibility and stringent definition of disability, which requires substantial documentation to prove, pose barriers to participation. Of the 29 percent of nonelderly adults with Medicaid who have a disability, only 41 percent have SSI (Musumeci and Orgera, 2021). Furthermore, only 35 percent of medical determinations were approved upon initial application; those that were appealed via an administrative law judge

___________________

¹⁶ Buck v. Bell, 274 U.S. 200 (1927).

¹⁷ 527 U.S. 581; 119 S.Ct. 2176 (1999).

¹⁸ SSA determines income based on an expansive definition that includes earned and unearned cash and noncash, including “in kind” support, such as food and shelter provided or paid by another person. The amount of SSI that a person is eligible to receive cannot exceed the maximum federal benefit rate ($794/month in 2021) minus their calculated income (SSA, n.d.).

hearing or a higher appeals level, processes that are challenging to navigate without legal representation, were successful 40 percent of the time. During the reapplication and appeals process for SSI, individuals may be at risk for income insecurity. Federal SSA policy includes provisions that suspend or terminate benefits to children who become involved in the criminal legal or child welfare systems; these are disproportionately members of racially and ethnically minoritized populations. Restrictive SSA criteria for asthma and sickle cell disease, illnesses that disproportionately affect Black children, create inequity (NASEM, 2020a). Because eligibility determination relies on other systems—schools, health care providers, child welfare, criminal legal system, and others—the structural racism and inequities that may affect those systems are likely to shape the administration of benefits (Community Legal Services of Philadelphia, 2020).

Sex and Gender

Sex and gender affect health and health outcomes in complex ways. Sex can impact disease risk and progression and treatment outcomes through genetic, epigenetic, and physiological pathways, including differences in reproductive autonomy, hormonal milieu, and metabolism (Crimmins et al., 2019; WHO, 2021). Much as race is a social construct,¹⁹ gender norms, access to resources due to the socialization of gender, perception of illness and health behaviors, access to health care, and biases in the health care system all affect health and health outcomes. For example, in a study assessing the influence of social support on nutritional risk, White women who lacked social support were less likely to be affected by poor nutrition than Black women who did (Locher et al., 2005). Black women are more likely than White women to develop CVD, and nearly 60 percent more likely to have high blood pressure (a risk factor for CVD) (CDC, 2023b; Nayak et al., 2020); however, research shows differences in referrals for cardiac care that are most evident for African American women, then African American men, compared with White men and women (IOM, 2003; Schwartz et al., 1999).

Furthermore, much of the discussion surrounding racial and ethnic health inequities mirrors that on the effect of sex/gender on health outcomes, including the importance of research that distinguishes between females and women from males and men, the historical role and power dynamics between men and women, and the role of policy across the same multiple domains that impact health outcomes. A salient example is AIAN women who are disproportionally affected by violence; 84 percent experience physical and sexual violence in their lifetime, with a majority of it perpetrated by

___________________

¹⁹ Genetic ancestry affects human health (for example, sickle cell); however, this is distinct from the impact of race, which is a social construct that has its foundations in systemic racism.

non-American Indian offenders (Rosay, 2016). Federal policies regarding public safety on reservation and village lands has resulted in an environment where tribal authorities are often unable to prosecute non-Indian offenders. Assumptions have been made that the violence only exists on tribal lands; however, a study of AIAN women in Seattle found that 94 percent of them had been sexually assaulted and or coerced into sex in their lifetime, with only 8 percent of those who reported to local law enforcement seeing their perpetrator convicted (Urban Indian Health Institute, 2018).

Societal prejudices and stereotypes of AIAN women and girls contribute to rape culture that creates further barriers to prevention, intervention, and justice. These factors mesh into a crisis of epic proportions, with women and girls going missing and being murdered at disproportional rates. A first-of-its-kind study in 2018 found that this was occurring at high rates in 71 cities across the country. However, it noted that the data were severely limited due to law enforcements’ noncollection and nonreporting of race and ethnicity, effectively making the crisis invisible. The study additionally found lack of media coverage; what did occur was rife with stereotypes of sex workers and runaways (Lucchei and Echo-Hawk, 2018). Through tribal advocacy and congressional support, recent legislation (Violence Against Women Act)²⁰ was passed to address this crisis. Yet, a 2021 Government Accountability Office report found the implementation was lacking and many of the tasks had yet to be completed by the Department of Justice, illustrating the need for accountability mechanisms in congressional and White House–mandated equity efforts (GAO, 2021).

Racially and Ethnically Minoritized LGBTQ+ People

As noted, structural racism can be experienced differently by different groups and compound other forms of structural discrimination, including structural sexism, heterosexism, and cisgenderism. Given the co-occurrence and mutual reinforcement of these three factors, racially and ethnically minoritized LGBTQ+ people often experience greater health inequities than their heterosexual and White counterparts (Hatzenbuehler et al., 2013; IOM, 2011; NASEM, 2020b), which plays out in sexual and reproductive health. For example, HIV and other sexually transmitted infections (STIs) are higher in many LGBTQ+ populations compared to their heterosexual and cisgender peers. Gay, bisexual, and other men who have sex with men have the highest incidence of HIV; they account for 66 percent of new infections each year even though they are only 2 percent of the population. However, the highest burden is among Black and Latino gay, bisexual, and other men who have sex with men (HIV.gov, 2023). Studies show that

___________________

²⁰ Title IV of the Violent Crime Control and Law Enforcement Act, H.R. 3355.

decades of not only sexual orientation–related but also racial and ethnic oppression, discrimination, and disenfranchisement compound STI risk and transmission (NASEM, 2021c). Moreover, STI rates among transgender women, particularly Black and Latina women, are some of the highest in the United States (Becasen et al., 2019; Poteat et al., 2014). Racially and ethnically minoritized transgender women possess multiple marginalized statuses and thus experience multiple forms of discrimination, including racism, sexism, and cisgenderism, all of which compound one another and simultaneously increase STI risk and undermine access to prevention, diagnosis, and treatment (NASEM, 2021c). Data on STIs among sexual minority women are not routinely collected, but existing data show that racially and ethnically minoritized women—who face sexual orientation–related stigma and racism—may experience a higher burden compared to their White counterparts (NASEM, 2021c).

In recent years, researchers have begun explicitly assessing the simultaneous impact of structural racism and other forms of structural discrimination, including structural heterosexism, on the health of racially and ethnically minoritized LGBTQ+ people. For example, English and colleagues (2022) examined associations between structural racism, structural heterosexism (modeled using anti-LGBTQ policies), and mental and behavioral health among Black and White young sexual minority men. The researchers found that for Black sexual minority men, state-level structural racism and state anti-LGBTQ+ policies were both independently positively associated with anxiety, heavy drinking, and other factors; the positive associations between structural racism and several psychological and behavioral health outcomes were stronger in states with more anti-LGBTQ+ policies, underscoring the compounding impact of both structural racism and heterosexism. In contrast, no association appeared between structural racism or heterosexism among White sexual minority men (English et al., 2022). However, policies such as sexual orientation–related nondiscrimination laws can be protective for Black LGBTQ+ populations, including Black lesbian and bisexual women; Everett and Agénor (2023) found that higher numbers of state-level sexual orientation–related nondiscrimination laws were associated with lower risk of maternal hypertension. These laws were also associated with a lower risk of maternal hypertension among White women, regardless of sexual orientation (Everett and Agénor, 2023). Passing, implementing, and enforcing laws and policies that prevent discrimination and advance equity are necessary, but LGBTQ+ racially and ethnically minoritized communities have many strengths that can be leveraged to advance health equity. For example, a 2020 survey of 38 LGBTQ+identified people of color in New York City (guided by the health equity promotion model) found common strengths identified by participants were safety, acceptance, and support; interconnectedness and resource sharing;

and advocacy, collective action, and community potential (Hudson and Romanelli, 2020). Moreover, Agénor et al. found, in two different studies, that social support from other LGBTQ+ people of color, health information sharing within LGBTQ+ communities of color, peer advocacy during clinical encounters, and receiving care from health care providers who shared their lived experiences facilitated access to and receipt of needed sexual and reproductive health care among transmasculine young adults of color (Agénor et al., 2022a,b).

Rurality

In the context of “place” as a social determinant of health (SDOH), residence in a rural area is strongly linked with poor health outcomes. Over at least the past 50 years, this has evolved into one the strongest determinants of excess mortality (Cosby et al., 2019). The population-based mortality disparity is on par with race and level of education as a predictor of early death (NASEM, 2018a). Poverty and rurality have a strong interaction as well (USDA, 2022). This rural health penalty is not equally distributed; measures of poor health and poverty are more concentrated among Black, Hispanic, and AIAN than White populations (Cosby et al., 2019; James et al., 2017). Death from heart disease, cancer, and stroke are considerably higher among rural Black people compared to those living in urban areas (Probst et al., 2020).

Resource constraints that drive health and economic inequities are often more severe in rural areas, demonstrated by the lack of transportation resources, inadequate economic opportunity, diminished access to health care, and a preponderance of food insecurity in rural areas (NASEM, 2021b; Romanello et al., 2022). Rural hospital closures are more likely in communities with higher percentages of Black and Hispanic residents, exacerbating pre-existing inequities (NASEM, 2018a). AIAN people in rural areas are affected by these same resource imbalances and further impacted by a woefully underfunded IHS (see Chapter 5 for more information on the IHS) (Heisler and McClanahan, 2020; NASEM, 2017).

Rurality is a strong predictor of poor health outcomes and health inequities (James et al., 2017). Areas with persistent poverty are not evenly distributed but are concentrated in the mid-South and AIAN reservations and villages. Although mirroring those in more populous areas, racial and ethnic inequities are more concentrated (NASEM, 2021b). Different strategies to address health inequities and to improve the overall level of care in rural areas are needed given the pre-existing deficit of wealth and public infrastructure. These approaches need to be community driven and funded in ways that acknowledge the unique implementation challenges of working with small populations.

Citizenship and Immigration Status

Citizenship and immigration status are powerful determinants of health status and life chances (NASEM, 2018b). U.S. citizenship is mostly understood as a legal membership at the national level, with such membership granted through birthright based on the Constitution’s 14th Amendment, 1924 Citizenship Act,²¹ and Immigration and Nationality Act of 1965.²² Federal law limits certain benefits and opportunities to U.S. residents that have significant implications for health outcomes, including on employment in most federal government jobs, voting in national elections, priorities in bringing family members to the United States, and protections from threats to immigration status renewal based on public charge considerations and deportations based on criminal convictions even for legal permanent residents (NASEM, 2015, 2018b). U.S. citizenship also forms the basis of access to many benefits and protections at the state level, including for government employment and voting (Citizenship and Immigration Services, 2020).

In addition to U.S. citizenship status, immigration status also plays a powerful role in limiting access to federal programs that can address key SDOH and increasing risks to immigration enforcement actions that may more directly impact health status (NASEM, 2015, 2018b). The federal government has exclusive authority to allow persons to enter the United States, deport or remove people, and provide work authorization and the ability to remain in the country for specified periods (American Immigration Council, 2021). The highest levels protections are accorded to legal permanent residents (often referred to as “green card” holders); they are able to access most federal benefits and state benefits without risk of deportation or removal, with renewal of status virtually guaranteed (American Immigration Council, 2021). Other statuses are temporary, tied to such factors as employment (H visas), education (F and M visas), and visitors on business or tourism (B visas), and they vary with respect to the ability to obtain work authorization and access to federal benefits. Finally, the federal government also provides certain persons with temporary protected status based on emergency conditions in their countries of origin, Deferred Action for Childhood Arrivals based on age and year of entry, and asylee or pending asylee status (American Immigration Council, 2021); see Chapter 5 for more information. These temporary statuses also vary greatly in the kinds of federal benefits and, by extension, state and local benefits they provide. Finally, people with no or expired legal status are ineligible for most federal benefits and protections and subject to detention or deportation (Colbern and Ramakrishnan, 2021). These immigration enforcement actions affect

___________________

²¹ Pub. L. 68–175.

²² Pub. L. 89–236.

the health status of not only the particular individuals involved but also family members living in multigenerational and multistatus households (Castañeda and Melo, 2014).

DATA GAPS AND OPPORTUNITIES

High-quality and accurate data are needed for several reasons in relation to racial, ethnic, and tribal health equity and policy making. One key reason is accountability—if data are poor, inequities can be hidden (intentionally and unintentionally) or ignored, and progress cannot be monitored (NCQA, 2021a; Rubin et al., 2018). The ability to conduct equity assessments to identify the impact of federal policies on health equity and then identify and remove barriers to accessing government programs is also contingent on having data available (Equitable Data Working Group, 2022). Accurate and high-quality data are also needed to guide the development and implementation of interventions. In addition, data can point to both health improvement and declines to guide resource allocation. If data are unavailable or inaccurate for a certain population, they may not receive the resources they need (Erickson et al., 2021; NCQA, 2021a; Rubin et al., 2018). Self-reported data on race and ethnicity are also important aspects, as they better reflect how people identify themselves.

Since 1977, the Office of Management and Budget (OMB) has issued minimum standards for maintaining, collecting, and presenting federal data on race and ethnicity. These were last revised in 1997, when the federal government separated “Asian or Pacific Islander” into “Native Hawaiian or Other Pacific Islander,” changed “Hispanic” to “Hispanic or Latino,” allowed respondents to self-identify with more than one race, and required federal agencies to report data on multiple racial categories rather than combining them into a general category of “multiracial” (OMB, 1997). These minimum standards inform all federal race and ethnicity data collection efforts and are critical to understanding patterns, conditions, and outcomes, including those related to health and health care and the SDOH. Not all federal agencies collect and report high-quality and accurate data on these minimum racial, ethnic, and tribal population categories, which is a significant barrier to achieving data equity. This lack of data also contributes to health inequities, as policy decisions are not fully informed by the data.

The OMB revised 1997 standards included minimum categories for race for AIAN, Asian, Black or African American, Native Hawaiian or Other Pacific Islander, and White, with individuals able to select one or more categories (OMB, 1997; OMH, 2021a). However, although these standards were to be effective/fully enforced by 2003, OMB has not done so (e.g., NHPI are often not separately reported and are combined with

Asian) (AAPI Data and National Council of Asian Pacific Americans, 2022; OMB, 1997). In addition, the OMB minimum categories for ethnicity are Hispanic or Latino/a and Not Hispanic or Latino/a.

In January 2023 of the Federal Register, OMB proposed important changes to the data collection on race and ethnicity, with a published notice on “Initial Proposals for Updating OMB’s Race and Ethnicity Statistical Standards.”²³ that follows over a decade of constituent engagement and questionnaire testing by the Census Bureau. Until a revision, however, the 1997 standards remain in place. Some of the important proposed changes include the following (see also Chapter 8):

1. Collect race and ethnicity information using one combined question, rather than have a question on Hispanic ethnicity separate from the question on race.
2. Add “Middle Eastern or North African” (MENA) as a new minimum category.
3. Require the collection of detailed race and ethnicity categories by default, “unless an agency determines that the potential benefit of the detailed data would not justify the additional burden to the agency and the public or the additional risk to privacy or confidentiality.”
4. Update terminology deemed to be inaccurate or archaic, including removing “Negro” from the Black of African American definition; removing “Far East” from the Asian definition—replacing with “East Asian”; removing the phrase “who maintain tribal affiliation or community attachment” in the AIAN definition; and removing “Other” in the Native Hawaiian and Other Pacific Islander definition.

The committee provides input on these proposed changes in Chapter 8.

The Department of Health and Human Services (HHS) has used the OMB minimum standards in most of its data collection efforts since 1997. Data are collected through surveillance systems, national surveys, administrative programs, and clinical trials. In 2011, HHS developed new standards for data collection of race and ethnicity to understand better the heterogeneous racial and detailed-origin subgroups that comprise the broader OMB categories (ASPE, 2011). The new HHS race standard disaggregates the Asian (Asian Indian, Chinese, Filipino, Japanese, Korean, Vietnamese, and other Asian) and Native Hawaiian or Other Pacific Islander (Native Hawaiian, Guamanian or Chamorro, Samoan, and other Pacific Islander) categories into more fine-grained detailed groups that correspond to ethnicity and country or place of origin. In addition, the new

___________________

²³ See https://www.federalregister.gov/documents/2023/01/27/2023-01635/initial-proposals-for-updating-ombs-race-and-ethnicity-statistical-standards (accessed March 8, 2023).

HHS ethnicity standard disaggregates the Hispanic or Latino/a category as follows: Mexican, Mexican American, or Chicano/a; Puerto Rican; Cuban; and other Hispanic, Latino/a, or Spanish origin (OMH, 2021a).

In contrast, the new HHS race standards do not disaggregate the Black or African American category by ethnicity, country, or place of origin (e.g., African, Caribbean, Haitian, Nigerian). Since 2000, the proportion of Black immigrants in the United States has increased steadily, from about 600,000 in 2000 to 2.0 million in 2019, with African immigrants representing the largest share of arrivals in 2019 and the Caribbean representing the most common region of birth. Because of pronounced social, economic, and health differences between U.S.-, African-, and Caribbean-born Black U.S. individuals, it is important for data collection efforts to capture the heterogeneity of the population. Not doing so undermines efforts to address the specific and unique needs of diverse communities and promote health equity for all Black populations (Tamir, 2022). Similarly, the AIAN category does not recognize distinct Tribal Nations (OMH, 2021a).

Given the relative recency of the revised HHS race and ethnicity standards, the vast majority of data on health and health care outcomes across and within racial and ethnic groups continue to use the OMB minimum standards. Unfortunately, the most reported groups remain non-Hispanic Black, non-Hispanic White, and Hispanic or Latino/a, for reasons discussed below (including the lack of capacity to report more granular data due to sample sizes too small to produce reliable estimates).

In addition to HHS data collection, the 2020 census collected detailed data on ethnicity or enrolled tribal affiliation for each of the OMB racial and ethnic categories. This represents a huge opportunity for (1) examining within-group disparities among OMB racial and tribal groups and (2) redesigning primary sampling units and more targeted data collection of hard-to-survey groups because of the more detailed 2020 census data on race and ethnicity population distribution (Census Bureau, 2019).

Gaps in Race and Ethnicity Federal Data Collection Efforts

Continued gaps in federal data collection on race and ethnicity pose a problem and contribute to data inaccuracy. These gaps, which have ramifications for data equity efforts, include (1) lack of disaggregation of diverse groups, (2) lack of oversampling, (3) inability to capture intersections with other social identity factors, (4) lack of granular data on small racial and ethnic groups, and (5) gaps in contextual data. Accurate data are necessary to inform knowledge and action on racial and ethnic health inequities and their social determinants.

The first major barrier is that data on OMB-defined historically underrepresented groups are not disaggregated (Panapasa et al., 2011). They reflect

a conglomeration of various ethnicities, countries or places of origin, or Tribal Nations. Failing to capture the diversity within OMB-defined racial and detailed-origin groups obscures significant within-group health inequities and hampers efforts to mitigate or eliminate health inequities between and within racial groups. These differences result from social and economic inequities based on ethnicity, country or place of origin, immigrant or refugee status, geography, language, skin color, historical context, or federal recognition.

This lack of disaggregation reflects both historical and contemporary norms, values, and practices rooted in racism, xenophobia, colonialism, and imperialism that view and treat racially and ethnically minoritized populations as monoliths and erase the complexity, nuances, and specificities of their social positions and lived experiences (see earlier sections of this chapter for more information). Multiracial people are a very heterogeneous population. They represent diverse social and economic positions, racial and ethnic backgrounds, histories, countries or place of origin, skin colors, cultures, tribal affiliations, and health and health care needs. All these concerns are masked by the lack of data disaggregated by race or ethnicity for this multifaceted group (Holland and Palaniappan, 2012; NCAI, 2021; Rodríguez, 2021; Rubin et al., 2018; Urban Indian Health Institute, 2020; Williams and Jackson, 2000; Zambrana et al., 2021). Data disaggregation can be costly with added confidentiality issues; however, prioritizing these populations is essential.

The second gap is the lack of oversampling, especially of AIAN and NHPI populations, in national health surveys and other relevant federal data collection efforts to produce reliable statistics on these minimum OMB categories (Faircloth et al., 2015; Galinsky et al., 2019; Panapasa et al., 2011; Wu and Bakos, 2017). In addition, adequate granular/detailed data are not collected on the minimum OMB categories (such as Asian, Native Hawaiian, and Pacific Islander) or by geography. This leads to inadequate data on the health of these diverse communities and precludes meaningful action to achieve health equity (Islam et al., 2010; Johnson et al., 2010; Nguyen et al., 2022). To offset the failure of federal agencies to oversample, researchers supported by federal data collection efforts and other funders conduct smaller specific group studies to address data gaps. However, the data on these smaller groups are extrapolated to other Asian, Native Hawaiian, and Pacific Islander groups or interpreted as representative of the total population; these smaller studies are not robust, are based on convenience samples, and suffer from bias, limiting generalizability to the population group (Holland and Palaniappan, 2012; Shah and Kandula, 2020; Yom and Lor, 2022). Despite barriers to expanding sample sizes and collecting data on small or geographically remote populations (e.g., cost, data reliability, privacy), expanding sampling frames to generate statistically reliable estimates of the population at varying levels of geography is important for policy and program development. Decisions to do so will need to consider a number of factors, such as population size, the magnitude of the

disparity based on scientific research, and cost and feasibility (see Chapter 8 for a recommendation and more details on oversampling).

The third gap is the lack of data on the intersection of race and ethnicity and racism with other axes of social identity and inequality, including gender, ableism, sexism, cisgenderism, sexual orientation, heterosexism, socioeconomic position and classism, nativity and xenophobia, and age and ageism. Together, these issues impact the health and health care outcomes of multiply marginalized individuals and populations (e.g., LGBTQ+ people, older adults, youth, low-income people) in compounding and unique ways (Bowleg, 2012, 2017; Collins and Bilge, 2020). However, the vast majority of federally collected health data are only reported in relation to a single axis of social identity or inequality at a time, thus rendering invisible the specific health experiences and needs of multiply marginalized groups and undermining the development and implementation of tailored actions that promote their health. Moreover, health surveys and other federal data collection efforts tend to include only some of the measures needed for an intersectional analysis of racial and ethnic health inequities. For example, although the National Health Interview Survey, the nation’s leading source of population health data from CDC, does collect information on sexual orientation identity, it does not include other measures of sexual orientation (e.g., sexual behavior) or gender identity. This inhibits the ability of researchers, policy makers, and program planners to identify racially or ethnically minoritized LGBTQ+ people in the data and formulate policies, programs, and practices that promote their health (Zambrana et al., 2021).

Moreover, substantial within- and across-group disparities are masked, and inappropriate interpretations and conclusions can be drawn when race and ethnicity data are not cross-tabulated with other social indicators. For example, poor average health outcomes among Black populations are often erroneously attributed to socioeconomic inequities alone. However, analyses that stratify data for racial and ethnic groups by socioeconomic status often show a more complex picture, implicating other factors related to interpersonal, institutional, and structural racism as essential drivers of health inequities (Williams and Jackson, 2000).

Relatedly, efforts are inadequate to collect data on persons speaking a language other than English at home and limited English proficiency. The census has reliable questions, but these are not typically administered as part of federal data-collection efforts (Flores, 2020).

The fourth gap pertains to the absence, or masking, of subgroups in data collection and reporting efforts. In particular, OMB standards do not include Arabs and Arab Americans or people of MENA descent as a distinct racial or ethnic category (Wiltz, 2014). These individuals have to select a category that may not reflect their personal or social identity, culture, history, or lived experience. Due to the lack of relevant options, some may

select the “White” or “other” options, which erases their identity and experience, erases their existence in the data, prevents generating data on their health outcomes and needs, and obscures data for other populations (Rubin et al., 2018). Similarity, ethnicity is treated separately from race in most data collection efforts, forcing Latino/a individuals to artificially choose a race in some cases (Flores, 2020).

Additionally, smaller populations, such as AIAN and NHPI populations, are often entirely excluded from data collection and reports due to their small numbers, limiting access to resources (mandated under treaty and trust responsibility for the AIAN population) (Panapasa et al., 2011). Survey data collection on the Pacific Islander jurisdictions is limited as well (Erickson et al., 2021; Friedman et al., 2023). National datasets routinely relegate these groups to the “other” racial category, which grossly misrepresents and denies their existence. This conceals their social, economic, and health experiences and needs and undermines attempts to create relevant, tailored action to improve the health of the various and diverse groups included in this nebulous “catchall” category (Holland and Palaniappan, 2012; NCAI, 2021).

Relatedly, the federal definition of AIAN,²⁴ which is limited to those registered in federally recognized Tribal Nations, not only underestimates but also misclassifies AIAN people who do not fit this definition. This further undermines the generation of high-quality and accurate data (NCAI, 2021; Rubin et al., 2018; Urban Indian Health Institute, 2020). One added complication, either unknown by data collectors or ignored in the categorization, is that in the eyes of the federal government, the tribes are first and foremost political bodies or domestic dependent nations (see earlier in this chapter for more details). An additional complication in reporting correctly and comprehensively on the AIAN population is that researchers conducting studies or collecting data might be unaware of this unique political status and lump them in with all other groups, usually as “other.” Use of data categories needs to respect the sovereign status of Tribal Nations to determine their own citizenship; federal, state, and county collection of tribal enrollment status is inappropriate without the express permission of a specific tribe. Collecting tribal affiliation is more appropriate, as it does not denote tribal citizenship and is cognizant of the historical context of disenrollment, descendants, and federal initiatives that separated AIAN people from their communities; however, it must be respectful of AIAN data sovereignty (Urban Indian Health Institute, 2020).

Due to extermination, reservation, termination, and assimilation policies, the AIAN population was splintered and now includes enrolled members, those purposively lost through assimilation, those “unenrolled,”

___________________

²⁴ Native Hawaiians are defined as descendants of the aboriginal people of State of Hawaii. Congress has not formally recognized them as a Tribal Nation.

and those who died/were killed. With these losses, any accountability and the capacity to generate accurate population counts were lost. Although the Self-Determination Era (characterized by an explosion in civil rights activism) began in 1975, AIAN people have seen few gains in health or SDOH. The same spirit of purposeful, targeted, strategic measures that caused these inequalities needs to now be applied to reversing the harmful policies’ effects.

Data related to inequities and resiliencies are severely lacking for the AIAN population, which directly impacts the allocation of resources legally owed through the treaty and trust responsibility. The COVID-19 pandemic, which disproportionately affected AIAN people (Arrazola, 2020; Hatcher et al., 2020), provides a poignant example of how inadequate data can drive health inequity. AIAN-specific data on infection rates and mortality were mostly missing, with great variation between states (Erickson et al., 2021; GAO, 2022). As a result, resource allocation to address disproportionate health needs was inadequate, as federal agencies, including CDC, relied heavily on these faulty data. Disaggregation of data is a priority of AIAN communities, but it is increasingly difficult to disaggregate small samples; in addition, the population is small because of past genocide and the current lack of proper data collection.

A fifth gap is that data that reflect groups’ social, economic, and historical contexts are not collected. In particular, national health surveys and other federal research efforts (e.g., clinical trials—see Chapter 5) fail to include measures of historical and contemporary structural, institutional, community, and interpersonal racism and xenophobia, which shape racial and ethnic health inequities (Brown and Homan, 2023).²⁵ As a result, available information on these inequities cannot be accurately contextualized, interpreted, and acted upon using health data alone. Additionally, although the census and other federal (e.g., administrative) data sources provide essential information on the SDOH of racial and ethnic health inequities (e.g., poverty, mortality, income, housing, education, and environmental risks), these data are typically not readily linked to or integrated with national survey, administrative, or clinical trial health data. This limits their usability and uptake in public health and health equity research (Knight et al., 2021; Rodríguez, 2021). The measures used would depend on the topic or SDOH area under study²⁶ (for examples of such measures, on a variety of SDOH, see Adkins-Jackson et al., 2022;

___________________

²⁵ For example, measures on explicit rules and laws, nonexplicit rules and laws, and area-based or institutional nonrule measures and expanding the use of structural measures that extend beyond the current psychosocial individual-level measures (Krieger, 2020).

²⁶ For example, to study the impact of structural racism on myocardial infarction, Lukachko and colleagues (2014) used the following indicators “(1) political participation; (2) employment and job status; (3) educational attainment; and (4) judicial treatment. State-level racial disparities across these domains were proposed to represent the systematic exclusion of Blacks from resources and mobility in society” (p. 42).

Agénor et al., 2021; Alson et al., 2021; Dougherty et al., 2020; Follis et al., 2023; Greenfield et al., 2021; Hardeman et al., 2022; Krieger, 2020; Krieger et al., 2016, 2017a; Mesic et al., 2018; Siegel et al., 2022; Wallace et al., 2015; Williams, 2016).

The census and other federal sources of data on SDOH do not include measures of historical or contemporary structural racism (see earlier). This would include measures of laws, policies, and institutional practices that minoritize Black, AIAN, Latina/o, and other communities across multiple social systems. Policies, such as Jim Crow laws, redlining practices, environmental zoning laws, and school funding policies, underlie and continue to drive racial and ethnic social and economic and ensuing health inequities (Chapters 3–7 provide examples). This undermines the ability of policy makers, researchers, and program officials to identify how historical and contemporary structural and social inequities shape racial and ethnic health inequities and, in turn, thwart action on these root causes at the federal, state, territorial, tribal, and local levels (Rodríguez, 2021). Moreover, failing to contextualize racial and ethnic health inequities by the historical and contemporary structural and social inequities that shape them fosters their interpretation as innate, inherent, and immutable. It supports an inaccurate view that they reflect individual-level biological and behavioral factors rather than the societal effects of racism and other forms of discrimination (Knight et al., 2021; NASEM, 2023).

Challenges in Race and Ethnicity Federal Data Collection Efforts

Contemporary race and ethnicity data collection practices stem from accounting, statistics, and epidemiology, which originated with the counting and monitoring of enslaved people and “imperial subjects” and were deployed in the context of eugenics in order to advance and justify slavery, colonialism, and imperialism (Gampa et al., 2020). In addition, due to structural inequities that undermine Black, AIAN, and other minoritized people’s access to education and employment, they have been systematically excluded from and underrepresented in the fields charged with collecting data on race and ethnicity, including data science, statistics, demography, and epidemiology (Downs, 2021). Data collection efforts and their interpretation, presentation, and use reflect the priorities, preferences, and objectives of White and Western perspectives, which diverge in important ways from the lived experiences of racialized people and communities (Knight et al., 2021). Conversely, race and ethnicity and data collection have been and can be used to undermine the health and well-being of minoritized individuals and groups. Historical and contemporary race and ethnicity data collection efforts have been used to actively surveil, exclude, and criminalize underserved communities (e.g., Black, Native, Asian, Latino/a,

Pacific Islander, and Arab communities) directly in conflict with reducing or eliminating health inequities and promoting health equity (Knight et al., 2021; Rodríguez, 2021).

As described by the Equitable Data Working Group, “federal collaborations with state, local, territorial and tribal governments can yield high-quality demographic data when all partners see value in its collection and use” (Equitable Data Working Group, 2022, p. 10). However, current race and ethnicity measures and data collection, interpretation, translation, and dissemination efforts do not reflect the relevant communities’ priorities, preferences, perspectives, and needs, which undermines their utility for informing community-centered actions that advance racial and ethnic health equity (Rubin et al., 2018; Urban Indian Health Institute, 2020). In addition, research studies that seek to investigate and address these issues in collaboration with impacted communities are usually sidelined from funding or publication in top-tier medical and public health journals, which further undermines the data and evidence available to inform equity-focused, community-centered solutions (Mervis, 2019).

To benefit Black, AIAN, NHPI, and other minoritized communities and advance health equity, race and ethnicity data collection efforts have to explicitly inform the equitable allocation of health-promoting resources to underresourced communities through laws, policies, programs, and community efforts. Additionally, care is required as data are disaggregated to avoid identifying and undermining the privacy of individual survey or study participants, particularly from small racial and ethnic groups while also achieving statistical reliability (Rubin et al., 2018; Urban Indian Health Institute, 2020). Disseminating point estimates by population group could include margins of error at the 95 percent confidence interval, with no predetermined cutoff on sample size as long as data privacy concerns are addressed.

A unique challenge for AIAN population data is the astounding levels of misidentification of race by health professionals, researchers, and medical examiners, for example, who collect data (Haozous et al., 2014). Trends in Indian Health, the official report of the IHS, includes multiple indications in the report acknowledging that the numbers are problematic, especially as received from states’ data misreporting of race (IHS, 2014). Furthermore, no report has been published since 2014. As bad as the known numbers for all aspects of AIAN health are for life expectancy, COVID-19 exposures, and death, for example, the actual burden of disease is underreported (Haozous et al., 2014).

Health Care Data

Opportunities exist to improve the plethora of health care data collected to improve data equity. These data reflect the individual, provider to

state/multistate, and national levels (see Figure 2-4) and are from various sources, including from surveys, billing records, and medical records. Entities collecting these data include payers and providers (hospitals, providers, clinics). A 2009 Institute of Medicine report provided recommendations on standardizing race, ethnicity, and language data to improve health care quality. In addition, the data can be used for multiple purposes: provider payment (see Chapter 5 discussion on value-based payment), a primary source for clinical trials and health services research, quality improvement initiatives, and incentives to improve individual behaviors.

Although health care data are rich, several gaps or issues are noted in the IOM (2009) report that continue to be relevant today:

• The entities listed above do not collect data on the entire population or have the capabilities to do so.
• Data are fragmented and not integrated.

• Health information technology can help streamline data collection and integration, but the maturity of technology varies across health care systems.
• Redundancy exists in the collection of race, ethnicity, and language data.
• Comparison groups cannot be equivalently stratified, given the lack of data standardization.

In addition, a 2022 systematic review of SDOH data in electronic health records yielded quality problems, such as misclassification bias of race, ethnicity, or country-of-origin data and incomplete data (Cook et al., 2022). Given these gaps and validity issues, data standardization in health care systems is important to achieving data equity. Several efforts are underway. In 2022, CMS’ Office of Minority Health released its framework for health equity 2022–2032, and one of the priorities is to expand the collection, reporting, and analysis of standardized data (McIver, 2022, p. 2). Several collaborations are cited, including partnership with HHS to improve interoperability and demographic and SDOH data standards.²⁷ In addition, the National Committee for Quality Assurance updated guidance for health plans to stratify quality data by race and ethnicity beginning in measurement year 2020 (NCQA, 2021b), and the Joint Commission, which accredits over 22,000 health care organizations, released new requirements to reduce health care disparities (Joint Commission, 2023); these include not only having the medical record reflect care, treatment, and services but also including race and ethnicity. CMS and the Joint Commission, recognizing the important role of these data in advancing health, are making early but limited efforts to ensure that these are captured for a subset of patients.

The United States Preventive Services Task Force (USPSTF),²⁸ the national organization that promulgates recommendations on clinical preventive services, has launched an initiative to address structural racism that can serve as a model for other regulatory agencies. Given the intrinsic deficiencies in representation of racial and ethnic minorities in clinical trials (see Chapter 5), prevention guidelines shared by USPSTF will reflect those same deficiencies. Disease processes manifest differently among various racial and ethnic groups,

___________________

²⁷ For example, interoperability with 2011 HHS Data Standards, U.S. Core Data for Interoperability standards, HHS Disparities Action Plan, CMS Accountable Health Communities Model, Consumer Assessment of Health Care Providers & Systems, and Standardized Patient Assessment Data Elements.

²⁸ Title IX of the Public Health Service Act mandated the creation of USPSTF to “review the scientific evidence related to the effectiveness, appropriateness, and cost-effectiveness of clinical preventive services for the purpose of developing recommendations for the health care community, and updating previous clinical preventive recommendations, to be published in the Guide to Clinical Preventive Services” (USPSTF, 2019). Each year, the task force makes a report to Congress that identifies critical evidence gaps in research related to clinical preventive services and recommends priority areas that deserve further examination (USPSTF, 2019).

due to genetic variation and the effects of structural racism (Gee and Ford, 2011; NASEM, 2022b). Disproportionately adverse clinical outcomes in diseases, such as cancer and CVD, reflect the different environmental exposures, health care access, and life experiences among racial and ethnic minorities (NASEM, 2017). Recommendations based on primarily White cohorts of study participants are not generalizable²⁹ to all populations (NASEM, 2022b). In 2021, USPSTF identified systemic racism as a factor undermining the utility of its current recommendations and the integrity of the recommendation development process. It has six core actions to address racism in future recommendations (Doubeni et al., 2021, p. 628):

1. Consider race primarily as a social and not a biological construct, and use consistent terminology throughout recommendation statements to reflect this view.
2. Promote racial and ethnic diversity in addition to gender, geographic, and disciplinary diversity in USPSTF membership and leadership, and foster a culture of diversity and inclusivity as an enduring value of USPSTF. This will be assessed annually before soliciting nominations for new members and internally assigning leadership roles.
3. Commission a review of the evidence, including an environmental scan and interviews with clinicians, researchers, community leaders, policy experts, other guideline developers, and patients from groups that are disproportionately affected to summarize the evidence on how systemic racism undermines the benefits of evidence-based clinical preventive services and causes preventable deaths.
4. Iteratively, update USPSTF methods to integrate the best evidence and consistently address evidence gaps for Black, Indigenous, and Hispanic/Latino populations. This includes measures to identify and track strategies to demonstrate progress in addressing health inequities regarding clinical preventive services.
5. Use a consistent and transparent approach to communicate gaps in the evidence related to systemic racism in preventive care in recommendation statements and the USPSTF annual report to Congress. This includes an ongoing assessment of how the effects of systemic racism on the quality of the evidence and receipt of clinical preventive services perpetuate health inequities.
6. Collaborate with other guideline-making bodies, professional societies, policy makers, and patient advocacy organizations on efforts to reduce the influence of systemic racism on health.

___________________

²⁹ For example, Black men are at greater risk than White men for lung cancer at lower pack-years of smoking. Therefore, as USPSTF is now recommending lowering the pack-year level to start screening, this may reduce racial inequities in lung cancer health outcomes if implemented with fidelity (Doubeni et al., 2021).

Developments to Improve Data Equity in Federal Agencies

In the last two decades, the federal government has made progress in advancing data equity by issuing EOs and creating interagency working groups to achieve improvements in data design, collection, analysis, and dissemination. For example, each administration, starting in 1999, has issued an EO establishing a commission that encourages greater participation of Asian American and Pacific Islander populations or businesses in federal programs, with improvements in data collections on program use as a key priority.³⁰ Other EOs have also prioritized data collection improvements, including those pertaining to Hispanic American (EOs 13230,³¹ 13555,³² 13935,³³ and 14045³⁴), Black (EOs 13621³⁵ and 14050³⁶), and AIAN (EOs 13336,³⁷ 13592,³⁸ 14049,³⁹ and 14053⁴⁰) people. Most recently, the Biden administration has issued several EOs that address improvements in data collection and dissemination of data to advance racial equity, including 13985 (updated by EO 14091⁴¹ Further Advancing Racial Equity and Support for Underserved Communities Through the Federal Government), 14035⁴² Diversity, Equity, Inclusion, and Accessibility in the Federal Workforce, and 14058⁴³ Transforming Federal Customer Experience and Service Delivery to Rebuild Trust in Government. In addition, the Equitable Data Working Group (established under EO 13985) studies and provides recommendations to the Assistant to the President for Domestic Policy to identify inadequacies “in existing Federal data collection programs, policies, and infrastructure across agencies, and strategies for addressing any deficiencies identified” and “support agencies in implementing actions, consistent with applicable law and privacy interests, that expand and refine the data available to the federal government to measure equity and capture the diversity of the American people” (The White House, 2021).

___________________

³⁰ See, for example, EOs 13125 (Clinton; 64 FR 31105 [June 1999]), 13216 and 13339 (George W. Bush; 66 FR 31373 [June 2001] and 69 FR 28035 [May 2004], respectively), 13515 (Obama; 74 FR 53635 [October 2009]), 13872 (Trump; 84 FR 22321 [May 2019]), and 14031 (Biden; 86 FR 29675 [June 2021]).

³¹ Exec. Order No. 13230, 66 FR 52841 (October 2001).

³² Exec. Order No. 13555, 75 FR 65415 (October 2010).

³³ Exec. Order No. 13935, 85 FR 42683 (July 2020).

³⁴ Exec. Order No. 14045, 86 FR 51581 (September 2021).

³⁵ Exec. Order No. 13621, 77 FR 45471 (August 2012).

³⁶ Exec. Order No. 14050, 86 FR 58551 (October 2021).

³⁷ Exec. Order No. 13336, 69 FR 25293 (May 2004).

³⁸ Exec. Order No. 13592, 76 FR 76603 (December 2011).

³⁹ Exec. Order No. 14049, 86 FR 57313 (October 2021).

⁴⁰ Exec. Order No. 14053, 86 FR 64337 (November 2021).

⁴¹ Exec. Order No. 14091, 88 FR 10825 (February 2023).

⁴² Exec. Order No. 14035, 86 FR 34593 (June 2021).

⁴³ Exec. Order No. 14058, 86 FR 71357 (December 2021).

The working group, cochaired by senior OMB leadership and the Office of Science and Technology Policy, issued a set of recommendations in April 2022 to advance data equity, including making “data disaggregation the norm while protecting privacy,” catalyzing “existing federal infrastructure to leverage underused data,” and building “capacity for robust equity assessment for policymaking and program implementation” (Equitable Data Working Group, 2022, p. 3; The White House, 2022). Some researchers and community organizations have characterized the Biden administration’s actions as a functional approach to data equity (focused on the extent to which data collections advance racially equitable outcomes), in contrast to a process approach (focused on the methods that are used to produce more equitable data collections, including community consultation on data design and development, collection and compilation, processing and analysis, and dissemination and preservation) (AAPI Data and National Council of Asian Pacific Americans, 2022). Given a delay of more than 2 years in the release of detailed-origin population counts from the 2020 census, community organizations have pushed for timely dissemination as an important aspect of data equity (AAPI Data and National Council of Asian Pacific Americans, 2022).

In addition to the standards set by the federal government, community organizations and researchers focused on data equity are pushing for even higher standards that follow on data stewardship principles. In 2016, a wide-ranging group of stakeholders, “representing academia, industry, funding agencies, and scholarly publishers,” published an article in the Nature journal Scientific Data outlining a set of principles on data stewardship known as FAIR (findability, accessibility, interoperability, and reusability) (Wilkinson et al., 2016). These recommendations placed “specific emphasis on enhancing the ability of machines to automatically find and use the data, in addition to supporting its reuse by individuals” and prompted the creation of several FAIR implementation networks to improve data collections (GoFair). In 2020, Data Science Journal published a supplementary set of data stewardship principles to address “concerns about secondary use of data and limited opportunities for benefit-sharing” among Indigenous communities (Carroll et al., 2020). The authors, representing a broad and diverse set of Indigenous data scholars and practitioners, encourage adoption of CARE principles (collective benefit, authority to control, responsibility, and ethics) as well to produce “data that reflect the realities of Indigenous Peoples, be useful for Indigenous purposes, and remain under Indigenous control, while promoting knowledge discovery and innovation” (Carroll et al., 2020, p. 8). These two principles seek to ensure advancements in data equity in not only functional (improving the production and availability of data collections that advance racial equity) but also process (ensuring racial and ethnic equity in the ways that data are collected, stored, analyzed, and disseminated) terms.

Data Gaps and Opportunities Conclusions

CONCLUDING OBSERVATIONS

This chapter provides an overview of the role of government, and the complex relationship between states, examples of health inequities and some of their root causes, and data gaps and opportunities. A common theme is inadequate data in many areas, preventing a full understanding of the extent of health inequities to inform policy actions. However, the chapter also shows the rich information that has been collected that points to the mechanisms for how the social and structural determinants of health have contributed to health inequities and are therefore key areas to focus federal action to advance health equity. Chapters 3–7 take deeper dives into SDOH and how they can positively and negatively impact health equity via federal policies: economic stability, education access and quality, health care access and quality, neighborhood and built environment, and the social and community context.

REFERENCES

This page intentionally left blank.