Interventions to increase racial and ethnic minority accrual into cancer clinical trials: A systematic review

Abstract

BACKGROUND:

Racial and ethnic minorities (REMs) continue to be underrepresented in clinical trials despite the 1993 National Institutes of Health’s Revitalization Act mandating the representation of women and underrepresented minority groups in clinical trials. Although Blacks represent 15% and Hispanics 13% of the cancer population, their clinical trial enrollment rates are disproportionately low at 4% to 6% and 3% to 6%, respectively. A systematic review exploring interventions aimed at improving cancer clinical trial (CCT) enrollment for REMs was conducted.

METHODS:

A systematic search of PubMed, Cochrane CENTRAL, and Ovid PsycINFO was conducted for English-language studies since 1993. Inclusion criteria included peer-reviewed, US-based studies with interventions aimed to recruit underrepresented minority adult patients into cancer clinical trials. REM groups were defined as Black, Hispanic, Asian, American Indian, and Native Hawaiian/other Pacific Islander.

RESULTS:

The systematic search identified 3123 studies, of which nine met inclusion criteria. Interventions included patient navigation/coaching (n = 4), a clinical trial educational video (n = 2), institutional research infrastructure changes (n = 1), a relationship building and social marketing recruitment model (n = 1), and cultural competency training for providers (n = 1). A statistically significant improvement in accrual was shown in three of the patient navigation interventions, one of the clinical trial educational videos and an institutional research infrastructure change.

CONCLUSIONS:

This systematic review illustrates several potential mechanisms by which to increase CCT recruitment for patients of REM backgrounds in various clinical settings More randomized controlled trials are needed to further explore. the benefits of these interventions for REMs.

INTRODUCTION

Despite breakthrough advancements in cancer treatment over the years, most studies leading to these advancements have not had adequate racial and ethnic minority (REM) representation. We are therefore less confident in how existing evidence-based treatments affect REM patients, which the US Census Bureau projects will encompass more than one-half of the US population by 2045. In 1993, the National Institutes of Health (NIH) passed the Revitalization Act to mandate the representation of women and REMs in clinical trials. In many clinical trials, as well as those specific to cancer, REM patients have been underrepresented. Although Blacks represent 15% and Latinos 13% of the cancer population, their clinical trial enrollment rates are disproportionately low at 4% to 6% and 3% to 6%, respectively, and this was noted to have declined since the NIH Revitalization Act.^1,2 For the purposes of this article, we will use the terms Black for African American/Blacks, Latino for Hispanic/Latino, and American Indians for Native Americans/American Indians/Alaska Natives without prejudice and the encompassing term REM in the United States for all aforementioned and Asian and Native Hawaiian/Other Pacific Islander. Furthermore, we will use the same terminology as the reviewed articles while respecting that the language around delineating various REM groups has evolved to become more accurately representative.

Patient-level barriers to clinical trial enrollment have been extensively studied and include mistrust, lack of access to information, cost, lack of transportation, and childcare.^3–8 More recent studies have found no racial/ethnic differences in willingness to enroll once patients are approached about clinical trials.^9–13 There has been substantial research understanding barriers and facilitators to enrollment. Physicians offer clinical trials less frequently to REMs because they fear losing trust, perceive that patients with low health literacy will have a poor understanding of the trial and consent process, and because of the perceived increased time to explain the trial process.^14–16 On a structural and systemic level, the lower availability of trials and supporting resources are critical challenges. This is especially true in underresourced institutions that often care for a higher number of REM patients.¹⁵

Several studies have evaluated perceptions of and intention to enroll in cancer clinical trials (CCTs), but few have directly assessed actual enrollment of REMs.^17–21 Studies that have assessed the latter include patients both with and without cancer, such as trials evaluating enrollment in cancer screening. In this systematic review, our aim was to identify and characterize mechanisms and strategies that could serve as interventions to improve representation of REM patients in CCTs. Despite the attention on this topic over the years, this systematic review is the only one to date evaluating interventions that examine actual accrual of REM patients with cancer into CCTs.

It is important to mention that socially defined races are not accurate reflections of biologic variation: within one race, there is significant variability in ancestry and genetics and that environmental pressures such as poor air quality, food insecurity, and access to care all play a role in the development of disease. Speaking to diversity in ancestry within a race, Deyrup et al. highlight how the average European ancestry in Blacks living in the United States is 16%; however, this can be as high as 30% in some states.^22,23 Because scientific research has thus far categorized diverse populations into socially defined races, we have used this designation in our review. We, however, believe that science must evolve from socially defined race categories and toward more accurate designations of biologic variation.

MATERIALS AND METHODS

We sought studies that evaluated interventions specifically aimed at enrolling REM patients with cancer into CCTs. Data were collected about race and ethnicity of participants and the types, duration, and location of interventions. Data are reported according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines.²⁴

Search strategy

Three databases were selected: PubMed (via National Library of Medicine’s PubMed.gov), Cochrane Central Register of Controlled Trials/Cochrane CENTRAL (via Wiley’s Cochrane Library), and PsycINFO (via Ovid). The search strategy was developed in PubMed by a research informationist (K.G.) in collaboration with the research team using Medical Subject Headings and keywords for four concepts: minorities, accrual, clinical trials, and cancer. Concepts were combined with the Boolean AND operator, and restrictions were added for English language and publication dates after the NIH Revitalization Act of 1993, which mandated the inclusion of women and minorities in clinical trials. The Cochrane Handbook filter was used to exclude animal-only studies.²⁵ A second research informationist performed a Peer Review of Electronic Search Strategies, and edits were implemented.²⁶ The search strategy was then translated to the other databases using available filters and controlled vocabulary (Supplement A). The three databases were comprehensively searched on May 21, 2020. Results were entered as RIS files in Covidence, a web-based software platform for systematic review development that includes the deduplication of uploaded records.²⁷

Inclusion criteria

Inclusion criteria for the review consisted of the following studies: (1) published empirical, peer reviewed research studies (i.e., no commentaries, editorials, conference abstracts, or dissertations); (2) published since 1993 (year of the NIH Revitalization Act); (3) conducted in the United States; (4) focused on improving accrual for CCTs targeting patients with cancer; (5) focused on adult populations (at least 18 years of age); (6) focused on improving accrual in REM; and (7) published in English. The rationale for conducting our systematic review is to provide a landscape evaluation of intervention trials complete with study details that are not provided by abstracts. Abstracts also do not provide sufficient information to evaluate the quality of a study. For these reasons, abstracts have not been included in our review.

Selection strategy

Using Covidence, study eligibility was determined by two investigators independently (R.N., C.M.) based on the manuscript titles and abstracts. A full text review was conducted by the four investigators on the study team (R.N., C.M., P.P., R.C.M.) (Fig. 1).

Figure 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses chart.

Data abstraction

An iterative process was used to create a standardized procedure to extract information from the eligible studies. During the full-text review phase, articles were reviewed twice using Covidence. For data extraction, a standardized data abstraction form was developed. This involved a primary reviewer, who completed the data abstraction form, and a secondary reviewer, who checked for accuracy and completeness. Then, the four reviewers resolved conflicts by using an iterative process and discussion to reach consensus. Data captured on the abstraction forms included first author, title, year of publication, study type, study setting, intervention type, length of intervention, number of study participants, percentage of underrepresented patient population, type of cancer, clinical trial type, primary and secondary end points, and statistical outcome.

Grading of the selected studies

After data extraction, two reviewers performed an independent assessment of the study’s quality by using a quality-assessment form based on the Downs and Black (1998) checklist for the assessment of the methodological quality of both randomized and nonrandomized studies of health care interventions.²⁸ This checklist was modified to more fully evaluate the types of studies included in this review (Supplement B). Study quality was evaluated based on three domains: quality of reporting (clear and detailed description of aims, outcomes, intervention and findings), external validity (representativeness), and internal validity (confounding and bias). The same procedure detailed for the data extraction phase was followed for the appraisal. Unresolved disagreements between reviewers were adjudicated in team meetings among the four reviewers. For each question, studies that met the criteria were assigned 1 point and those that did not were assigned 0 points. When a question was nonapplicable, it was assigned “N/A” and a point was removed from the tallying denominator. A score of 80% or higher was deemed as “higher quality” and lower than 80% as “lower quality.”

RESULTS

The systematic search identified 3123 studies, of which nine met inclusion criteria (see Fig. 1 for study selection). The most common reason for exclusion was the lack of focus on improving patients with cancer’s accrual into CCTs. Many studies instead focused on willingness, barriers, and facilitators to enrollment. Given that this is an emerging area of research, we decided to conduct a systematic review to characterize the landscape on interventions aimed at improving CCT accrual.

Study characteristics

Table 1 describes the study characteristics and results. All studies were prospective interventional studies to increase REM enrollment into CCTs, and the majority took place in hospital-based cancer centers (National Cancer Institute [NCI]-designated institutions [n = 4], non–NCI-designated cancer centers [n = 1], private practice settings [n = 1], and a mixture of settings [n = 3]). Many took place in urban settings (n = 4) and some in rural (n = 1), suburban (n = 1), and a mixture of settings (n = 3). The types of CCTs recruited into various types and included therapeutic trials (n = 3), a behavioral trial (n = 1), both therapeutic and nontherapeutic (n = 3), and some unspecified types of CCTs (n = 2). Three studies focused only on patients with breast cancer and the rest included a mix of cancer types. Some studies focused solely on Black patients (n = 4), one a homogeneous American Indian population (n = 1), and the rest a mix of various races and ethnicities (n = 4). Only two of the studies were randomized control trials (RCTs) and the rest were prospective non-RCTs. The comparison groups in the non-RCTs were the institutions’ prior accrual numbers, national averages, or those cited in the literature. The various study interventions included patient navigation/coaching (n = 4), clinical trial information videos (n = 2), institutional structural change (n = 1), a relationship building and social marketing approach (n = 1), and cultural competency training (n = 1).

TABLE 1.

Study characteristics

Study	Study characteristics	Participant characteristics	End points	Outcome
Guadagnolo B.A., 2011	Study type: Single arm Study setting: Rural cancer institute Intervention: Patient navigation Length: 5 y	N = 332 Cancer type: Mixed Demographics: 100% Native American Clinical trial type: Therapeutic and nontherapeutic	Primary: (1) Patient navigation utilization (2) Potential impacts on clinical care (treatment interruptions and clinical trial enrollment)	22% recruited (compared with the average in the literature of <1%)
Heiney S.P., 2010	Study type: Single arm Study setting: Urban/suburban mix, mixed health care settings Intervention: Relationship-building and social marketing recruitment model Length: 2 y	N = 133 Cancer type: Breast Demographics: 100% Black Clinical trial type: Nontherapeutic (behavioral)	Primary: Clinical trials recruitment	66% recruited (no comparison group)
Holmes D.R., 2012	Study type: Single arm Study setting: Suburban community physician offices Intervention: Nurse navigator Length: 2 y	N = 59 Cancer type: Breast Demographics: 100% Black Clinical trial type: Not specified	Primary: Clinical trials recruitment	86% recruited institutionally. Recruitment rates of Black patients to clinical trials increased from 3% to 7%
Robinson B.N., 2017	Study type: Single arm Study setting: Urban/suburban mix, hospitals, and cancer institutes Intervention: 15-min culturally tailored enrollment barriers video Length: 1.5 y	N = 200 Cancer type: Breast Demographics: 100% Black Clinical trial type: Therapeutic	Primary: Clinical trials recruitment Secondary: Intervention’s influence on population’s (1) intentions to participate in a clinical trial and (2) attitudes toward clinical trials	13.5% of sample recruited to trial, 7.5% increase in recruitment from the institution’s 2012 baseline of 6%
Skinner J.S., 2019	Study type: Randomized control trial Study setting: Urban NCI comprehensive cancer center Intervention: 10-min clinical trial educational video Length: 1 y	N = 63 Cancer type: Mixed Demographics: 51.6% White, 34.4% Black, 4.7% Asian, 4.7% American Indian/Alaskan Native, Other 4.7% Clinical trial type: Not specified	Primary: Clinical trials recruitment	Nonsignificant difference in recruitment
Wells J.S., 2017	Study type: Single arm Study setting: Urban/suburban mix, mixed health care settings Intervention: 4-h cultural competency video or in-person training for clinical research team Length: 1 y	N = 67 Cancer type: Mixed Demographics: Not specified but noted that clinicians estimated to have a 20% Black, 5% Latino patient population. Clinical trial type: Therapeutic	Primary: (1) Develop a cultural competency program for research team (2) Determine if program increases cultural competency (3) Determine impact on minority patients to clinical trials recruitment	Nonsignificant difference in recruitment
Fracasso P.M., 2013	Study type: Randomized control trial Study setting: Urban NCI comprehensive cancer center Intervention: Coaching intervention by trained layman Length: 6 months	N = 75 Cancer type: Mixed Demographics: 97% Black, 3% Asian Clinical trial type: Therapeutic or nontherapeutic	Primary: Clinical trials recruitment	Nonsignificant difference in recruitment
Fouad M.N., 2016	Study type: Single arm Study setting: Urban NCI comprehensive cancer center Intervention: Patient navigation Length: 7 y	N = 272 Cancer type: Mixed Demographics: 100% Black Clinical trial type: Therapeutic	Primary: Clinical trials recruitment	Recruitment increased from 9% to 16% between 2007 and 2014
Anwuri V.V., 2013	Study type: Single arm Study setting: Urban NCI comprehensive cancer center Intervention: Structural changes on four levels: (1) leadership support; (2) center-wide policy change; (3) infrastructural process control, data analysis, and reporting; and (4) follow-up with clinical investigators. Length: 5 y	N = Not clearly stated Cancer type: Mixed Demographics: Black, Indian/Alaska Native, Asian, and Native Hawaiian/Pacific Islander (individual percentages not reported) Clinical trial type: Therapeutic and nontherapeutic	Primary: Clinical trials recruitment	Significant absolute increase in therapeutic trial recruitment of 60% (p = .04). When considering that over this period, the number of cancer patients seen proportionately increased, recruitment remained stable (14.7% to 15%)

Abbreviation: NCI, National Cancer Institute.

Patient navigation/coaching

Four studies used a patient navigation/coaching intervention.^29–32 Three consisted of a homogeneous sample (two Black and one American Indian),^29,30,32 whereas one was heterogeneous.³¹ Generally, patient navigators/coaches provided support throughout the cancer care continuum by assisting with activities such as coordinating appointments, facilitating transportation, clinical trial education, and enrollment. Guadagnolo et al. studied how a patient navigator program that included culturally competent navigators and community research representatives affected cancer care, particularly enrollment in CCTs for American Indian patients.²⁹ Notably, these community research representatives were embedded within these communities and served as liaisons between the hospital staff and community. This study recruited 22% of patients into CCTs over a 5-year study period, which compared with a < 1% rate cited in the literature for American Indians. Holmes et al. used a single nurse navigator, from an NCI-designated cancer center, who made weekly visits to community oncology sites to recruit Black female patients with breast cancer. They successfully recruited 86% of eligible patients they identified. Institutional accrual rates of Black patients increased from 3% to 7% over the 2-year study period. Fracasso et al.’s and Fouad et al.’s studies involved racially concordant navigators that were trained lay individuals. Fracasso et al. conducted an RCT of a coaching intervention (vs. usual care) where a single coach provided care to a racially mixed study population. The results showed no difference between the two interventions over the 6-month study period. Fouad’s single-arm study found an increase in enrollment from 9% to 16% of all Black patients with cancer at the institution over the 7-year study period.

Clinical trial information video

Two studies used video interventions for patients to increase accrual into CCTs.^33,34 Robinson et al. conducted a single-arm prospective study showing Black women with breast cancer a brief culturally tailored video with race concordant patients and clinicians (Black patients and clinicians) that addressed common barriers to trial enrollment.³³ Although not described in the Materials and Methods section as an intervention component, a patient navigator was present from enrollment until 6 months afterwards to provide support. At 6 months postintervention, 14% of those who saw the video enrolled in a therapeutic CCT, which represented a 7.5% absolute increase over the previous years. Skinner et al. conducted an RCT that compared an informational video that was viewed at a visit versus taken home by patients from a mix of racial demographics and cancer types.³⁴ Both arms received a clinical trial education booklet to take home. Randomization was not stratified nor was sample size calculation adjusted by ethnicity; 33% of subjects were White and the total sample size was small at 63 patients, making it difficult to draw conclusions about REMs. Unlike Robinson et al.’s study, there was no mention of race concordance among video participants. At 1-year after intervention, there was no difference found between the study arms or between White and REM patients. It was reported that 53% of the video at home arm did not view the video.

Cultural competency training for clinical trials team

One study used a cultural competency training for clinical research associates and physician investigators in an attempt to increase Black and Latino accrual to CCTs.³⁵ It was a 4-hour training delivered either in person at a Radiation Oncology Therapy Group meeting or online. The training educated providers on the barriers, myths, beliefs, and norms within Latino and Black cultures that may hinder clinical trial enrollment. Additionally, educational materials for several Radiation Oncology Therapy Group CCTs focused on disease sites with high incidence in Latino and/or Black populations were created. After the training, there was a nonsignificant 1.5% increase in accrual.

Relationship building and social marketing

Heiney et al. used relationship building and social marketing to recruit Black patients with breast cancer into a behavioral trial.³⁶ Authors built relationships with community members and leaders who helped design their study brochures, logo, and marketing material. They also engaged in activities to establish a trusting bond between the research team and potential study subjects such as sending letters from the principal investigator and communicating with empathy. Accrual occurred via self-referrals (from social marketing initiatives), community partner referrals (including physicians), and contact information through Health Insurance Portability and Accountability Act (HIPAA) waivers. Using these strategies, 66% of eligible patients were recruited into the trial. However, there was no comparison group in this study. Most patients recruited came from participating institutions, especially via the HIPAA waiver, and a small portion came directly from responses to social marketing.

Institutional structural changes

Anwuri et al. attempted to increase REM accrual to CCTs through system-based changes in the research department of an NCI-designated cancer center.³⁷ Changes focused on leadership support, center-wide policy, infrastructural process control, data analysis, and reporting. The program was presented as an NCI requirement and crucial success factor for both scientific and cancer center goals. A centralized accountability system was implemented for monitoring REM accrual. If accrual targets were not being met, a study was at risk of being closed unless a plan was developed to remedy this. The program worked with investigators to help overcome barriers to accrual. They reported a 60% increase in minority accrual to any type of CCT over the 5-year study period; however, this was an absolute increase and, when adjusting for the proportionate increase in REMs seen in the clinic over that time, the accrual rates remained stable.

Quality appraisal

All studies met criteria for clear reporting of a hypothesis, intended measurable outcomes, and study findings. A majority (>50%) clearly described patient characteristics, interventions of interest, and delineated findings by race/ethnicity. Most studies (>50%) also addressed internal validity bias through accurate outcome measures and external validity through a participant pool representative of the population from which they were recruited. Few (<30%) addressed confounding variables in the participant pool (e.g., socioeconomic status, health literacy) or selection bias through randomization of intervention groups. Two studies were not applicable for question 10 entirely (Were losses of patients to follow-up taken into account?) because they were unable to be randomized.^36,37 In all, three of the nine studies met criteria for “higher quality” with scores at 80% or higher.^29,31,32 The remaining six studies were deemed “lower quality” with scores of 50% to 70% (Fig. 2).^30,33–37

Figure 2.

Quality assessment scores.

DISCUSSION

Our systematic review, to the best of our knowledge, is the only review to date that has evaluated studies of interventions to increase REM accrual into CCTs. Only nine studies have been identified since the 1993 NIH Revitalization Act, six of which targeted patients, two targeted providers, and one was a hybrid in which both patients and providers were targeted. Five of the nine studies demonstrated a significant impact on accrual. Each of these studies uniquely addressed well-known barriers and facilitators to study accrual that include, but are not all inclusive of, trust, education, communication, support, cultural competency, and access.^6,7,8,38 A minority, three of the nine studies, met high-quality standards per our quality criteria and, interestingly, all of them were patient navigation/coaching studies. Despite most studies being of lower scientific rigor, we find the evidence promising and useful, especially for the development of future research in this area.

The majority of the reviewed studies were on a navigator/coaching approach (four studies), which has been shown in the existing literature to improve diagnostic resolution and follow-up and access and adherence to cancer treatment and screening.^39–42 The studies included in this review had significant heterogeneity with regard to navigator type and work experience, frequency of contact, cultural congruence, community engagement, environment, and length of study. For example, study navigators consisted of a trained layperson, a nurse navigator, a patient navigator, and a coach; all with varying levels of experience. All studies except for one of the coaching interventions had positive results. As stated by the study authors, the coaching intervention without a positive result could have been due to the shorter duration of intervention compared with the other studies and low frequency of navigator contact. What these studies had in common was that navigators helped overcome critical barriers on multiple levels such as transportation, health literacy, support, and financial concerns. Overall, these results illustrate the potential for patient navigation to improve CCT enrollment as well as the potential for training people without prior experience to fulfill this role.

Two video intervention studies were included in this review that addressed educational and trust barriers to enrollment by conveying the perspectives of those with lived experiences as well as those of health care professionals. Patients spoke about their experiences in CCTs, how they arrived at their decisions, and how their concerns were addressed. Physicians spoke in basic terms about CCTs and, in one of the studies, staff and clergy were also included. Each of the interventions varied in their trial methodology, study population and content. Unlike Skinner et al.’s study, Robinson et al.’s study showed an increase in accrual which could be attributed to racially concordant study coordinators and video participants and a patient navigator who provided follow-up support to the patient for 6 months. Multiple studies have emphasized the importance of culturally tailoring content and cultural congruency in clinic and research staff.^6,43,44 The American Cancer Society Action Network Barriers to Patient Enrollment in the Therapeutic Clinical Trials for Cancer Landscape report recommends maintaining a diverse workforce reflective of underrepresented populations.⁴⁵ This can provide a sense of trust and alleviate attitudinal barriers such as the fear of being a “guinea pig,” which is so often perceived by minority groups.⁴⁶ It is unclear if Skinner et al.’s video intervention included racially concordant patients and staff. Based on Robinson et al.’s study, we believe this modality is promising because it provides an efficient way to address educational and trust barriers. More studies should be conducted using racially concordant experiential patient education videos.

Cultural competence is another important aspect in the success of the provider–patient relationship and clinical trial enrollment. It was defined in an article by Cross et al. in 1989 as “a set of congruent behaviors, attitudes, and policies that come together in a system, agency, or among professionals and enable that system, agency, or those professionals to work effectively in cross-cultural situations.” ⁴⁷ Research on the effects of cultural competency on patient satisfaction is still in its infancy but has shown improvement in the cultural competence of health care providers and increases in patient satisfaction.⁴⁸ One study examined strategies used by US Cancer Centers of Excellence in successfully recruiting REMs into clinical trials and cultural competency training for staff was one of those mechanisms.⁴³ We believe such training is critical to understanding the needs and health care barriers of different populations and serves as a conduit to enhancing provider–patient communication. A study in this review intervened with a 4-hour cultural competency training for clinical research associates and physician investigators. At the 1-year follow-up, there was a nonsignificant increase of 1.5% in CCT enrollment at participant sites versus sites without trained providers. Although the results were nonsignificant, they are still meaningful. This study highlights that merely educating a clinical research group on barriers and cultural beliefs of a community, even without providing the tools to facilitate overcoming these barriers, may improve accrual. Clinical research teams might become more motivated to seek resources to overcome barriers once aware of these issues. Perhaps if there were a formalized process provided in training to overcome barriers, there would have been a significant increase in accrual. This is an area that will require more exploration and would benefit from a study with a larger sample size and training to overcome barriers.

Community engagement was a centerpoint in one study that used a framework that combined relationship building and social marketing to recruit into a therapeutic teleconference group for Black survivors of breast cancer. Overall, it is difficult to assess the impact of this intervention because the number of people exposed to social marketing could not be quantified and there was no comparison group. However, this methodology was innovative because most patients were recruited via institutions and especially via the HIPAA waiver mechanism, which speaks to the importance of building relationships with research teams and focusing on building the research team–participant relationship. Through the HIPAA waiver, potential subjects were contacted over the phone by recruiters who were trained in empathetic communication. The authors note that it is possible that approaching individuals in the comfort of their own homes versus in a clinic setting may have put them more at ease and made them more willing to enroll in trials.

Provider and system level biases, such as provider attitudes, trial eligibility criteria, provider communication, and perceived increased time in explaining trials are known barriers for REM accrual,^7,15,16,49 which was addressed in one study at an NCI-designated institution. A central accountability system was key in mandating and monitoring REM accrual targets. When looking at the results, there was an increase in absolute numbers recruited but not a proportionate increase as the number of REMs seen at the center increased over time. This difference was not addressed by the authors. Nevertheless, this study illustrates the importance of not only shifting an institutional culture toward more inclusivity of REMs in trials but also of creating an accountability system that enforces this policy and provides support in dealing with barriers. We feel that shifting an institutional culture toward inclusivity via cultural competency trainings and mechanisms to address barriers is of paramount importance to ensure equity and patient-centered decision-making. This will ensure that research teams are not merely driven to meet accrual targets but are motivated to enroll in the spirit of equity. This framework is certainly one that other institutions can adapt and improve on.

Limitations

Our systematic review was limited by including only articles written in English within the United States and by beginning our search criteria in 1993 and not before that date.

Conclusion

Despite heterogeneity of study methods and predominance of lower quality evidence, this review highlights potentially effective interventions to improve REM accrual in CCTs. Interestingly, the interventions ranged from resource and time intensive (patient navigation) to a short intervention at one point in time (15-minute informational video), which brings to light the possibility of improving accrual at resource-limited clinical environments. Focus should be especially placed on building interventions for such environments because they are usually where most REMs receive their care. The common thread to many of these interventions was race concordance, community engagement, and addressing barriers throughout the cancer care continuum. Existing literature on overcoming barriers in willingness to enroll in trials also supports implementing these aspects in accrual efforts.^2,6,8,43,50 Therefore, we recommend incorporating them into programs aimed at increasing REM CCT accrual. We believe patient navigation is one comprehensive way to address these facilitators and barriers throughout the cancer care continuum and should be an active area of future research in this arena. Our systematic review highlights the critical issue of insufficient REM representation in clinical trials and urgently calls for the development of effective enrollment interventions. Future studies should consider these previously mentioned aspects and explore the studied interventions we have presented in RCTs with larger sample sizes, minimal bias, more diverse populations, and longer study periods.