Understanding Hispanic/Latino Participation in Clinical Trials and Observational Studies, and Strategies to Increase Participation: A Targeted Literature Review

Abstract

Hispanic/Latino representation in medical research remains poor. We describe factors affecting rates of recruitment, participation, adherence, and retention of Hispanics/Latinos in clinical studies in the United States and characterize proposed strategies to improve these rates. A targeted literature review was conducted. Relevant studies were identified from Embase, MEDLINE^®, and CENTRAL from January 1, 2010 to September 4, 2020. Sixty-eight studies were included. Key facilitators to research involvement were establishing trust between research staff and participants, incorporating familism, and using culturally appropriate language. Common elements of successful strategies for improving research involvement included incorporating community partners, bilingual and culturally competent research staff, continuous engagement and building relationships between participants and staff, and incorporating Hispanic/Latino cultural values. There is no universal strategy to improve research involvement of Hispanics/Latinos. The best strategy is likely a combination of key elements from several strategies, tailored to each unique study population. Further research is needed.

The underrepresentation of ethnic and racial minorities in clinical trials is a persistent issue in the United States (U.S.), despite the National Institutes of Health (NIH) Revitalization Act of 1993 stipulating the mandatory recruitment of women and minorities in all NIH research.^1,2 A recent report from the U.S. Food and Drug Administration (FDA) on its 2020 Drug Trial Snapshots showed a notable disparity between the representation of minorities and the representation of non-Hispanic Whites in clinical research, with the latter making up only 62% of the U.S. population, but 76.3% of research participants within clinical trials for new molecular entities and therapeutic biologics.³ Additionally, Hispanics/Latinos/Latinas/Latinx (H/Ls) represent 18.5% of the U.S. population, but only 11% of trial participants.^3,4 Since the NIH Revitalization Act of 1993, initiatives to recruit participants from minority groups have grown more common. These initiatives have had varying success, as seen by the decreasing H/L participation rate in a study assessing participation in the National Institute of Neurological Diseases and Strokes trials before 1995 (7.5% participation rate) and after 1995 (5% participation rate, p <.001).⁵

Within the context of research, recruitment is generally considered the first chronological step in engaging and enrolling participants into the study (before the study has begun), participation is the second step where recruited participants provide baseline data and begin study follow-up, adherence is the act of a participant following study protocol and committing to the assigned intervention throughout the study, and retention pertains to keeping enrolled participants in the study for its duration. Each step in participant involvement is crucial to the success of clinical trials. Furthermore, the health of individual patients suffering from diseases may hinge on their involvement in research and compliance with treatment protocols. For example, Mexican Americans have disproportionately high rates of type 2 diabetes and experience more diabetes-related complications compared with non-Hispanic White Americans.^6,7 However, rates of inadequate treatment adherence to oral medications has been reported to be between 51.7% and 60.2% for H/Ls, compared with 37.5% for non-Hispanic Whites.⁸ In this way, patients who are most in need of treatment may not be receiving the proper education and care they need to manage their health.

Achieving high participant involvement in research can be challenging, especially in communities that have historically been underrepresented in research and where mistrust of researchers is common.^9–11 Thus, proposed solutions to address clinical trial/study recruitment in the H/L population should be evaluated and described to better serve these patients. Thus, the objective of this review is to identify and describe factors that affect rates of recruitment, participation, adherence, and retention of H/Ls in clinical trials and observational studies in the U.S. Additionally, we aim to characterize proposed strategies to improve these rates.

Methods

A targeted literature review was conducted. Study eligibility criteria were defined using a modified PICO (Population, Intervention, Comparator, Outcomes) framework. Eligible studies included clinical trials or observational studies that reported factors affecting study recruitment or participation, adherence or compliance to treatments or study protocols, or study retention of H/L adults or children. Selected publications were focused on data from the U.S., available in English, Spanish, or Portuguese, stating that at least a portion of their sample population included H/Ls. Abstract and full-text selection, full-text review, and data extraction were performed by a single senior reviewer, and all steps were overseen and quality-checked by a lead senior investigator. Results were narratively summarized.

A literature search was conducted by searching for articles published between January 1, 2010 and September 4, 2020 in MEDLINE^®, Embase, and Cochrane Central Register of Controlled Trials (CENTRAL) according to pre-defined search strategies. Relevant conference abstracts published in the years 2018–2020 were also identified via Embase. In addition, gray literature searches of funding agency websites (i.e., NIH; https://nihpublications.od.nih.gov/) and Google Scholar, and bibliographies of included literature review articles were hand-searched to capture relevant publications. The reviewer extracted all relevant data from the final set of included studies. The following data were extracted where available: (a) study name, year of publication, study authors, study design, study setting, sample size, follow-up length, and participant exclusion/inclusion criteria; b) name and description of the strategy to improve recruitment, participation, adherence, or retention rates of H/Ls; c) characteristics of the study participants including age, sex, race/ethnicity, country of origin, spoken languages, disease or therapeutic area; (d) factors affecting recruitment, participation, adherence, or retention rates of H/Ls, and effectiveness of strategies to improve these rates. Strategies and factors affecting these rates were categorized by a senior reviewer to facilitate the summary of information from the most frequent categories. Categorization of studies by what stage of research involvement they were investigating (i.e., recruitment, participation, adherence, or retention) was based on the language used in the studies themselves.

Results

Study selection and characteristics

A total of 8,332 citations were identified from the electronic databases and from gray literature searches (Figure 1). Following full-text screening, 68 publications were included.

Of the 68 included publications, six were available as conference abstracts and the remaining were available as full-text articles. Most studies used cross-sectional (n = 23) or descriptive study designs (n = 16).

Twenty-one studies reported the length of follow-up, which ranged from one week¹² to six years⁸ (median: 12 months). Forty-two studies were multi-center, 19 were singlecenter, and seven did not report this characteristic. As per our eligibility criteria, all studies were conducted at least in part within the U.S. California was the most common state (n = 14), followed by Florida (n = 9) and New York (n = 6). The number of participants in each study ranged from nine¹³ to 30,838.⁸ A common inclusion criterion across studies was proficiency in either English or Spanish (n = 13).

Participant characteristics

Sociodemographic characteristics

Mean age of H/L participants was reported in 24 studies, ranging from 14.1¹⁴ to 75¹⁵ years (median: 49.9). Across 23 studies, the proportion of female participants ranged from 19%¹¹ to 100%¹⁰ (median: 66.6%). The proportion of female participants was 50% or higher in 18 studies.

The proportion of H/L participants was reported in 51 studies — 22 of these consisted of only H/Ls; in the remaining 29 studies, 11 aimed to enroll ethnic/racial minorities, and in these, H/Ls were between 20%¹⁶ and 93%¹⁷ of the total sample (median: 44%).

Figure 1.

PRISMA flow diagram.

Studies that did not aim to enroll only ethnic/racial minorities (n = 17) reported proportions of H/L between 3%¹⁸ and 98%¹⁹ (median: 16%).

Country of origin among H/L participants was reported in 10 studies. The U.S. was most frequently reported (n = 7), ranging from 10%²⁰ to 63%²¹ (median: 23.1%). Other countries of origin included Mexico (n = 5), Cuba (n = 3), Nicaragua (n = 3), Colombia (n = 2), Guatemala (n = 2), Honduras (n = 2), Peru (n = 2), Puerto Rico (n = 2), Argentina, Bolivia, Costa Rica, Ecuador, El Salvador, Venezuela, and the Caribbean (n = 1 each). The proportion of H/L participants who spoke English was reported in six studies, ranging from 3%²⁰ to 50%¹⁵ (median: 30%). Nine studies reported on the proportion of H/L participants who spoke Spanish, which ranged from 45.9%²² to 100%¹⁰ (median: 70%).

Disease or therapeutic area

A wide variety of diseases were represented in the set of 68 included studies. The most reported disease in the studies was cancer (n = 24), followed by mental health disorders and addiction (depression, psychosis, mental health issues, and substance abuse, including smoking; n = 9), diabetes (n = 6), HIV/AIDS (n = 6), and Parkinson's disease (n = 4). Across the cancer studies, general cancer (n = 16) and breast cancer (n = 6) were the most frequently reported diseases.

Strategies to increase recruitment, participation, adherence, or retention

Thirty studies aimed to test strategies for improving the recruitment, participation, adherence, or retention rates of H/Ls in clinical trials or observational studies. Each strategy was unique to a single study. A table summarizing the strategies that were quantitatively measured for their effectiveness in improving recruitment, participation, adherence, or retention rates (i.e., according to a pre-specified quantitative goal or statistically significant result versus a control) appears as Table 1 (n = 16). Strategies were manually categorized into those that demonstrated quantitative effectiveness and those that did not. The remaining studies only qualitatively described the effectiveness of their strategy (n = 14).^{15,18,19,23–33} Notably, 10 studies only described their strategy to improve study involvement, but did not test its effectiveness.^20,34–42 This section of the manuscript focuses on the 16 studies measuring their strategy's effectiveness quantitatively.^43–48

Recruitment strategies

A total of 18 strategies aimed at improving recruitment of H/Ls into research were identified. Of these, five were shown to be effective according to a pre-specified quantitative recruitment goal or statistical test versus a control. Strategies were community-based participatory research approaches (use of community partners; engagement of community; communication strategies);⁴⁹ an engagement, recruitment, and retention process involving formative assessment and community engagement specialists;⁴⁵ digital marketing campaigns;⁴⁸ a recruitment strategy involving four bicultural, bilingual clinical research assistants;¹³ and recruitment strategies involving a community advisory board, bilingual flyers, and presentations and exhibits at community events.⁴⁶

Noteworthily, three strategies were identified that were not shown to be effective in improving recruitment rates. The first strategy was an office-based clinical trial educational video and the study showed no difference in the proportion of participants who enrolled in a clinical trial after viewing the video compared with the control of no in-office education after one year (3 vs. 2 enrollees; p = .69).⁴³ The second strategy was a minority-based community clinical oncology program which failed to meet its accrual targets of 60 patients in the first year of recruitment (only 26% of this goal was reached), 80 in the second year (37.5%), and 100 in the third year (48%).⁴⁴ The third and final strategy that was deemed ineffective was the use of recruitment letters specialized for the recipients' racial/ethnic group.⁴⁷ These letters did not improve screening or enrollment rates in the overall sample nor the racial/ethnic groups (p ≥ .48) compared with standard recruitment letters.⁴⁷

Participation strategies

Eleven strategies developed for improving participation of H/Ls into research were identified. Of these strategies, none were found to directly improve participation rates (i.e., through presentation of data showing increased number of participants following application of the strategy). However, two demonstrated effectiveness in improving outcomes related to participation, including understanding or knowledge of clinical trials, and perception of benefits or barriers to clinical trial participation. The first of these strategies was called Choices, and was described as a bilingual multi-component strategy including an educational interactive video, a low-literacy booklet, and care coordination by patient navigation.²² Following receipt of Choices, perceived understanding of clinical trials (p = .033) and consideration of clinical trials as a treatment option (p = .008) improved compared with controls.²² The second strategy was a single-session slide presentation with embedded videos called the Adaptation of Clinical Trials Education Program for Hispanic Americans.^50–53 Benefits

Table 1.

SUMMARY OF QUANTITATIVELY EFFECTIVE OR NOT EFFECTIVE INTERVENTIONS/STRATEGIES TO IMPROVE RECRUITMENT, PARTICIPATION, ADHERENCE, AND/OR RETENTION OF SUBJECTS INTO RESEARCH

following the presentation included increased mean knowledge of cancer clinical trials (p = .034), and improved perceived benefits (p = .019) and barriers (p =.013) to cancer clinical trials.

One strategy was identified that did not statistically improve outcomes related to participation of H/Ls.⁴³ The strategy was embedded within a 15-minute survey, in the form of three clinical trial educational questions that recipients had to answer. The first three questions on the survey assessed the initial willingness to participate in clinical trials, and the subsequent three questions were the clinical trial educational questions. The final question again assessed whether recipients were willing to participate in a clinical trial. Following the survey, the proportion of H/Ls who were willing to participate did not significantly change (8 [57%] vs. 11 individuals [79%]).

Retention and adherence strategies

A total of seven studies described strategies to improve retention in health research or adherence to medication. Of these, only two strategies were considered effective in improving retention rates for H/Ls.^45,46 The first retention strategy included the use of a promotora (promoter), a lay H/L community member who has been trained to educate the community on basic health principles and who serves as a liaison between the community and health care professionals.⁴⁶ The promotora made weekly reminder phone calls to participants, provided contact information for questions or concerns, and gave weekly homework assignments on content discussed during that session. Twenty-five of 38 (65.8%) participants completed the entire five-month follow-up, meeting the retention goal of 20 participants for the promotora group. The retention rate was lower in the control group, where only eight out of 20 (40%) individuals completed the entire five-month program. In the second study, a combined engagement, recruitment, and retention strategy involving telephone reminders, increased site visits, and over-scheduling of follow-up interviews, was used.⁴⁵ Authors reported that retention strategies yielded 100% retention rates for all recruited sites throughout the study.

One strategy aimed at improving adherence to medication—a cultural strategy featuring H/L values delivered by Spanish-speaking nurses—was deemed effective after those in the cultural strategy group were shown to take a significantly greater number of doses of treatment compared with the historical control group (157 pills [5.2 months' worth] vs. 129 [4.3]; p = .028).^54,55 One strategy was ineffective at improving adherence to medication rates among H/Ls. The use of PictureRx, an illustrated medication list, was not associated with increased self-reported adherence compared to the usual care group (Adherence to Refills and Medications Scale Scores: 10.3 PictureRx vs. 9.9 usual care).¹² Notably, the authors found that patients randomized to PictureRx had better overall understanding of their medications (p < .001), including greater ability to report the drug indication (p < .01), strength (p < .05), dosing (p < .01), and frequency of administration (p < .001). However, this increased understanding did not translate to improved adherence.

Factors affecting recruitment, participation, retention, and adherence

The remaining non-interventional studies contributed to the objectives of this review by reporting perceptions of or attitudes towards research from H/Ls, as well as factors affecting recruitment, participation (or willingness to participate), retention, or adherence.

Facilitators

A summary of commonly reported (i.e., reported by two or more studies) facilitators to study recruitment, participation, retention, and adherence to the study protocol or medication is provided in Table 2.

Facilitators related to communication and the relationship between the potential participant and their physician and research staff were reported by 14 studies. Emerging categories of facilitators included the establishment of the participant's trust in their treating physician, research staff, or the health information they have received (n = 9). Additionally, language was a strong driver of successful participation as well as adherence for H/Ls (n = 8). Several studies described how language concordance with the participants' physician or research staff was a facilitator for successful participation. The provision of Spanish-language study materials was also deemed necessary, and use of culturally appropriate terms (e.g., study may be thought of as schoolwork, and does not translate the concept of a research study), tone, and non-verbal communication such as body language was suggested. The level of literacy of the participants should also be considered when communicating with potential participants or when providing study-related materials.

Participant-related facilitators to participation, adherence, and retention were reported by 11 studies. Among these, family support or participation was identified as one of the most reported facilitators to H/L participation and adherence (n = 5). For example, family members assisting in organizing medications in boxes and reminding participants to take them was described as a key facilitator to H/L participants.^56–68 Education level was a commonly reported sociodemographic factor that was often linked to participation and adherence rates of studies (n = 5). Generally, participants who had an education level of some college or more had higher chances of participating and being retained within the study, whereas those with education levels of high school or lower did not. Other sociodemographic characteristics related to participation and retention were being married (n = 3), as well as higher self-reported baseline quality of life (n = 2).

Facilitators of participation and adherence related to participant beliefs were reported by nine studies. The most common belief-related facilitator to participation was the belief that the potential participant's contributions to the study would benefit others or their family members (n = 6). Like the motivator of helping others, perceived personal benefits (e.g., benefits from receiving treatment for their ailment during the trial) was also a commonly reported facilitator to participation (n = 3).

Lastly, 11 studies reported study-related facilitators to recruitment, participation, or retention. Among these studies, adequate knowledge of or familiarity with the clinical trial or study at hand or clinical trials/research in general was an emerging category, reported by six studies. Helping the participant's understanding of the purpose behind conducting the research and their roles as a research participant, clarifying any misconceptions they may have regarding clinical research and health information, and relieving uncertainty of clinical trials in general were described as facilitators. Another study-related facilitator to participation was the incorporation of monetary incentives for potential participants (n = 4). Most studies reported that participants appreciated this gesture in exchange for their time and efforts. Notably, however, one study found that monetary incentives reduced the likelihood of healthy individuals participating in an influenza vaccine trial.¹⁷ The authors described this finding as a paradox, since

Table 2.

SUMMARY OF COMMONLY REPORTED FACILITATORS TO RECRUITMENT, PARTICIPATION, ADHERENCE, AND/OR RETENTION.

financial benefit is generally a key motivator for the participation of healthy individuals in clinical trials. Nevertheless, a possible explanation of this finding was provided, which stated that potential clinical trial participants may have associated the monetary incentive with increased perception of risk. In line with this finding, another commonly reported study-related facilitator for both recruitment and participation was an adequate description of the clinical trial (including purpose, and role of the participant) or associated risks of contributing to the study (n = 3). Adequate description provided by the research staff may help the potential participant feel more secure and reduce stress or concerns.

Barriers

A summary of commonly reported (i.e., reported by two or more studies) barriers to study recruitment, participation, retention, and adherence to the study protocol or medication is provided in Table 3.

Common societal barriers were reported by 12 studies and included those related to costs associated with participation (or income; n = 5), lack of accessible transportation to the trial sites (n = 5), lack of accessible childcare or elder care, and time demands associated with participation (n = 2). These factors were seen as addressable at a societal level, by creating more accessible or affordable public transportation and child or elder care; however, at a study level, provision of transportation or monetary incentives may also help to combat these barriers.

Common barriers related to participant's beliefs were reported by 10 studies, and the most common was fear of adverse effects of study interventions (generally applicable to pharmaceutical interventions; n = 5). Three studies reported that some participants had the perception that they would be used as guinea pigs to test new drugs or treatments that were not adequately evaluated for safety.

Finally, common participant-related barriers included the age of participants (n = 4), where those who were older tended to refuse to participate. Immigration status was another commonly reported barrier (n = 2). Participants in a focus group suggested that researchers and study advertisements should make sure to clearly state that immigration status will not be documented during the study and will not prevent participation.² Importantly, the study stated that questions about the length of residency in the U.S. can spark feelings of mistrust in the potential participants regarding the intentions of the researchers.

Three studies did not describe commonly reported barriers or facilitators to recruitment, participation, retention, or adherence, and are reported here separately.^{70–72,73–75} Chang et al. examined factors associated with willingness to be contacted for research on depression among H/Ls and other racial/ethnic groups.⁷³ Authors found that among H/Ls, those whose primary language was Spanish were more likely to agree to be contacted for research compared with those whose primary language was English (odds ratio 1.741; p< .001). Diehl et al. evaluated accrual patterns across 10 clinical trials conducted by the American College of Surgeons Oncology Group, with results stratified by racial/ethnic group.⁷⁴ Authors reported that clinical trial protocols that were successful in recruiting H/Ls had the following features in common: studies were designed to recruit patients with regional or advanced-stage disease, and studies involved an investigational systemic therapy. Authors concluded that study design can influence recruitment patterns. Lastly, the study by Langford et al. aimed to examine

Table 3.

SUMMARY OF COMMONLY REPORTED BARRIERS TO RECRUITMENT, PARTICIPATION, ADHERENCE, AND/OR RETENTION

the association of race/ethnicity with clinical trial awareness and to inform future communication campaigns to increase clinical trial awareness and participation.^75,76 Investigators reported that H/Ls were less likely to be aware of clinical trials compared with non-Hispanic Whites (odds ratio: .62; 95% confidence interval .49–.78), and that lower awareness of clinical trials is at least a partial explanation for lower minority participation in trials.

Discussion

The aim of this targeted review was to identify and describe factors affecting rates of recruitment, participation, adherence, and retention of H/Ls in clinical trials and observational studies in the U.S. We also aimed to characterize proposed strategies to improve these rates.

Among the studies that reported strategies to improve recruitment, several patterns emerged among the successful strategies. The first theme that emerged was the use of community-based approaches to enroll potential participants. Use of community partners or community champions (i.e., an individual who helps link the community, has good interpersonal skills, and can serve as a broker on behalf of the health initiative within the community) were common approaches that were generally associated with successful recruitment efforts. These community specialists may help recruitment efforts by shaping the strategy or study materials to match the community's needs, advocate for the community's participation into the study, and provide recommendations to study investigators on recruitment strategies. Ideally, community members should be engaged throughout a research study, from identifying the issue or question to be tested, all the way to translating the results back to the community.⁷⁷ In a previous systematic review that aimed to synthesize the existing evidence on interventions to enhance enrollment of vulnerable populations (minority, underserved, poor, rural, urban, or inner city) into health research studies, the most common strategies for recruitment were found to be social media, community outreach, health system recruitment, and referrals.⁷⁸ Although most of their included studies focused on Black participants, H/L populations were the focus of 46% of studies. Like our own review, the rigor of methodologies employed by the included studies varied: less than a quarter of studies compared interventions using statistical analysis. Authors concluded that further research using robust study designs is needed.

For participation, no strategy emerged as objectively successful. However, of the two strategies that demonstrated effectiveness in improving knowledge of clinical trials and the participant's perception of benefits or barriers to participation,^22,50 a common theme of bilingual information and staff emerged. Additionally, both strategies used a similar format for delivering their information, by means of a slideshow presentation or video. However, it should be noted that although not explicitly tested, it is likely that several of the strategies identified to improve study recruitment would also be applicable for improving study participation. In many studies, stages of recruitment and participation are chronologically close together. Thus, successful strategies such as community engagement may also be a good option to consider for keeping patients engaged in the early stages of the study. In support of this, a previous review focused on identifying strategies to improve participation among racial/ethnic minorities (one of which was H/Ls) showed that community involvement was a common approach that helped to facilitate participation.⁷⁹

In terms of retention, the two strategies that demonstrated effectiveness both had several key components in common that were likely associated with their success.^45,46 In both studies, a community engagement specialist or promotora was involved to support the study and to act as a point of contact for questions or concerns, which helped to facilitate trust in the relationship between researchers and participants. Additionally, both strategies used phone call follow-ups to remind participants about the study. This finding is in alignment with a previously conducted review focused on broader minority populations⁷⁹ suggesting that community involvement is an effective strategy not unique to H/Ls. Finally, one strategy to improve adherence of H/Ls was considered to be quantitatively effective.⁵⁵ Authors used a cultural strategy that included key H/L values.²⁸ Integration of these cultural values into research with H/Ls may be effective in improving adherence.

In addition to the patterns seen across the strategies to improve recruitment, participation, adherence, and retention rates of H/Ls, several themes emerged across the facilitators and barriers to these rates. One key theme was that of H/L mistrust towards the research staff, physicians, or health information providers. This theme tied into several commonly reported barriers including the belief that participants are being used as guinea pigs in research, where unsafe treatments may be administered to these groups to test them before they are received by the general public.^10,61,69,79 In line with this, the fear of adverse events associated with the administered treatments (e.g., pharmaceutical products) was also a commonly reported barrier,^10,61,69,79 highlighting the need for clear explanations of the risks associated with the treatments to be provided and the establishment of rapport between the research staff and participants. A fear that the immigration status of the participants might be revealed to government bodies was also a trust-related barrier, and it was recommended that researchers clarify that immigration status will not be documented during the study. Medical mistrust and perceived discrimination can contribute to lower satisfaction with health care among H/Ls.⁸⁰ Additionally, it can lead to patients not receiving proper care as individuals may be less likely to take medical advice, attend follow-up appointments, or continue prescriptions.⁸¹ Although this issue is complex and often informed by long histories of discrimination, medical mistrust may be mitigated by consistently improving the health research literacy of H/Ls over time. Specifically, mistrusting H/Ls may benefit from being assured of their rights and protections for human trial participation to relinquish any fears of being a experimented on, addressing concerns about potential side effects of interventions, and having a safe space to learn and inquire about the research at any point in the study.

Another key theme in the facilitators and barriers to research was familism, including family support for participating into the study or family involvement in the study.^{14,59,61,68,69,81} Reminders from family members to take their medication or family acceptance of the individual's participation into the study were strong facilitators to H/L involvement in research. Additionally, the perceived belief that participation in research would indirectly benefit one's family members was a commonly reported facilitator for participation. These findings emphasize the importance of family as a valuable resource that should be incorporated into clinical trials and studies of H/Ls. Lastly, the theme of appropriate use of language in research was a key facilitator to H/L involvement in research. Key factors included concordance of spoken language with the physician or research staff, provision of Spanish-language study materials that are at the appropriate literacy level and include culturally relevant terms and images, and use of appropriate tones and non-verbal communication when interacting with participants. It is important to consider each of these aspects of communication to promote successful involvement of H/Ls into research.

Overall, for clinical trial and observational study participants, the ability to decide whether to participate is based on a combination of factors, including having knowledge about science, understanding of the research process, being able to seek out and evaluate health information, and being able to trust that the messages coming from the providers and the system are appropriate for the populations served. Among the many factors that act as barriers to participation in research, few are easily modifiable by the clinical investigator, but consciously developing tools to improve communication and education among minority populations may have an impact.^76,81

Several exploratory questions we initially hoped to answer in this review could not be addressed due to lack of consistency across studies. Specifically, our review did not identify any meaningful trends suggesting that effective strategies to increase recruitment, participation, retention, and adherence rates may differ across study designs. Although this finding could indicate that strategies effective for improving study involvement and adherence in observational studies would be the same as those that would benefit clinical trials, we cannot rule out the possibility that this lack of a trend was due to the substantial variation seen even within study design categories (e.g., there were many different types of observational studies). In light of a growing body of evidence on this topic, future reviews may choose to restrict their inclusion criteria to certain study designs to be able to draw stronger conclusions for successful participation strategies within randomized controlled trials, for example. Furthermore, we did not identify any apparent trends in the success of strategies to improve study involvement and adherence across different therapeutic areas. Diseases being investigated tended to vary across included studies, and no consistent themes emerged that were specific to certain disease areas. Similarly, future research focused on specific disease areas may yield clearer trends in solutions to H/L participation.

Our review has several notable strengths. First, to our knowledge, this is the first literature review to comprehensively assess the factors that affect rates of recruitment, participation, adherence, and retention of H/Ls in both clinical trials and observational studies in the U.S. Existing literature reviews are often focused on broader racial/ethnic populations which do not necessarily reflect H/Ls well.^{77–79,82–84} Additionally, our search methods included extensive searches of major electronic databases, gray literature sources, and bibliographies of captured reviews. However, several knowledge gaps and limitations of this review were identified. First, the included studies were largely heterogeneous in terms of sample size, disease or therapeutic area, inclusion and exclusion criteria, and study design. Consequently, comparison and summary of study findings was challenging, and the summarized findings may not be generalizable to all diseases and populations. Additionally, some of the included studies enrolled mixed populations of H/Ls and non-H/Ls (other minorities). Findings from these studies were considered applicable to both H/Ls and non-H/L minority groups as the strategies were designed for increasing participation of underrepresented groups overall. However, it cannot be determined whether such strategies to improve study involvement are effective specifically in H/L s without within-study stratification of results by race/ethnicity. Second, the present review was limited by the reporting style of the included studies. Specifically, many studies did not provide a clear quantitative assessment of the effectiveness of their strategy to improve recruitment, participation, retention, or adherence of H/Ls, and therefore, strong conclusions on the relative efficacy of each strategy could not be drawn. Similarly, many barriers and facilitators to research involvement of H/Ls were assessed qualitatively by study authors, and as a result, it was not possible to determine the relative weights of these factors, and therefore which factors should be considered most significant. Further research using robust study designs—for example, randomized controlled trials to assess the effectiveness of participation strategies—are warranted to identify the most effective strategies to improve study involvement of H/Ls in various disease areas. Finally, although we kept our inclusion criteria open to any type of research that involved this population, emergent from the findings of our review, it became evident that the majority of the literature is focused on medical studies. Future research on participation of patients in non-medical fields (e.g., social sciences) would also be important to fully understand factors affecting rates of study involvement.

Conclusion

Though a substantial amount of literature exists on the barriers and facilitators to recruitment, participation, and retention of H/L and ethnic/racial minority patients in research, there are still areas that could be improved. Due to substantial heterogeneity in the study designs, populations, and objectives, as well as lack of explicit reporting, there appears to be no universal strategy to improve recruitment and participation of H/Ls into research. The best strategy is likely a combination of key elements from several strategies, tailored to suit each study population. More research is needed on more effective ways to raise awareness about participation in research among H/Ls. Some suggestions for themes to be further explored in future research are: the use of technology and social media to recruit and enroll patients in research, the potential role of nurses in the recruitment of patients for research, modification to the training given to recruiters, education and engagement with physicians to better promote recruitment among H/Ls, and building awareness of clinical trials through promoting trials in clinics, at community events, and within the larger health system.