Abstract
This report describes implementation of a clinical trial testing a behavioral approach to reduce stroke risk factors in Black men with stroke or transient ischemic attack (TIA). A survey of research team members identified macro-, mediator-, and micro-level barriers and facilitators to research engagement and participation and identified approaches to address these barriers. Not being aware of the study was the key macro-level barrier. The team used multiple methods of outreach to inform participant candidates including in-person and virtual community events. Key mediator-level barriers were lack of phone and transportation access, and clinicians not referring patients. A multi-pronged strategy was employed, including in-person meetings (for those who lived in the region) often at locations that were accessible to men in their neighborhoods. To help with transportation challenges the team provided ride-share services, bus fare or parking vouchers. At the micro-level, the team also used multiple approaches to help participants manage barriers related to stress or family life circumstances and acute or chronic health conditions. Being able to conduct assessments remotely was critical to being able to reach this community of men who often had significant functional impairments, such as difficulty walking or in speech fluency because of the effects of stroke. When possible, staff encouraged the involvement of care partners. Successful research outreach and engagement addresses barriers at the macro-, mediator-, and micro-levels, and use mixed-method (in-person + virtual) approaches targeted to the specific challenges individuals may be facing.
Introduction
In 2022, the U.S. Health and Human Services and U.S. Centers for Disease Control and Prevention participated in an event that acknowledged the 50th anniversary of the end of the United States Public Health Service Study of Untreated Syphilis in the Negro Male at Tuskegee and Macon County, Alabama, 1932 to 1972 (Centers for Disease Control and Prevention [CDC], 2024). Unethical practices in the Tuskegee study led to substantial and long-term damage for participants in the study as well as their families and the widely publicized information about these abuses have contributed to enduring medical mistrust in the broader Black community.
It is perhaps not surprising that previous abuses in research have contributed to ongoing research hesitancy (Shavers et al., 2002), although more recent data note that historical abuses are no longer cited as a primary driving factor in research underrepresentation (Rencher & Wolf, 2013). Black communities are diverse and face multiple challenges to research participation including competing life commitments and time demands. Some individuals may hesitate to participate in research because they fear it could confirm negative racial stereotypes (George et al., 2014; Lang et al., 2013). While Black communities overall express a broad interest in research participation (especially among those with previous research exposure) a robust body of data across diverse health conditions suggest that there continue to be multiple challenges to recruit and retain Black research participants in clinical studies (Aaron, 2016; Compadre et al., 2018; Diaz, 2012; Suarez et al., 2020).
There is a robust body of research that notes that some of the same factors that caused delays or access problems among minorities seeking and obtaining health care are also barriers to participating in medical research (George et al., 2014). Research hesitancy may be particularly pronounced among older individuals (Strayer et al., 2023). The COVID-19 pandemic, which began spreading across the world in 2020 and imposed significant multi-year health impact, had devastating effects including death, disability and unprecedented human suffering (Dattani & Rodés-Guirao, 2024). Clinical research was also heavily impacted by the COVID-19 pandemic (Sathian et al., 2020).
Other reports on research barriers, including a model originally outlined by Levkoff and Sanchez (2003) and further tested by Loue and Sajatovic (2008) and Kanuch et al. (2016), have framed barriers to participation in research in the context of macro-, mediator-, and micro-level barriers and facilitators. Macro-level barriers are those that exist in the environment or broader society (such fear of confirming racial stereotyping), mediator-level barriers are those that exist in relation to health services (such as medical mistrust), while micro-level barriers are specific to the individual.
The interplay between medical mistrust, medical research hesitancy and ongoing health disparities among Americans are apparent in common conditions, such as cardiovascular disease and stroke. While stroke is the fifth leading cause of death in the U.S. and a leading cause of long-term disability (Imoisili et al., 2024) stroke burden is not distributed equally across communities. Stroke prevalence is higher among adults aged ≥65 years and among racial minorities including Black adults (Imoisili et al., 2024). A recent report found that hypertension was the strongest contributor to ischemic stroke among the Black individuals, particularly among Black men, where the population-attributable risk was 45.8% (Aradine et al., 2022).
This report focuses on the experience of implementing a clinical trial testing a novel behavioral approach to reduce stroke risk factors, especially hypertension, in Black men who have had stroke or transient ischemic attack (TIA; Still et al., 2021). We focus on Black men in this study because stroke is a leading cause of death and disability in this group and there is limited information on best approaches to address potentially modifiable risk factors, such as uncontrolled blood pressure. The analysis aimed to identify macro-, mediator-, and micro-level barriers and facilitators to research engagement and participation among Black men with stroke or TIA and to describe approaches that might minimize these barriers. The report is intended to assist researchers to recognize and address potential barriers to engagement and recruitment in this community both generally and in relation to public health events that may occur in the future.
Method
Clinical Trial Description
This analysis is part of an ongoing larger 6-month prospective randomized controlled trial (RCT) evaluating the effects of a stroke risk reduction approach called
Clinical Trial Inclusion Criteria and Overall Recruitment Approach
All study participants experienced a stroke or TIA within 10 years. Eligible participants also had the following inclusion criteria: (1) self-identified as Black male, (2) current Barthel index (BI) score of ≥ 40 (Kasner, 2006), and (3) able to participate in study procedures. Participants were excluded if they were unwilling or unable to provide informed consent or if their stroke was related to sickle cell disease. Individuals were recruited using a variety of methods including solicitation of referrals from local/regional medical care providers, outreach via free or paid advertising, such as Research Match (Harris et al., 2012), querying the electronic health records (EHRs) of two medical health systems, use of IRB-approved advertising and in-person outreach.
Identification of Barriers to Engagement and Recruitment
This analysis was informed by the research barriers model of Levkoff and Sanchez (2003). The study team’s experience in identifying and addressing each of the barrier levels were collated and ranked to inform a set of approaches that other clinical trial teams might consider in research that seeks to engage and recruit Black men with cardiovascular disease.
For this analysis, an anonymous online survey was conducted among members of a clinical trial team to identify, rank order, and address barriers to engagement and recruitment. The survey was conducted after enrollment was completed while clinical trial data collection was still in process. Multidisciplinary team members of the clinical trial group who had any role in outreach planning/outreach activities independently identified factors they experienced as barriers to participation among their sample. The survey was administered anonymously to minimize bias among staff with differing responsibilities. Study team members based their responses upon what they perceived to be barriers to engagement and enrollment based upon what they were told by participants and the cumulative interactions they had with study participants.
A general list of 11 barriers previously identified in a clinical trial enrolling people with diabetes and mental health conditions (Kanuch et al., 2016) comprised the survey items. Team members were also asked to identify additional barriers encountered in their experience. The 11 general barriers included lack of phone access, transportation difficulties, chronic physical health problems, schedule conflicts, acute health issues, mental health issues, weather, disagreement with diagnosis, stress from family or living circumstances, potential subjects being unaware of the study and clinical providers not referring patients due to competing demands.
Team members were asked to identify barriers to both initial engagement (when individuals are initially approached) and to eventual enrollment (consented to participate in the study). Team members were also asked to independently rank barriers to either initial engagement or eventual study enrollment. Rankings were operationalized on a continuum of “very common” (3 points), somewhat common (2 points), or “not very common” (1 point). Study team members were also asked to independently summarize how they and other study team members dealt with each of the barriers.
Once all final survey responses were collated, the team met by videoconference to review /discuss the summative list of barriers, group the barriers into environmental/social (macro-), health care system/public services (mediator-), or individual (patient-) level barriers and to summarize the approaches that the team used to address each barrier. Finally, team members provided input on any additional approaches to address barriers that may not have been included in the survey collation list.
Results
Clinical Trial Participants
Mean clinical trial sample age was 61.19 years (standard deviation/SD of 9.47 years. The overwhelming majority (n = 142, 88.8%) had at least a high school education. Just over one-third (n = 55, 35.26%) lived alone. There were 65 individuals (41.40%) on Medicaid insurance. The majority experienced a stroke (n = 114, 71.25 %) while just over one-quarter (n = 46, 28.75%) experienced a TIA. Figure 1 shows referral sources for individuals that were enrolled and randomized.
Referral Sources for Enrolled Participants
Study Team Sample
There were nine current or past study team members who responded to the anonymous survey. Staff roles were diverse and included study principal investigators (PIs), research coordinators, and outreach staff.
Barriers to Initial Engagement and Recruitment
Figure 2 shows relative ranking of barriers to initial engagement in the study and Figure 3 shows relative ranking of barriers to eventual recruitment in the study. The top barriers to initial engagement reported by at least one-third of study staff members were clinicians not referring their patients to the study (66.7%, n = 6), potential subjects not being aware of the study (55.5%, n = 5), lack of phone access among clinical trial subject candidates (55.5%, n = 5), transportation difficulties (44.4%, n = 4), schedule conflicts (44.4%, n = 4), stress from family or living circumstances (44.4%, n = 4), and chronic physical health problems (33.3%, n = 3).
Barriers to Initial Participant Engagement
Barriers to Eventual Enrollment
The top barriers to eventual enrollment reported by at least one-third of study staff survey respondents overlapped substantially with barriers to initial engagement and included schedule conflicts (55.5%, n = 5), chronic physical health problems (55.5 %, n = 5), transportation difficulties (44.4%, n = 4), stress from family or living circumstances (33.3%, n = 3), and acute health issues (33.3%, n = 3).
Strategies to Address Barriers at the Macro-, Mediator-, and Micro-Levels
Table 1 shows strategies that study team members used to address environmental/social (macro-), health care system/public services (mediator-), or personal (micro-) level barriers to engagement or recruitment. The majority of barriers identified by at least 1/3 of survey respondents were at the mediator-level (lack of phone access, lack of transportation access, and clinicians not referring their patients to the research team) and the micro-level (family or life stressors, acute, and chronic physical health problems).
Research Team Approaches to Address Macro-, Mediator-, and Micro-Level Barriers to Clinical Trial Initial Engagement and Eventual Enrollment.
Not being aware of the research study was the key macro-level barrier. The team used multiple methods of outreach to inform participant candidates about the study including in-person and virtual community events. Most virtual events took place when travel and social contact restrictions were in place during the COVID-19 pandemic and events transitioned to an in-person format once these restrictions had lessened or resolved.
Key mediator-level barriers were lack of phone access, lack of transportation access and clinicians not referring their patients. As with approaches to address barriers at the macro-level, a multi-pronged strategy was employed by the study team. Contact approaches other than phone were used including in-person meetings (for those who lived in the regions) often at locations that were accessible to men in their neighborhoods (e.g., a nearby public library). To help with transportation challenges the team provided the availability of ride-share services, bus fare or parking vouchers. All research procedures that could be done remotely were done using videoconferencing or by phone. This was especially helpful during the early days of the COVID-19 pandemic. In some cases, the research team was able to piggyback research visits in with same-day clinical appointments. This did require some coordination with clinical teams to find appropriate settings to meet with study participants.
To help generate clinician referrals the team met with care providers in multiple different formats including medical staff meetings, Grand Round presentations at an academic medical center and email and phone/text follow-up. For providers who gave permission and who were in the same medical practice as the study PI, research staff checked provider schedules and sent provider prompts to discuss possible referral with their patients. A weekly clinical trial focused email was sent to all local neurology clinicians with study inclusion/exclusion criteria and referral instructions.
At the micro-level, the study team also used multiple approaches to help participants manage barriers related to stress or family life circumstances and acute or chronic health conditions that made it challenging to complete screening and enrollment procedures. Being able to conduct assessments remotely was critical to being able to reach this community of men who often had significant functional impairments, such as difficulty walking or in speech fluency because of the effects of stroke.
When possible/available, the staff encouraged involvement of care partners. Some men had a family member that helped them with communication and study procedures. The study team members were also flexible in scheduling and, when appropriate, spaced completion of assessments over time when needed for those who were fatigued or felt unwell. In some cases, study team members helped provide community resource information (such as contact information for a local foodbank). In some cases, the team did home visits for individuals who were unable to leave their home. The team had a set of safety procedures for home visits including always having two study team members present during these assessments.
Discussion
RCTs are conducted to generate evidence for efficacy and safety for interventions prior to being implemented across large populations. Recruitment is a critical part of the clinical trial process, and often one of the most challenging. Many trials fail to meet their target sample size due to delays in recruitment (Johnson, 2015). Clinical trial enrollment may be particularly challenging in marginalized communities and the COVID-19 pandemic greatly exacerbated enrollment difficulties and clinical trial completion (Andraska et al., 2021). This report describes a research team’s ultimately successful effort to enroll 160 Black men with stroke or TIA. A survey approach and follow-up discussion by study team members identified macro-, mediator-, and micro-level barriers and facilitators to engagement and participation among this population and approaches that were used by the research team to minimize these barriers. This study team’s experience in engaging a community that has been historically “hard-to-reach” may offer value to future research groups.
The study staff used a variety of methods to overcome or work around barriers to both initial engagement and eventual recruitment of participants. There was no single “one size fits all” approach and staff would often try a sequence of different types of attempts to reach/engage individuals including in-person, virtual and other types of outreach. A recently published scoping literature review of strategies to improve enrollment in mental health clinical trials (another potentially “hard-to-reach” community; Iflaifel et al., 2024) along with a qualitative study on this topic (Iflaifel et al., 2023) similarly suggests the use of greater consideration should be given to online or mixed (offline/in-person + online/virtual) recruitment strategies that adopt a tailored approach, and offer flexibility and choice.
The main macro (environment/social) barrier in the study presented here was individual’s not being aware of the research study as a possible opportunity. Because the COVID-19 pandemic curtailed in-person visits to the clinical or medical center, opportunities for face-to-face engagement were limited. The team distributed IRB-approved flyers describing the study via posting in locations where older men might be likely to see the information. Over time, as recruitment continued, the team also used paid advertising and worked with a clinical trial recruitment vendor that had a broad national audience.
A particularly useful approach to reaching individuals who were home-bound was to solicit their interest via the EHR, either by reaching out directly to individuals via the EPIC MyChart patient communication portal, or by sending them a letter via U.S. postal service and following up by phone. During the 5-year course of the trial, an increasing proportion of individuals registered for an EHR portal as an avenue of communication. By the end of the study, outreach via the EHR was one of the most productive referral sources.
There were two key areas for mediator-level barriers, those having to do with access problems (transportation and phone availability) and the fact that clinical providers were often too busy/stretched to make patient referrals, even though they were generally supportive of the study and welcomed having their patients in the TEAM program. The study team quickly pivoted to virtually-administered research procedures during the initial months of the COVID-19 pandemic and the ability to conduct activities remotely was critical to successful enrollment. A recent review on health management strategies for stroke survivors during the rehabilitation period (Zhu et al., 2024) suggests that remote delivery using telehealth technology is an effective and efficient method of delivering care to people who have survived stroke and that these technologies are increasingly used by individuals in stroke recovery. The review also suggests that telehealth can overcome multiple physical and logistic challenges and improve access and convenience, and can reduce stress, thereby improving patient satisfaction. However, it is important to note that not everyone is able to benefit from technology-facilitated health communication, especially older people and those who may have problems with internet access (Gajarawala & Pelkowski, 2021). On the topic of clinician outreach, the study team spent considerable time reaching out to clinicians with referral solicitations. Presentations (especially those conducted in virtual formats) had very limited success. Referrals were more likely to be obtained from clinicians who were actively engaged in the inpatient setting with in-person access to patients with acute stroke.
At the micro-level, the study team also used multiple approaches to help participants manage barriers related to stress or family life circumstances and acute or chronic health conditions. Being able to conduct assessments remotely was critical to being able to reach this community of men who often had significant functional impairments, such as difficulty walking or in speech fluency because of the effects of stroke. When possible/available, the staff encouraged involvement of care partners who facilitated activities, such as getting set up for videoconferencing or completing study forms.
While the study team’s experience in conducting a clinical trial enrolling Black men with stroke or TIA could help inform successful approaches to engage hard-to-reach communities there are some important caveats in interpreting our findings. First, while the study was conducted mainly during the COVID-19 pandemic, the study team survey evaluated barriers broadly and did not attempt to parse out barriers that were worsened specifically due to COVID-19. Some unique challenges and solutions that the study team identified may be less relevant in the post-COVID era. In addition, our study had only minimal exclusion criteria, given that the experimental intervention being tested was an adjunct to standard stroke treatment. Outreach approaches for clinical stroke trials with narrower entry criteria (e.g., drug trials) may need to be different. Our survey also did not attempt to parse out differences in outreach for men with stroke versus TIA, although some of the recommendations on managing engagement of individuals with acute or chronic health challenges might be more relevant for men with stroke as opposed to TIA. It should be noted that attribution of weather as a possible barrier to research is not aligned with the prior literature and could potentially obscure other reasons for research hesitancy. It is possible that medical mistrust might make it especially likely that someone could use bad weather as a convenient reason to not attend. Similarly, filtering input on research barriers through study staff could introduce bias and may have diluted or altered the information received. This is a limitation of the study methods and future work should include input from study participants on barriers and how these might be addressed. Finally, as was seen over time during the COVID-19 pandemic, health care systems continue to have an increased emphasis on technology-facilitated care and outreach approaches and in-person methods may be less relevant.
In conclusion, there are multiple barriers to enrolling Black men with stroke or TIA into clinical trials. Successful outreach and engagement methods need to address barriers to research at the macro-, mediator-, and micro-levels, and use mixed-method (in-person + virtual) approaches targeted to the specific challenges individuals may be facing.
Footnotes
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Sponsorship: This project was supported by a research grant from the U.S. National Institutes of Health IR01NR018023-01A1 (PIs Sajatovic & Still).
Declaration of Conflicting Interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Sajatovic: Research grants within past 3 years: Neurelis, Intra-Cellular, Merck, Otsuka and Alkermes. Consultant in the past year: Otsuka, Lundbeck, Janssen, Teva, and Medscape. Royalties in the past year: Springer Press, Johns Hopkins University Press, Oxford Press, and UpToDate Compensation for preparation of/participation in CME activities past year: American Physician’s Institute (CMEtoGo), American Epilepsy Society, and Clinical Care Options. Authors Conroy, Lute, Krehel-Montgomery, Surdam, Fuentes-Casiano, Whitlow, Broadnax, Moore, Still, and Sundararajan have no conflicts to report.