3

Policies to Improve Clinical Trial and Research Diversity: History and Future Directions

The analysis draws substantially from the research paper by Dr. Laura Bothwell, Ph.D., and Aaron Kesselheim, M.D., J.D., M.P.H, which was commissioned for this study. The full research paper can be found online at: nap.nationalacademies.org

In this chapter, the committee describes major federal policies designed to improve the inclusion of underrepresented and excluded populations in clinical trials and analyze the benefits and limitations of these policies with the aim of improving them. The history of trial diversity policies is deeply embedded in the broader historical context of work toward equity and inclusion. Given the statement of task, the committee decided it was most appropriate to get a scope of what political action has been taken to include more diversity in clinical trials and clinical research. However, the committee would like to acknowledge that there have been many federal policies throughout history that have contributed to racial and ethnic groups being excluded, such as census policies related to race and ethnicity (Pratt et al., 2010). Although out of scope for this report to cover in great detail, the committee deemed it important to acknowledge that federal policies have historically both increased inclusion and exclusion.

EARLY HISTORY

Race and Ethnicity

The Civil Rights Act of 1964 (P.L. 88-352) was perhaps the earliest occasion when legislators or regulators set policies on racial diversity in clinical research. In compliance with the law, in 1965, National Institutes of Health (NIH) General

Clinical Research Centers added new notices to grant applications warning that racial discrimination was illegal. Eventually, all domestic U.S. grant applicants to the Department of Health and Human Services (HHS) had to file with the HHS Office for Civil Rights an assurance of compliance with Title 6 of the act, which prohibits discrimination based on race, color, religion, national origin, or sex in services and establishments that receive federal funding, including hospitals and medical facilities.

Since that time, enforcement of Title 6 has been partial and inconsistent, and racial and ethnic minority populations groups have continued to experience inadequate treatment in clinical care and research at both federal and state levels (Yearby, 2014). National attention was drawn to problems of racism in research in 1972 with the revelation of the 40-year Tuskegee Syphilis Study conducted by the U.S. Public Health Service observing the progression of syphilis among untreated low-income African American men long after treatment had become available (Brandt, 1978). In response, Congress passed the National Research Act of 1974 (P.L. 93-348), which established the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research (Vargesson, 2015). The commission published the Belmont Report: Ethical Principles and Guidelines for the Protection of Human Subjects of Research, which laid the groundwork of principles and guidelines for research involving human subjects, identifying three basic ethical principles for human subject experimentation: respect for persons, beneficence, and justice. The report pointed out that “the selection of research subjects needs to be scrutinized to determine whether some classes (e.g., welfare patients, particular racial and ethnic minority population groups, or persons confined to institutions) are being systematically selected simply because of their easy availability, their compromised position, or their manipulability, rather than for reasons directly related to the problem being studied” (National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, 1979).

Sex and Gender

In the mid-20th century, alongside growing awareness of the value of protecting vulnerable populations, many began to draw attention to a long-held bias in the field of clinical research: the “male norm,” as later summarized in a 1994 report by the Institute of Medicine (1994). Healthy, young, or middle-aged males, frequently who were white, were considered to be the “norm” study population; by contrast, females were thought to confound trial results with their fluctuating hormone levels and reproductive potentials (IOM, 2001; Pinn, 2003). When news broke from Europe and Canada in the early 1960s that widespread maternal exposure to the sedative thalidomide during pregnancy led to fetal death and birth defects, policy makers took the stance that pregnant women were a “vulnerable population” who should be shielded from the potential reproductive adverse effects of drug exposures in trials (Vargesson, 2015). In response to the tragedy, the

U.S. Congress passed the Kefauver-Harris Amendment in 1962 (P.L. 87-781) to strengthen the authority of the Food and Drug Administration (FDA) in overseeing drug development and pre-market evaluation. Some years later, in 1977, the FDA created a guideline, “General Considerations for the Clinical Evaluations of Drugs,” that banned women of childbearing potential from Phase 1 and early Phase 2 trials, except for life-threatening conditions (FDA, 1977). The policy strictly excluded women who used contraception, who were single, or whose husbands had vasectomies (FDA, 1993). In 1979, the Belmont Report further stipulated that pregnant women should be considered vulnerable research subjects and should be protected at all costs (National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, 1979). The 1979 FDA Labeling Rule established the first classification system for identifying the risks prescription drugs posed to pregnant women, fetuses, and breastfeeding infants (FDA, 1979).

Age

The phrase “therapeutic orphan” was coined by Harry Shirkey, M.D., in 1963 to describe the lack of modern drug therapy targeted toward children (Shirkey, 1968). Most authors have attributed this state of affairs to the shortage of relevant drug research in children, as private-sector sponsors deemed the introduction of therapies targeting children to have little potential for profit (MacLeod, 2010). In 1978, the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research published a report on research involving children, discussing the fundamental ethical permissibility of pediatric research, particularly research not benefiting the child involved (NCPHS, 1978b). Prior to this report, philosophers and ethicists held opposing views: some repudiated any ethical justification for research with a healthy child (Ramsey, 1976), while others claimed that even children bear a certain obligation to benefit society, justifying a presumption of their consent to experiments of minimal risk (McCormick, 1974). The commission contended that children might be entered in research entailing more than minimal risk and promising no individual benefit when (1) the risk entailed represents “a minor increase over minimal risk,” (2) the experience presented by the intervention is “reasonably commensurate with those inherent in the actual or expected medical, psychological or social situation” of the subject, and (3) the research is likely to yield generalizable knowledge about the subject’s condition that is “of vital importance for the understanding and amelioration of the condition” from which that class of subjects suffers (Jonsen, 1978). These recommendations were later adopted by the Department of Health and Human Services, including the FDA, in its regulation titled “Additional Protections for Children Involved as Subjects in Research” in 1983 (HHS, 1983).

In addition to creating risk classifications for drugs taken by pregnant women and lactating mothers, the 1979 FDA Labeling Rule sought to improve

the safety and efficacy of drugs intended for diverse ages by requiring labeling content under “Pediatric Use” and “Geriatric Use.” Notably, the rule did not outline specific requirements for risk information provided under the pediatric and geriatric sections as it did for data on pregnant women and lactating mothers (FDA, 1979).

Diversity among Investigators

The Office of Minority Programs (OMP) was established in the NIH Office of the Director in 1990 (NIMHD, 2022). Two years later, the OMP co-funded various projects, including training for faculty and students at all stages along the educational pipeline. It also funded a National Academy of Sciences study that focused on evaluating NIH training programs for underrepresented students (NRC, 2005). The study found that while many NIH programs were helpful in providing students research experience, funding, and mentoring, there was a sharp drop-off among “minority trainees” at the postdoctoral and junior faculty levels. The OMP eventually became the National Center on Minority Health and Health Disparities in 2000, and was redesignated as the National Institute of Minority Health and Health Disparities in 2010.

In addition to the OMP, individual institutes have their own ongoing initiatives. For example, the NIH Office of Diversity and Health Disparities (ODHD) within the National Institute on Drug Abuse was established more than 20 years ago and serves to strengthen a more diverse and robust extramural research workforce, attracting and retaining talented individuals from all backgrounds, and supporting research aimed at the NIH mission of reducing health disparities. Among its numerous endeavors, it provides funding to recruit and support high school, undergraduate, and graduate/clinical students, postdoctorates, and eligible investigators to work on an existing NIH-funded project in a particular area of interest (NIH, 2021d). This opportunity is also available to investigators who are or become disabled and need additional support to accommodate their disability to continue to work on the research project.

MODERN POLICIES

National Institutes of Health

The NIH is responsible for providing direction to research programs with goals to improve the health of the nation, and to that end, the NIH creates policies to improve the nation’s well-being (NIH, 2021a). The NIH is the largest federal sponsor of clinical trials in the United States, devoting about $3 billion per year to funding trials (NIH, 2017c). Its stewardship over clinical trial policies has a substantial impact on the rigor, transparency, and effectiveness of the clinical trial enterprise (Hudson et al., 2016).

The first significant work toward inclusive clinical trial policies at the NIH emerged in response to the 1985 report of the U.S. Public Health Service Task Force on Women’s Health Issues outlining how underrepresentation of women in clinical trials had led to suboptimal women’s health care (Women’s health, 1985). The task force recommended increased participation of women in clinical trials, including women of childbearing potential. It also recommended that research should emphasize diseases that are more prevalent in women (Liu and Dipietro Mager, 2016). In response to this report, the NIH adopted the Inclusion of Women and Minorities in Clinical Research policy in 1986 (NIH, 1987). The major goal of this policy was to ensure that research and clinical trials were designed to provide information about sex and race/ethnicity differences. Response to this policy was slow; guidance for its implementation was not developed until 1989 when a memorandum on inclusion announced that research solicitations should encourage the inclusion of women and minority population groups and stipulated that a rationale should be provided when women and minority population groups were excluded (NIH, 1989).

In 1990, the General Accounting Office (GAO), later known as the Government Accountability Office, a legislative branch agency that provides auditing, evaluation, and investigative services for the U.S. Congress, investigated the NIH implementation of the guidelines for the inclusion of women and minority population groups. In its report, the GAO revealed that the 1986 Inclusion of Women and Minorities in Clinical Research policy had been poorly communicated and inconsistently applied before the 1990 grant review cycle. The GAO identified two major limitations of the policy. The policy only pertained to extramural research conducted by investigators who had been awarded NIH grants, but not intramural research overseen by scientists employed by the federal government. In addition, the policy provided little incentive for researchers to analyze study results by gender (Nadel, 1990). As criticism mounted in response to the GAO report, the Congressional Women’s Caucus took legislative action by passing a package of bills known collectively as the Women’s Health Equity Act of 1990 (S. 2961, 101st Congress (1989–1990)). Responding to this new legislation, the NIH founded the Office of Research on Women’s Health (ORWH) in the same year (P.L. 103-43). The ORWH helped the research community understand the importance of inclusion of women in clinical trials by monitoring and promoting NIH-wide efforts to ensure the representation of women and by prioritizing diseases, disorders, and conditions that primarily affect women. The ORWH also supports initiatives to advance women in biomedical careers and ensures that women are included in clinical research funded by the NIH (P.L. 103-43).

The establishment of these offices and lessons learned from the original inclusion policy contributed to the development of the NIH Revitalization Act of 1993 (P.L. 103-43), which became an updated version of the original inclusion policy but also provided additional guidance on the inclusion and reporting and analysis of sex/gender and racial/ethnic differences in intervention effects for

NIH-defined Phase 3 clinical trials (Night, 2009). The act emphasized that the NIH should ensure that women and minority population groups be included in all clinical research, that Phase 3 clinical trials had sufficient numbers of participants to allow for analysis, that populations were not to be excluded from trials due to cost, and that the NIH must maintain outreach efforts to include women and minority population groups in clinical studies. The law was designed to ensure that clinical research determines whether an intervention differently affects men, women, or members of a minority population (Liu and Dipietro Mager, 2016). Scientists at the time largely supported the act, and it sparked discussions about the importance of appropriate trial design and subsequent subgroup analyses. After implementation of the act, women and minority population groups were increasingly included in clinical trials (Boissel et al., 1995; Freedman et al., 1995). Currently, females make up 49 percent of subjects in NIH-funded clinical trials (Blehar et al., 2013). Under the act, the Office of Minority Programs also changed its name to the Office of Minority Health Research (OMHR). At this point, the OMHR did not have grant-funding authority.

Key gaps remained regarding inclusivity of NIH-funded trials. A study comparing the ethnic distribution of patients enrolled in trials funded by the National Cancer Institute in 2000 through 2002 with those enrolled in 1996 through 1998 found that the proportion of minority trial participants did not change significantly and that the proportion of participants who were Black had declined. After adjusting for age, cancer type, and sex, patients enrolled in 2000 through 2002 were 24 percent less likely to be Black than those enrolled in 1996 through 1998 (Murthy et al., 2004). Ten years into the NIH Revitalization Act’s implementation, another GAO report found that although women were taking part in clinical studies in greater numbers than men and more funding was available for studying diseases that disproportionately affected women, only a small fraction of publications based on NIH-funded research reported findings stratified by sex (Helmuth, 2000). Twenty years post–NIH Revitalization Act, another study concluded that minority population groups remained disproportionally underrepresented in cancer clinical trial enrollments in 2014. In addition to persistent barriers for minority participation in cancer clinical trials, the study reported a dearth of cancer clinical trials that focus primarily on racial/ethnic minority populations, as well as a lack of usable trial data about racial/ethnic minority populations (Chen Jr. et al., 2014). The analysis of NIH-funded trials commissioned by the Committee on Women in Science, Engineering, and Medicine showed that demographic trends in NIH-funded clinical research have not changed much over the years (see Appendix B) and that these trends can vary widely by institute.

In 1998, the NIH Guide for Grants and Contracts published guidelines for including children in research supported by the NIH, unless there were scientific or ethical reasons not to include them (NIH, 1998). The goal of the policy was to obtain appropriate data on treatment outcomes in children. This policy applied to all initial applications/proposals and intramural projects submitted to the

NIH, and it provoked discussions among investigators and ethicists surrounding the ethical dilemma of balancing improving access and recruitment of children in clinical trials with the need to protect this vulnerable population (Glantz, 1998; Kopelman, 2000; Tauer, 2002). The impact of this guideline seemed to lag behind those targeting women and minority populations’ enrollment. A survey was conducted in 2008 to assess NIH Scientific Review Group (SRG) members’ experiences with and attitudes about the NIH inclusion guidelines for women and minority population groups and children, released in 1994 and 1998, respectively. While about half of the SRG members surveyed agreed that the inclusion guidelines resulted in an increase in the number of underrepresented and excluded populations enrolled in clinical research, less than one-third responded that the guidelines expanded the inclusion of children (Taylor, 2008).

In 2000, with the passage of the Minority Health and Health Disparities Research and Education Act (P.L. 106-525), the office became the National Center on Minority Health and Health Disparities (NCMHD). The act gave NCMHD the authority to fund grants and called for the development of a comprehensive NIH strategic research plan and budget for health disparities research. The center was again redesignated as the National Institute on Minority Health and Health Disparities in 2010 with the passage of the Patient Protection and Affordable Care Act (P.L. 111-148), or ACA (NIH, 2010). The office gained authority to plan, review, coordinate, and evaluate the minority health and health disparities research and activities conducted and supported by the NIH institutes and centers (Kneipp et al., 2018).

In 2001, the NIH policy and guidelines on the Inclusion of Women and Minorities as Subjects in Clinical Research from 1986 were updated (NIH, 2001a). The original purpose of the 1986 policy was to ensure the inclusion of women and minority groups in NIH-funded clinical research and that these research findings should be generalizable to a broad population (IOM, 1994). However, the policy lacked a clear definition of clinical research and did not require specific analyses by racial groups to be included when reporting population data. Thus, the updates provided guidance on clarifying the definitions of racial and ethnic categories and reporting analyses of sex and racial minority population groups in clinical trials (Nours, 2021). These updates included the Office of Management and Budget (OMB) Directive’s racial and ethnic categories that are to be used to monitor population data for clinical trials. Though the directive claimed that “the categories in this classification are social political constructs and should not be interpreted as being scientific or anthropological in nature,” scholars argued that standards reflected an important step in moving beyond a simplistic concept of race and its impact on health and provided state and federal public health agencies with an important opportunity to collect, tabulate, and analyze data on program participation and community health that more accurately reflected the racial and ethnic nuances of contemporary American society (Friedman et al., 2000; Hattam, 2005). According to a 2015 GAO report, however, the reporting

of the racial/ethnic composition of study participants did not improve since 2004 (GAO, 2015).

In 2009, the NIH commissioned the Institute of Medicine to conduct a study on the health of lesbian, gay, bisexual, and transgender (LGBT) individuals. The resulting report, The Health of Lesbian, Gay, Bisexual, and Transgender People: Building a Foundation for Better Understanding, concluded that major knowledge gaps exist in the health needs of LGBT people and urged the NIH to support additional research (IOM, 2011). The NIH LGBT Research Coordinating Committee was established to develop and coordinate the NIH’s LGBT research and training, expand knowledge of LGBT health, and improve methods to reach these populations through specific trial networks such as the Adolescent Medicine Trials Network for HIV/AIDS Interventions, HIV Prevention Trials Network, HIV Vaccine Trials Network, and Microbicide Trials Network (NIH, 2015a). In 2015, the Sexual and Gender Minority Research Office was created to coordinate and support sexual and gender minority research activities across the NIH.

To support transgender inclusion, the trial networks adopted a two-step method in data collection forms, separating birth sex and gender identity into two variables (Sausa et al., 2009). In addition, the trial networks updated protocol design with language for transgender inclusion, implemented staff training for cultural sensitivity, consulted with transgender individuals, and conducted new research on transgender individuals (Siskind et al., 2016).

In his 2015 State of the Union address, President Obama announced the Precision Medicine Initiative. In response to this initiative, the NIH created an ethnically diverse research cohort amounting to 1 million or more Americans who had agreed to have their clinical data tracked for research purposes (Collins and Varmus, 2015). This effort was accompanied by workshops hosted by the NIH that examined the reproducibility and transparency of clinical research and aimed to maximize cohort diversity, inclusion, and attention to health disparities (ACD, 2015; NIH, 2015b, 2015c). To further catalyze diversity in research, analysts suggested that the NIH should be empowered to set and enforce recruitment of diverse research populations by race and ethnicity as the default and require scientific justification for limited or selected study population enrollment, similar to what had been created for sex balance (Clayton and Collins, 2014; Oh et al., 2015).

The NIH revisited its policies on age in response to the passage of Section 2038(H) of the 21st Century Cures Act in 2016 (P.L. 114–255). This act instructed the NIH to hold a workshop accounting for differences across the lifespan, publish guidelines addressing consideration of age in clinical research, and ensure that researchers conducting applicable Phase 3 clinical trials report results of analyses by sex/gender at ClinicalTrials.gov (Nours, 2021). As a result of these efforts, the Inclusion Across the Lifespan (IAL) policy was created. This policy required that NIH-funded studies include individuals of all ages (including older adults and children) in clinical trials unless age-based exclusions are scientifically or ethically justified. The policy outlined when certain age groups may be

excluded and noted that grantees are required to annually report on the age at enrollment of their participants along with sex/gender, race, and ethnicity (Nanna et al., 2020; NIH, 2017a, 2017b, 2020a).

The IAL policy is still in its nascent stages, and more data are needed to assess its impact (the policy went into effect in January 2019) (Nanna et al., 2020). Further policy work may also be warranted, as solely extending the age of eligibility for clinical trials is insufficient to make a study truly representative of the general population because social factors such as socioeconomic status may influence access to trials by marginalized groups (Lauer, 2020).

An IAL workshop held in 2017 resulted in several publications related to people with disabilities, including a JAMA article reporting that a present-day review of 338 Phase 3 and 4 NIH-funded actively recruiting studies in ClinicalTrials.gov found that most of the trials did not mention individuals with disabilities in either the inclusion or exclusion criteria (greater than 90 percent did not mention physical disabilities and greater than 80 percent did not mention intellectual disabilities) (Lockett, 2017; Spong and Bianchi, 2018). Explicit exclusion was mentioned in 12.4 percent of the studies for those with intellectual or developmental disabilities (including criteria based on IQ, defined intellectual disability, or cognitive impairment). Explicit exclusion was mentioned in 1.8 percent of studies for those with physical disabilities (including inability to ambulate, extreme immobility, and paraplegia) (Spong and Bianchi, 2018). Further, there are non-explicit barriers to trial participation for people with disabilities, as those with cognitive impairment may be limited by lack of ability to comply with the study protocol or procedures, and individuals with physical disabilities can face limited access to study facilities or face challenges with physiological measurements.

The Sex as a Biological Variable (SABV) policy, which was passed in January 2016, plays an important role in consideration of preclinical research and the design of clinical trials. It established the expectation not only that gender be considered when volunteers sign up for a study but also that investigators balance the proportion of males and females in preclinical investigations from the earliest stages of study design (Arnegard et al., 2020). The policy requires researchers to take sex into account when creating research questions, designing experiments, analyzing data, and reporting results (Nours, 2021). In the 6 years since the NIH enacted SABV, progress has been made (Clayton, 2021). A survey of NIH study section members revealed growing favorability toward the policy, despite some unsupportive perspectives. The number of grant applications that appropriately consider SABV also has increased (Woitowich and Woodruff, 2019).

Regarding diversity among investigators, the NIH Advisory Committee to the Director Working Group on Diversity was formed in 2013 in response to the Working Group on Diversity in the Biomedical Research Workforce (WGDBRW) recommendations. The WGDBRW includes a subgroup on individuals with disabilities that focuses on systematically identifying data, strategies, and experiences

of individuals with disabilities in the scientific workforce to address the multiple barriers they face. In 2017, the WGDBRW established a second subgroup, the Diversity Program Consortium, which supports numerous initiatives designed to build infrastructure leading to diversity, research mentorship for diverse scientists, and awards and resource support (NIH, 2013).

Despite these efforts, the 2021 Women, Minorities, and Persons with Disabilities in Science and Engineering report found that even though the share of science and engineering degrees awarded to underrepresented populations increased over the past decade, several disparities remained. Scientists and engineers with disabilities have an unemployment rate much greater than their peers, and even greater than that of the U.S. general labor force. Female scientists and engineers have lower median salaries than do their male counterparts in most broad occupational groups. Underrepresented populations also hold a small (8.9 percent) share of academic positions, which is considerably lower than their share of the population (NSF, 2021).

Enforcement and Accountability

To ensure the success of NIH inclusion policies, internal monitoring systems include offices, working groups, and committees established across the NIH. An example of a committee used to monitor the progress of NIH policies on the inclusion of underrepresented and excluded populations across the lifespan in clinical research is the Inclusion Governance Committee, which is responsible for monitoring NIH extramural grants and ensuring diversity reporting (Nours, 2021). The NIH also seeks information and advice from the public and hosts workshops that provide researchers with evidence-based approaches in meeting these policies. For example, the Inclusion Across the Lifespan-II workshop provided researchers information about the inclusion of pediatric and older populations in clinical studies in meeting the IAL policy (NIH, 2020a).

Accountability to inclusion policies occurs differently for intramural and extramural clinical research. Intramurally, monitoring for adherence to these policies occurs primarily at the scientific or chief director level. Extramurally, researchers applying for NIH grants must justify their study populations as part of the process to be considered for funding. Extramural researchers must also work with NIH staff to resolve any issues concerning lack of inclusion of certain populations prior to grant approval. Progress reports on a study’s development are monitored by NIH program officers to ensure that all principal investigators meet an acceptable threshold for the number of participants and inclusion criteria in the study’s population. Phase 3 clinical trials are required to report results of sex/gender and race/ethnicity data into ClinicalTrials.gov so that this information can be monitored (Nours, 2021).

The NIH also requires that funded researchers submit a Research Performance Progress Report, or RPPR, annually that asks grantees about their current

accomplishments for the project, upcoming plans, and significant changes regarding human or animal subjects (NIH, 2018a). This information is entered into eRA Commons and is used for accessing and sharing information over the life of a study (NIH, 2016). These progress reports must be approved by NIH for continued funding. NIH then externally reports their inclusion data in a format that is disaggregated by Research, Condition, and Disease Categorization (RCDC) categories. These RCDC data can be found on the NIH Research Portfolio Online Reporting Tools (RePORT) website (Nours, 2021).

Throughout the history of NIH policies, adaptation has been critical, as some policies have not been sufficient to encourage scientists to broaden their study inclusion criteria. Although current policies are encouraging, underrepresented and excluded populations are still underrepresented in clinical trials of some diseases, such as cardiovascular disease, hepatitis, digestive diseases, HIV/AIDS, Alzheimer’s disease, and chronic kidney disease. Improper analyses and disaggregated data in publications exist due to the lack of inclusion in clinical research, impeding the generalizability of scientific findings to the broader population (Nours, 2021). Therefore, further adaptation is needed to adequately diversify clinical trial participation. Investigator bias must also be addressed. In a 2018 study to evaluate compliance with inclusion and assessment of women and racial and ethnic minority population groups in randomized controlled trials, it was found that both male and female researchers perform equally poorly during analysis and reporting of women in clinical studies, and both male and female participants show the same amount of gender bias in decision making (Geller et al., 2018).

Much work also remains to achieve compliance with existing policies. Organizations such as the ORWH, which monitor compliance, are crucial. The ORWH has created resources such as the Inclusion Outreach Toolkit to help principal investigators fulfill their responsibility to conduct inclusive research (Mistretta and Mistretta, 2016; Nours, 2021). Furthermore, the NIH created three free e-learning courses as well as a high-level quarterly publication called Women’s Health In Focus at NIH to raise awareness of the health of women and marginalized populations (Nours, 2021). Strengthening ORWH and other institutional accountability mechanisms could likely improve achieving inclusivity objectives.

Food and Drug Administration

The FDA has been working for decades to ensure that people of different ages, races, ethnic groups, and genders are included in clinical trials. The official stance of the FDA is that clinical trial participants should be representative of the patients who will ultimately use the medical products that the FDA evaluates, because people of different ages, ethnicities, or races can react differently to medical products for a variety of reasons (NIH, 2020b). The agency has primarily promoted

diversity by publishing guidelines that inform sponsors and drug manufactures of the FDA’s current thinking and regulatory interpretations (FDA, 2021a).

In 1985, the FDA introduced “Content and Format of a New Drug Application” (21 CFR 314.50 (d)(5)(v)), its first guidance on analyzing specific subgroups such as pediatric, geriatric, and renal failure patients to evaluate whether dosing modifications were necessary in these populations (FDA, 1985). The inclusion of renal failure patients could be considered early progress for individuals with disabilities. This regulation did not include gender and race as subgroups.

An early breakthrough for gender inclusion came following the work of the first HHS task force on women’s health, established in 1983, which produced a 1985 report on women’s health issues encouraging reexamination of extant policies excluding women of childbearing potential from clinical trial participation (HHS, 1985). The FDA responded with the 1987 publication of a guidance for industry, “Guideline for the Format and Context of the Nonclinical Pharmacology/Toxicology Section of an Application,” which set an expectation that both sexes of animals should be used to provide valuable information in preclinical drug safety studies (FDA, 1987). In the following year, the FDA released “Guidance for the Format and Content of the Clinical and Statistical Section of an Application” in which it recommended analyzing data from clinical pharmacology studies for safety and efficacy by sex, race, and age (FDA, 1988). In addition, the FDA issued a 1989 guidance aimed at drugs used in the elderly that included “Guidelines for the Format and Content of the Clinical and Statistical Sections of an Application” (FDA, 1989). This guideline recommended the analysis of safety and efficacy data to determine the influence of demographic factors such as age and sex in Phase 2 or Phase 3 trials (the final two stages of clinical testing prior to drug approval).

Although the 1988 and 1989 guidance documents aimed to promote evaluation of drug effectiveness based on gender, a landmark GAO report in 1992 concluded that women were nonetheless being underrepresented in clinical trials and trial data were often not analyzed for differences in therapeutic response by sex (GAO, 1992). This report was prompted by a request from Congress based on studies in the medical literature that women tended to metabolize antihypertensive and cardiovascular drugs at a slower rate than men, and that drug interactions with female hormones and use of oral contraceptives could have caused different responses putting women at risk if the FDA approved drugs on the basis of clinical trials in which women were underrepresented (GAO, 1992; Tamargo et al., 2017). The GAO report found that for greater than 60 percent of drug trials, the representation of women in the trial population was less than the proportion of women in the population with the corresponding disease. The GAO concluded that the FDA had not issued adequate guidance for drug manufacturers to determine the extent and sufficiency of female representation in Phase 1 and 2 trials. For example, the FDA did not define the term representative, and drug manufacturers were uncertain of FDA expectations around that term.

While the 1992 GAO report did not evaluate the “appropriateness” of the FDA policy of excluding women of childbearing potential, in 1993 the FDA withdrew its restriction on the participation of women in early clinical trials (GAO, 1992). This retraction was believed to have been prompted by analyses of published clinical trials that showed that trials of aspirin or antianginal drugs had few or no women in them, which made it uncertain how they worked in women (FDA, 1993a; GAO, 1992). In addition, there had been concerns that the 1977 policy may have led to a general lack of participation of women in drug development studies (DiPietro and Liu, 2016). Concerns about the efficacy of drugs in women also arose at a time when the FDA and the scientific community were focusing the need for individualized treatment, and there had been a lack of specific studies of pharmacokinetics in women even when gender-related differences may be expected or important, such as differences due to menopause or the menstrual cycle, or oral contraceptive use, or differences based on body fat percentage, weight, or muscle mass. In addition, the 1977 policy had prevented the gathering of early information on drug response in women that could be used in the design of Phase 2 and 3 trials and may have delayed discovery of gender-based variation in drug effects (FDA, 1993). Earlier participation of women in clinical trials could have led to making appropriate gender-based adjustments in larger studies, such as doses based on weight rather than fixed doses. Still, the FDA did not require that women be included in trials (Wood, 2021). The agency merely stated that it would expect careful, gender-based characterization of drug effects, such as quantifying differences in dose-response and maximum size of effects. The FDA also recommended pharmacokinetic and pharmacodynamics screening in women as a tool to detect differences and analyses of safety and efficacy by sex.

In 1994, the FDA Office of Women’s Health was established to guide the agency on policies for the inclusion of women in clinical trials (HHS, 1994). Within the same year, an Institute of Medicine report, Women and Health Research: Ethical and Legal Issues of Including Women in Clinical Studies, Volume 2, Workshop and Commissioned Papers, called attention to the forms of historical gender bias in the design and implementation of trials (IOM, 1994). Spurred by these concerns, Congress released a 1997 regulation, “FDMA Section 115: Clinical Investigations (b) Women and Minorities Regulation,” that required the FDA and NIH to review and develop guidance on the inclusion of women and minorities in clinical trials.¹ To comply with this regulation, the FDA issued the Demographic Rule in 1998, revising the New Drug Applications (NDA) content to require safety and efficacy data to be presented by gender, age, and racial subgroups and dosage modifications to be identified for specific subgroups (FDA, 1998). This rule gave the FDA the authority to refuse any NDA that did not ana-

___________________

¹ Food and Drug Administration Modernization Act of 1997, S. 830, 105th Congress, November 21, 1997.

lyze safety and efficacy data appropriately. It also required data on participation in Investigational New Drug (IND) applications to be presented by sex, age, and race so that any potential deficiencies in the NDA submission could be identified. In addition, a 1999 FDA guidance recommended the use of population pharmacokinetics to help identify differences in drug safety and efficacy among population subgroups (FDA, 1999). In 2000, Congress passed a law titled Amendment to the Clinical Hold Regulations for Products Intended for Life-Threatening Diseases (21 CFR 312.42) that permitted the FDA to place clinical holds on IND studies if men or women were excluded due to reproductive potential from clinical trials on a serious or life-threatening illness.

Although these policies aimed to increase the inclusion of underrepresented and excluded populations in trials, there were important shortcomings. The 1998 Demographic Rule had the force of law but lacked specificity relative to previous guidance, as it did not include criteria to determine the number of women to be included. The guidance issued in 2000 also did not require the inclusion of any particular number of men or women. A 2001 GAO report that examined these policies found that around one-third of NDAs and 39 percent of IND documents failed to meet the requirements of the 1998 FDA regulation. Although the FDA had the authority to suspend research if women were excluded for their reproductive potential, it had never done so. In addition, the report found that women were only 22 percent of the participants in the small-scale safety trials in which dosage levels were set. There also was no management system to track the inclusion of women in trials or to monitor compliance with existing regulations. The FDA had no criteria to determine whether reviews of NDAs adequately addressed sex differences, and FDA medical officers had not been required to discuss sex differences in their own reviews of NDAs. Thus, the FDA lacked tools to enforce its own regulations and ensure that its reviewers consistently documented sex differences in NDAs (GAO, 2001).

To address some of these limitations, between 2002 and 2005, the FDA issued multiple recommendations for the inclusion and safety of pregnant women in clinical trials. In 2002, an FDA regulation on establishing pregnancy exposure registries provided guidance on monitoring the outcomes of pregnancies exposed to specific medical products with the goal of providing clinically relevant data to medical providers for treating patients who are pregnant (FDA, 2002). In a 2004 guidance, the FDA provided a basic framework for designing and conducting pharmacokinetic and pharmacodynamics studies in pregnant women, and provided instructions on how to assess the influence of pregnancy on pharmacokinetics and pharmacodynamics of medical products (FDA, 2004). A final draft guidance was released in 2018 titled “Pregnant Women: Scientific and Ethical Considerations for Inclusion in Clinical Trials” that supported an informed and balanced approach for gathering data on the use of medical products during pregnancy by encouraging judicious inclusion of pregnant women in trials and careful attention to fetal risk (FDA, 2018).

A push for diversity spurred Congress to pass Section 907 of the FDA Safety and Innovation Act of 2012 (FDASIA) (P.L. 112-144) that directed the FDA to investigate how well demographic subgroups were included in clinical trials and whether subgroup-specific safety and efficacy data were available (FDA, 2012). This law also required the FDA to provide Congress with an action plan that addressed improving the completeness and quality of data analyses on demographic subgroups. To fulfill these directives, the FDA drafted a report, Collection, Analysis, and Availability of Demographic Subgroup Data for FDA-Approved Medical Products, to address the extent to which demographic subgroups participated in clinical trials and whether the relevant subgroup analyses were performed in a manner consistent with FDA regulations (FDA, 2013). The FDA found variability across medical product types in the extent to which demographic data were analyzed. In some applications, subgroup analyses were limited by low sample size. Racial minority population subgroups were often underrepresented in trials. Communication of demographic information to the public also tended to vary for medical devices compared with drugs and biologics due to differences in the FDA regulatory frameworks.

In 2014, the FDA released Action Plan to Enhance the Collection and Availability of Demographic Subgroup Data, as necessitated by Section 907. This action plan outlined recommendations for the inclusion of demographic data in labeling and the public availability of these data. It also included new guidelines to encourage greater demographic subgroup inclusion in trials, plans to work with sponsors to improve information on demographic subgroups in NDAs and INDs, and intentions of strengthening FDA reviewer training by adding education in inclusion, analysis, and communication of clinical data (FDA, 2014b). In another 2014 guidance, the FDA outlined its expectation for sex-specific patient enrollment, analysis of the data, and reporting of the study information with the intention of improving the quality and consistency of data. Through this guidance, the FDA encouraged sponsors to investigate reasons for the lack of enrollment of women and suggested measures to correct this imbalance. For example, if women’s participation dropped substantially after the initial trial screening, then the study criteria may have to be examined to reduce the unintentional exclusion of women. The guidance also provided recommendations to improve enrollment such as targeting investigation sites where women could be more easily recruited, considering alternative communication strategies for recruitment, and maintaining open enrollment for women until a target proportion has been achieved (FDA, 2014a).

Although several FDA regulations in the 1990s addressed the inclusion of women in trials, fewer regulations specifically targeted the inclusion of racial and ethnic minority population groups. When the Office of Women’s Health sought to raise the issue of terminology for race in the 1990s, the FDA initially exempted itself from OMB definitions of race. The Office of Women’s Health raised the issue again in 2004 when it drafted the first guidance around inclusivity on race

that adopted the OMB definition of race and ethnicity for reporting trial populations. This draft was not finalized until 2016 (Wood, 2021).

The Office of Minority Health was only established in 2010 as part of the ACA to advise the FDA on reducing health disparities among racial and ethnic groups (DiPietro and Liu, 2016). In 2016, the FDA released the guidance titled “Collection of Race and Ethnicity Data in Clinical Trials: Guidance for Industry and Food and Drug Administration Staff” to provide instructions on the use of standardized terminology for demographic information (age, sex, gender, race, and ethnicity) based on OMB directives, to ensure that subgroup data was collected consistently (FDA, 2016). This was a very limited guidance that only discussed the terms used to describe “non-white” populations but did not explicitly encourage inclusion of these populations in trials. A guidance for medical devices was released in 2017, “Evaluation and Reporting of Age-, Race-, and Ethnicity-Specific Data in Medical Device Clinical Studies,” in which the FDA provided recommendations for the evaluation and reporting of demographic-specific data in clinical studies (FDA, 2017). The guidance covered why diverse representation was important and identified potential barriers to enrollment, as well as provided recommendations to overcome those barriers.

Around this time, the FDA also began to require the use of Drug Trials Snapshots that provided information about the populations that participated in FDA-supported clinical trials, and highlighted whether there were any differences in benefits and/or side effects by sex, race, ethnicity, and age (FDA, 2020a). Although the use of Snapshots has marked progress in ensuring that demographic information is transparent and made available to the public, it has had limitations. The Snapshots only cover 2015 onward, and only provide information from Phase 3 studies or products that already have been approved (Wood, 2021). A final guidance was issued in 2020 titled “Enhancing the Diversity of Clinical Trial Populations – Eligibility Criteria, Enrollment Practices, and Trial Designs Guidance for Industry,” specifically addressing the need to enhance diversity of clinical trial populations by modifying eligibility criteria, enrollment practices, and trial designs (FDA, 2020c). The guidance advises that drug sponsors have a “plan for inclusion of clinically relevant populations no later than the end of the Phase 2 meeting.” Through this guidance, the FDA satisfied a mandate under Section 601 (a)(3) of the FDA Reauthorization Act of 2017 (FDARA) (P.L. 115-52) to broaden and develop eligibility criteria with no unnecessary exclusions for clinical trials, improve trial recruitment so that trial participants reflect the population that will use the drug, and apply these recommendations to clinical trials (FDA, 2020c).

This guidance also aimed at promoting enrollment of individuals with disabilities in clinical trials. For individuals such as older adults, disabled, or cognitively impaired individuals who need caregiver help or transportation, a requirement to make frequent visits to trial sites can be problematic and hinder participation. To make participation in clinical trials less burdensome, the FDA

has proposed measures such as reducing the frequency of visits, considering whether visits could be replaced by telephone or virtual means, making participants aware of reimbursements for travel and lodging expenses for trial participation, and using mobile medical professionals to visit and evaluate participants or collect blood samples (Stephenson, 2020).

While historically the FDA has often used sex and gender synonymously, this guidance noted that the FDA recognizes that for some clinical trial participants, gender and sex may not be concordant. Still, it stated that discussion of this topic falls outside the scope of the guidance. Sexual minority populations remain largely overlooked by FDA policies (FDA, 2020d).

Enforcement, Incentives, and Accountability

Thus far, the FDA has undertaken various measures to improve diversity in clinical trials primarily via guidance documents and the use of Drug Trials Snapshots. Despite these efforts, certain demographic groups have remained underrepresented in many trials. The measures taken by the FDA have limitations that must be addressed so that populations participating in trials reflect the diversity of the population at large that will be using the drugs/medical products. One such limitation is the lack of enforcement of FDA guidance. Notably, most guidance documents contain the following disclaimer:

The FDA encourages inclusivity but lacks the power to enforce recommendations made in published guidance documents. Although the FDA can make recommendations, it has limited capacity to enforce them. This dynamic may have developed in part because the FDA does not fund investigational drug trials, but rather assesses them, giving the agency less bargaining power compared with the NIH, which can provide funding. However, this is not an insurmountable barrier, especially since the 2000 law gave the FDA the power to put a clinical trial on hold if men or women were being excluded due to reproductive potential. Although the FDA does not appear to have ever put a trial on hold for this reason, such regulations can be leveraged to ensure that sponsors do everything in their power to improve enrollment practices and increase ease of enrollment so that clinical trials can be more inclusive. In addition, FDA also has authority within the IND and Investigational Device Exemption processes to provide feedback to sponsors of clinical trials. The IND process is needed for sponsors to be able to ship the investigational drug across state lines, and involves determining whether the product is reasonably safe for initial use in humans and that the pharmacology

of the compound justifies its use commercially (FDA, 2021a). Given the safety concerns of not including a diverse population in drug testing, this process could be leveraged by the FDA to hold investigators accountable.

FDA progress has also been slow and tends to be reactive to congressional prodding. Developing a guidance is a long and resource-intensive process, and often there has been a lack of bandwidth to produce guidelines to improve inclusivity in trials. Such guidelines have been created in response to laws from Congress, but in the absence of such legal directives there has been limited planning to promote diversity. Changes have tended to arise not from leadership within the agency but from congressional action. The lack of internal initiative and leadership to independently create policies for inclusion has hampered progress toward diversity. In addition, too often if gender or race/ethnicity differences are not proven (e.g., difference in symptoms of a heart attack in women), then they have been treated as though absent (Wood, 2021). A more cautious approach could help to increase concern for the lack of diversity in clinical trials. Finally, the lack of standardization of submission of NDAs and INDs has been problematic. Since every sponsor tends to design trials differently and can label trial characteristics differently in the database (Wood, 2021), there has been no clear standard of submission. Thus, analyzing data and obtaining intersectional data from the FDA has been challenging. Setting guidelines for submission across trials of medical products and drugs could help standardize data collection, which could further aid in analysis. Although the FDA has made major progress in ensuring more inclusivity within clinical trials, it must continue to try to enforce its recommendations to ensure that trial populations reflect the diversity of populations at large.

In addition to enforcement and accountability measures, FDA could do more to incentivize industry to have more diverse clinical trials. This would encourage industry to fill an unmet need in drug development and there is precedent for these policy incentives. In 1983, Congress passed the Orphan Drug Act (ODA) to encourage industry to develop new drugs for rare diseases. The ODA provides tax credits to offset the cost of research and development for these rare diseases, waiving of Prescription Drug User fees, as well as a 7-year extension of market exclusivity for eligible products. This has undoubtedly led to an increase in the number of drugs for rare diseases, with over 800 drug indications approved between 1983 and 2019, compared to only 38 drugs prior to the passage of the ODA (Aitken et al., 2019). An analysis found that the ODA led to “a 69% increase in the annual flow of new clinical trials for drugs for ‘traditional’ long-established rare diseases” (Yin, 2008). Similar incentives could be developed to increase the diversity of clinical trials and clinical research and incentivize developing research infrastructure that includes communities from the outset of the research design.

However, these policies would need to be developed carefully to minimize costs for the public and patients, with an observation for lessons learned from the ODA. While the ODA has done a great deal for rare diseases, there are

growing concerns about its costs. As the number of treatments for orphan drugs increases, there are increased concerns about the high costs of orphan drugs, their potential to threaten insurance premium levels, and their accessibility for many patients due to their high costs (Pearson et al., 2022). In addition, a 2018 GAO report found inconsistent and incomplete reviews in the process of designating medicines as orphan drugs, indicating that drug companies may be overusing the classification to maximize profits (GAO, 2018b).

The committee feels that some of these incentives could still be implemented with some guardrails in place that may minimize some of the downsides of the ODA. For example, a recent white paper analyzed some policy options to prevent some of these costs, including establishing a maximum revenue threshold to be eligible for ODA incentives, using sliding scale bonuses or refunds depending on outcomes for a particular drug, using volume-based contracts to purchase large volumes of drugs for rare conditions, and more (Pearson et al., 2022). These provisions could be considered and written into any potential new legislation to minimize the risk of harm while still providing incentives to private industry to diversify clinical trials and clinical research.

Centers for Disease Control and Prevention

The Centers for Disease Control and Prevention (CDC) sponsors a wide range of public health initiatives, including providing funding for local and state public health agencies. The research mission of the CDC is to support public health studies, and it often sponsors retrospective examinations of public health issues. Funding clinical trials makes up a small portion of the CDC’s research budget, but the agency still regulates recruitment of diverse demographics.

CDC policies promoting inclusion of diverse populations in research have been largely catalyzed by federal laws. The 1993 NIH Revitalization Act did not govern the CDC, but the legislation spurred the CDC to create its own policies on inclusion of women and diverse races and ethnicities in research participation (Geller et al., 2011). In 1995, the CDC issued the “Policy on the Inclusion of Women and Racial and Ethnic Minorities in Externally Awarded Research,” applying to extramural research activities (CDC, 1995). In 1996, the CDC released “Inclusion of Women and Racial and Ethnic Minorities in Research,” which applied to intramural research (CDC, 1996). The ordinances stated that women and members of racial and ethnic minority population groups should be adequately represented in all CDC research involving human subjects, in the absence of a compelling reason for exclusion. They further stipulated that women of childbearing potential should not be routinely excluded from research without proper cause. The policies provided guidance on how these goals for inclusion could be met by investigators. For example, if all diverse groups could not be included in a single study, multiple studies could be conducted. The policies also stated that it was not necessary to provide the statistical power to test hypotheses in all

groups separately, but that if differences between groups are plausible, this should be tested in the study design. Further, study proposals should include discussion of the inclusion or exclusion of minority groups (CDC, 1995, 1996).

The CDC policy on inclusion of women and ethnic and racial minority populations was closely followed by the more general policy in 1997, “CDC Procedures for Protection of Human Research Participants” (CDC, 1997). This new policy restated an abridged version of the 1996 ordinance, placing it into context of protections of other policies for the protection of human subjects.

The CDC revised the 1996 “Inclusion of Women and Racial and Ethnic Minorities in Research” policy in 2010. This policy united the past separate policies on extramural and intramural research into a single policy. The revised policy strengthened the call for representation in clinical research by stating that direct efforts should be made to actively recruit and enroll women and minority population groups in all funded research (CDC, 2010).

Protections for children in clinical research were codified in the 1997 “CDC Procedures for Protection of Human Research Participants.” The 1997 guidelines deferred to local and state laws on medical consent for minors to determine whether it is appropriate for a minor to participate in a research study. However, the policy also stated that the minimum requirements for consent in a local jurisdiction do not necessarily authorize a minor’s involvement in a research study. The policy gives latitude to the study’s institutional review board (IRB) to determine the ethical parameters for a minor’s involvement in research. The IRB should weigh risks to the children in the study with the benefits the research may provide to children as a group. The ordinance also stresses the importance that research poses a “minimal risk” to children, as adjudicated by the IRB. It outlines minimal risk research activities that are usually acceptable for children, such as urinalysis, venipuncture, electroencephalography, and allergy scratch tests (CDC, 1997).

The CDC issued an explicit policy on children in research in 2006 (updated in 2011) titled “Inclusion of Persons under the Age of 21 in Research.” According to this policy, research proposals must include a rationale to include or exclude persons under 21 in intramural or extramural research—similar to the provisions in the “Inclusion of Women and Racial and Ethnic Minorities in Research” policy (CDC, 2011).

While the CDC’s public health research and initiatives have received a plethora of academic attention, there has been less research on the clinical trial policies of the CDC. The CDC devotes a smaller proportion of its funding to clinical trials, but further attention to this issue may be useful.

Centers for Medicare & Medicaid Services

In addition to the CDC, other federal agencies that sponsor or regulate clinical trials include the Department of Defense, the Agency for Healthcare Research

and Quality (AHRQ), and the Centers for Medicare & Medicaid Services (CMS) (HHS, 2017). The role that CMS plays in funding diverse clinical trials is especially important. While costs of routine care to participants in trials are usually covered by Medicare, many costs are borne by the participants, decreasing participation especially among financially disadvantaged patients (Medicare advantage, 2009). For participants receiving Medicaid benefits, there were historically no federal mandates for clinical trial coverage, making it prohibitively expensive for many Medicaid beneficiaries to participate in clinical trials (Winkfield et al., 2018).

This changed in 2000, when President Clinton issued an executive memorandum directing the secretary of HHS to cover the routine patient care costs associated with clinical trials, as well as costs due to any medical complications (The White House, 2000). The Health Care Financing Administration (the predecessor to CMS) responded to the executive order with a policy specifying that Medicare would cover “routine” costs that accompany clinical trial participation, including diagnostic tests, hospital charges, and provider fees, but excluding reimbursement for “items and services provided solely to satisfy data collection” (CMS, 2000). The exception is thought to have created major barriers that prevented community providers from participating in clinical trial enrollment, which in turn disproportionately affected racial and ethnic minority population groups. CMS updated its clinical trial policy in July 2007 with clarifications and some additional coverage items. However, the policy still excludes items and services for data collection (CMS, 2007).

In 2014, CMS released an updated guidance that allows CMS to determine coverage of an item or service only in the context of a clinical study. In its coverage with evidence development for transcatheter mitral valve repair, for example, CMS explicitly noted, “study protocol must explicitly discuss subpopulations affected by the treatment under investigation, particularly traditionally underrepresented groups in clinical studies, how the inclusion and exclusion criteria affect enrollment of these populations, and a plan for the retention and reporting of said populations on the trial” (CMS, 2014b). Such requirements by CMS have the potential to change the landscape of clinical trial representation because millions of Americans served by Medicare who have been traditionally underrepresented in clinical trials could now be recognized and actively recruited. Additionally, the $2.3 trillion omnibus spending and relief package passed by Congress in 2020 guaranteed, for the first time, routine costs for clinical trials for Medicaid recipients by 2022, expanding access for many low-income participants (Takvorian et al., 2021). Still, transportation, time away from work, and other ancillary costs remain barriers that will need to be addressed for participants despite their coverage status. In addition, the committee is not aware of any studies that specifically examine the impact that these reimbursement policies have had on accessibility for participation in clinical trials, the extent that they are utilized by participants, and any barriers that remain for participants accessing these coverage options.

Finally, while the updated version of this policy mandated that peer review of study protocols should assess “the adequacy of plans to include both genders, minorities, children, and special populations as appropriate for the scientific goals of the research,” it omitted sexual and gender minority populations (NIH, 2019b).

CMS and the Center for Medicare and Medicaid Innovation also conduct demonstration projects, which are pragmatic clinical studies testing the effectiveness of different strategies and financial/reimbursement incentives for quality or outcome improvement. These studies involve data collection across sites, and achieving diverse enrollment has been a priority.

Agency for Healthcare Research and Quality

The AHRQ is an agency within HHS that has a mission to improve the safety and quality of America’s health-care system. The Healthcare Research and Quality Act of 1999 (P.L. 106-129) established the Office of Priority Populations within AHRQ. Subsequently, the AHRQ has required that all AHRQ-supported research includes priority populations unless a compelling justification is provided against inclusion. Priority populations initially included women, children, and racial and ethnic minority population groups; populations with special healthcare needs (chronic illness, disabilities, and end-of-life care needs); and elderly, low-income, inner-city, and rural populations (AHRQ, 2021).

In 2021, President Biden signed Executive Order 13985, titled “Advancing Racial Equity and Support for Underserved Communities through the Federal Government” and defined underserved communities as individuals who have been denied “consistent and systematic fair, just, and impartial treatment.” Specifically, it identified “Black, Latino, and Indigenous and Native American persons; Asian Americans and Pacific Islanders and other persons of color; members of religious minorities; lesbian, gay, bisexual, transgender, and queer (LGBTQ+) persons; persons with disabilities; persons who live in rural areas; and persons otherwise adversely affected by persistent poverty or inequality” as underserved communities (Executive Office of the President, 2021). Subsequently, the AHRQ updated its Policy on the Inclusion of Priority Populations in Research (NOT-HS-21-015) to expand its definition of priority populations to match those groups identified in Executive Order 13985 (AHRQ, 2021).

The Patient-Centered Outcomes Research Institute (PCORI) is a U.S.-based nonprofit institute created through the ACA (IRS, n.d.). Its funding comes from the Patient-Centered Outcomes Research Trust Fund (PCORTF) authorized by Congress in 2010 under the ACA, and reauthorized again in 2020 under the Further Consolidated Appropriations Act (P.L. 116-94). With PCORTF funding, in 2014, PCORI launched PCORnet, a national patient-centered clinical research network (Fleurence et al., 2014). PCORnet published “Diversity and Inclusion in PCORnet: Need and Recommendations” to set guiding principles for diversity and inclusion in PCORnet. The guidance called for inclusion and prioritizing

underrepresented groups affected by the outcomes of research, including “people of color, rural/inner-city populations, pregnant and lactating women, gender and sexual minorities, individuals with disabilities, and other audiences commonly underrepresented in clinical research” (PCORI, n.d.). PCORI has also undertaken numerous measures recently to expand its work toward diversity, equity, and inclusion. The agency created an internal steering committee that is developing a comprehensive action plan to enhance diversity, equity, and inclusion. PCORI is also developing a data collection strategic framework with attention to diversity and inclusion.

The Common Rule

Many federal agencies are subject to the Common Rule (45 CFR 46), most recently revised in 2018 (HHS, 2017). The Common Rule is the blanket federal policy for the protection of human subjects that IRBs are expected to follow. The Common Rule requires that selection of research subjects be equitable, but it does not further specify inclusive recruitment of diverse subpopulations. Rather, the rule states that IRBs should be particularly cognizant of the special problems of research with subjects vulnerable to coercion or undue influence, such as children, prisoners, individuals with impaired decision-making capacity, or economically or educationally disadvantaged persons, and that additional safeguards should be included in studies to protect the rights and welfare of these subjects. The rule requires that IRBs should be composed of diverse individuals, and if an IRB regularly reviews research that involves a category of subjects that is vulnerable to coercion or undue influence, the IRB should consider including one or more members knowledgeable about and experienced in working with these categories of subjects. The Common Rule also specifies particular ethical regulations for children and pregnant women. One notable change in the 2018 revision was the removal of pregnant women from the “vulnerable” category of research subjects. This was in response to criticism that women were being unfairly excluded from research studies, to the detriment of designing treatments for women (NIH, 2019a). The revision aimed to increase the participation of women in research studies and to improve the recommendations for prescribing interventions for pregnant women (Hurley, 2017).

Enforcement and Accountability of CDC, CMS, AHRQ, and Common Rule

U.S. government agency policies on inclusion are enforced through different means. For example, CMS has an expectation that all supported clinical studies demonstrate adherence to inclusivity requirements, and that the agency would not anticipate approving a study that does not meet the requirements (CMS, 2014). The AHRQ peer review regulation requires that reviewers of grant and contract

applications include their assessment of the proposed inclusion plan for priority populations in evaluating the overall scientific and technical merits of applications. Similarly, the CDC requires that grant reviewers abide by CDC policies on inclusion and diversity. Beyond grant review, accountability to inclusion policies also comes from a research applicant’s IRB and oversight from HHS’s Office for Human Research Protections (OHRP). The Common Rule grants latitude to IRBs to review research projects under the Common Rule’s requirement that subject recruitment is equitable (HHS, 2017). However, the extent to which this is actually accomplished by IRBs in practice can depend upon the individual IRB’s commitment to and interpretation of this objective. While OHRP does not directly oversee compliance of individual research studies, it does issue written assurances of compliance to research institutions, such as universities and academic medical centers, which allow investigators to perform research within the institutions. If an institution is noncompliant with the Common Rule, OHRP can revoke its assurance.

SPECIAL POPULATIONS

Including special populations in research is critical for addressing research questions. According to a 2018 review, “when special populations have been included into clinical trials, numerous age-dependent, community, cultural and genetic features have come to light.” However, including these populations on clinical research requires consideration and use of best practices, such as building trust, conducting clinical trials that are relevant to special population, providing incentives and compensation, as well as offering options for participants to easily opt-out of the research (Winter et al., 2018).

Who are special populations? “It was not until the establishment of NIAAA [National Institute on Alcohol Abuse and Alcoholism] in the early 1970s, that using the concept of special populations as a categorizing strategy (both for funding and for the development of treatment programs) became prominent” (IOM, 1990). There is no standard definition of special populations; depending on where the term is used—in social work, education, medicine, criminal justice, or human services—the groups included will be defined by different terms.

Sovereign Nations

An unknown, legal, complex, and unique relationship exists between the United States and tribal nations. Hundreds of treaties, the Supreme Court, the President, Congress, executive orders, and laws have created a fundamental

contract between 574 sovereign tribal nations and the United States (Federal Register, Vol. 86, No. 18), which recognize tribal nations as sovereign “domestic dependent nations under the protection” of the United States, and as sovereign nations, “exercise inherent powers over their members and territory” (Executive Order 13084, May 14, 1998). It is important to understand sovereignty and how it will play a role in research. In 2004, Kalt and Singer wrote Myths and Realities of Tribal Sovereignty: The Law and Economics of Indian Self-Rule, wherein the effects of tribal sovereignty reflected a three-decade resurgence in Indian country (Kalt and Singer, 2004). What happened 30 years earlier that engendered the resurgence? The passage of the Indian Self-Determination and Education Assistance Act of 1975 (P.L. 93-638) provides for maximum Indian participation in the government and education of the Indian people; provides for the full participation of Indian tribes in programs and services conducted by the federal government for Indians, and encourages the development of human resources of the Indian people; establishes a program of assistance to upgrade Indian education; supports the right of Indian citizens to control their own educational activities; and for other purposes. The act also provides tribal nations with the opportunity (other purposes) to manage health-care services in their local area, providing medical, dental, and behavioral health-care needs to improve the health and well-being of tribal members.

Indian Health Services

Specific challenges exist when trying to recruit American Indians related to both their physical locations and the structure of the Indian Health Service (IHS). Greater than 75 percent of American Indians live in urban areas, while only 1 percent of IHS funds are allocated to clinics in these areas; opportunities to interact with possible researchers is thus significantly limited (HHS, 2022a). Additionally, most studies require IHS and/or tribal approval, in addition to standard IRB approval, requiring additional care and consideration by investigators for navigating this system (Giuliano et al., 2000).

Millions of acres of tribal lands have been given up to the federal government; nearly all the land was acquired via treaty or agreement with tribal nations. In return, the federal government promised to provide health, education, and general welfare for reservation residents. These promises are known as “the United States trust responsibility to all Indians.” It is the federal government’s responsibility to protect tribal treaty rights, lands, assets, and resources, as well as carry out the mandates of federal law concerning tribal nations (112th Congress, 2nd Session). In April 2020, the National Tribal Budget Formulation Workgroup requested an “adequate level of funding” for the IHS Fiscal Year 2022 Budget. The President’s Fiscal Year 2022 Budget includes an increase of $2.2 billion dollars in discretionary funding for IHS, or 36 percent above FY 2021, which is the largest single-year funding increase for IHS in decades (IHS, 2021). Prevent-

able and treatable diseases, lack of basic health system infrastructure, and past failed policies all have contributed to avoidable health disparities, decreased life expectancies (see Table 3-1), and maintains impoverished health conditions for reservations and tribal members (NIHB, 2020). The IHS is the federal agency that oversees and provides health care to tribal communities through Indian tribes, tribal organizations, and urban Indian organizations.

Before COVID-19, the IHS was already so underfunded that expenditures per patient were just one-fourth of the amount spent in the Department of Veteran’s Affairs health-care system and one-sixth of what is spent for Medicare. IHS facilities are, on average, understaffed by 25 percent (GAO, 2018a). Direct care, or medical and dental care that American Indians and Alaska Natives receive at an IHS or tribal medical facility, is covered through health benefits from the IHS. When a patient requires care that is not available at the IHS or tribal clinic, purchased referred care (PRC) funds are used, approval of which depends on several factors, including confirmation of tribal affiliation, medical priority, and funding availability. However, IHS remains severely underfunded, which contributes to its inability to meet its mission of raising the health status of American Indian/Alaska Native people.

During FY 2019, the Oklahoma City Area Indian Health Service (OCA IHS) was an example of severe underfunding. The per-person PRC funding level was $311.20. This led to OCA IHS having to operate at a Priority I service care level, also known as life-or-limb service (NIHB, 2020). To qualify for PRC care, a patient must be in peril of losing either their life or a limb. Additionally, there are 42 Urban Indian Organizations (UIOs) for health care in the United States providing care to the 78 percent of American Indians/Alaska Natives who are not living on a reservation either permanently or temporarily. These UIOs receive less than 1 percent of the IHS budget, which is currently underfunded at less than 50 percent of need (NCUIH, 2019). The reauthorization extension of the Indian Health Care Improvement Act (IHCIA) (P.L. 94-437) was passed in 2010 as part of the ACA. However, after 4 years, provisions of the act continued to be unfunded (NCAI, 2016). It is not difficult to understand why tribal members experience significant health disparities.

Due to the factors listed above and centuries of mistreatment, mistrust among American Indians and Alaska Natives has grown. To reach these communities, it

TABLE 3-1 2020 Life Expectancy at Birth

SOURCE: U.S. Department of Health and Human Services, Office of Minority Health, https://minorityhealth.hhs.gov/omh/browse.aspx?lvl=3&lvlid=62.

is critical that researchers involving tribal communities understand tribal sovereignty, as it is a fundamental root of a tribe.

As sovereign nations tribes have legal rights and privileges not afforded to other groups, tribes can regulate research involving tribal members and the use and ownership of their data. What is or is not agreed upon with one tribe will not necessarily transfer to another tribe. And, there are no tribal- or federal-wide agreements for the inclusion of or data protections for tribal members who individually enroll in clinical studies not specific to tribes or tribal nations, such as the 19,806 American Indian/Alaska Native who participated in NIH-funded research in 2017. There is little information available to educate investigators about the concerns surrounding tribal participation and therefore little to guide them on how to approach inclusion of Indigenous individuals in clinical studies that are not specific to these groups and their Native nations (Kalt and Singer, 2004).

Therefore, regardless of the difficulty, it is critical that research, education, and outreach in Indian Country continue to be brought to the tribal nations.

Winter et al. (2018, p. 58), in the section entitled “History, Context and the Ephemeral Nature of Trust,” describe the importance of understanding the history and context for special populations to anticipate behaviors and attitudes from research participants. This applies to any culture or population throughout history.

___________________

² Frank Charles Dukepoo was an acclaimed geneticist at Northern Arizona University and the first Hopi to earn a Ph.D.

This page intentionally left blank.