At the intersection of trust and mistrust: A qualitative analysis of motivators and barriers to research participation at a safety‐net hospital

Abstract Introduction The underrepresentation of Black, Indigenous, and People of Color (BIPOC) individuals in healthcare research limits generalizability and contributes to healthcare inequities. Existing barriers and attitudes toward research participation must be addressed to increase the representation of safety net and other underserved populations. Methods We conducted semi‐structured qualitative interviews with patients at an urban safety net hospital, focusing on facilitators, barriers, motivators, and preferences for research participation. We conducted direct content analysis guided by an implementation framework and used rapid analysis methods to generate final themes. Results We completed 38 interviews and identified six major themes related to preferences for engagement in research participation: (1) wide variation in research recruitment preferences; (2) logistical complexity negatively impacts willingness to participate; (3) risk contributes to hesitation toward research participation; (4) personal/community benefit, interest in study topic, and compensation serve as motivators for research participation; (5) continued participation despite reported shortcomings of informed consent process; and (6) mistrust could be overcome by relationship or credibility of information sources. Conclusion Despite barriers to participation in research studies among safety‐net populations, there are also facilitators that can be implemented to increase knowledge and comprehension, ease of participation, and willingness to join research studies. Study teams should vary recruitment and participation methods to ensure equal access to research opportunities. Patient/Public Contribution Our analysis methods and study progress were presented to individuals within the Boston Medical Center healthcare system. Through this process community engagement specialists, clinical experts, research directors, and others with significant experience working with safety‐net populations supported data interpretation and provided recommendations for action following the dissemination of data.

by an implementation framework and used rapid analysis methods to generate final themes.
Results: We completed 38 interviews and identified six major themes related to preferences for engagement in research participation: (1) wide variation in research recruitment preferences; (2) logistical complexity negatively impacts willingness to participate; (3) risk contributes to hesitation toward research participation; (4) personal/community benefit, interest in study topic, and compensation serve as motivators for research participation; (5) continued participation despite reported shortcomings of informed consent process; and (6) mistrust could be overcome by relationship or credibility of information sources. (BIPOC) is present in multiple arenas, and the medical field is no exception. [1][2][3][4] In 1993, the National Institutes of Health acknowledged the need to increase diversity in healthcare research when they mandated the inclusion of women and marginalized groups in NIH-funded clinical trials. 5 However, there is a clear lack of progress in the implementation of this policy as nearly 20 years later, BIPOC individuals are still underrepresented in healthcare-related research studies. [6][7][8] For example, of more than 300,000 enrollees in trials submitted to the US Food and Drug Administration between 2008 and 2017, a mere 4% represented the Black population. 9 Clinical studies that exclude large segments of the population are not generalizable and do not lead to a knowledge base that is inclusive of the entire population. Without inclusivity, the medical system cannot deliver on the promise of tailored medicine or targeted therapies. In addition, within this infrastructure, underrepresented communities do not share equally in the benefits of research which in turn exacerbates health inequities. [10][11][12][13][14][15] This issue became clear early in the coronavirus disease 2019 (COVID- 19) pandemic when US death rates among Black and Latino individuals were threefold higher compared with White individuals. 16 This trend is exhibited more chronically in diseases such as breast cancer, where African American women have higher mortality rates than their White counterparts despite having lower incidence. 17 The consequences of underrepresentation call for immediate action and an increased focus on the inclusion of BIPOC individuals in healthcare research. [18][19][20][21][22] While the need for better inclusion of BIPOC populations in healthcare research is clear, a myriad of barriers to achieving sufficient representation exist and lead to challenges with the recruitment and retention of samples that are adequate for making race-stratified conclusions. 6,15,[23][24][25][26] Barriers include, but are not limited to, language/literacy, socioeconomic status, the complexity of informed consent, and limited access to technology. 23,25,[27][28][29][30] Further, systems of oppression and historical maltreatment of minoritized communities have resulted in a culture of mistrust in scientific institutions that contributes to hesitancy toward participating in healthcare research. 8,28,31 Safety-net hospitals and clinics are facilities that primarily serve patients from traditionally marginalized communities, including individuals of lower socioeconomic status, who are uninsured, who receive Medicaid benefits, and who identify as BIPOC individuals. 23 34 Nearly 75% of BMC patients identify as medically underserved, and approximately 65% utilize public insurance. 34,35 The BMC patient population reflects the racial and ethnic diversity of Boston, where approximately 30% of patients do not speak English as a primary language, about 30% identify as Black, and about a quarter identify as Latino. [34][35][36] Additionally, 7% of patients at BMC experience unstable housing or homelessness. 35 As an academic medical centre focused on caring for a safety-net population, we know that research is important to help us provide high-quality, equitable, and evidence-based care to our patients.
Many of our faculty conduct research with our patient population. [37][38][39] Additionally, BMC has an active clinical trials network office that focuses on conducting studies with our patients. Research opportunities are made available to patients through a variety of platforms including direct referral/conversations during clinical visits, flyers posted in clinic exam rooms and waiting rooms, and opportunities presented on the website.
We created and used a semi-structured interview guide to gain better insight into participants' previous experiences with research, understanding of the research process, and experiences with informed consent. We also encouraged participants to reflect on the process of learning about research studies, preferences for future research, and factors that influenced the decision to, or not to, participate in a research study.
We recruited participants from a pool of 74 individuals who had previously either participated in a randomized control trial (APPEX) or qualitative interviews assessing experiences with the hospital discharge process (CIIS Equity) at BMC and had agreed to be contacted for future research opportunities. 37,40 Of the 74 eligible individuals, 74% (n = 55) were able to be contacted, and 60% (n = 44) agreed to participate in the current study. For those willing to participate, we scheduled an interview time and sent informed consent information via email or mail.
We conducted interviews with 38 individuals by telephone in either English or Spanish. We recorded all interviews and documented verbal consent at the time of the interview. We provided participants with a $50 gift card as compensation for their participation. The Boston Medical Center and Boston University Medical Campus Institutional Review Board approved all study procedures.

| Data analysis
A professional transcription company transcribed verbatim all audio recordings; interviews completed in Spanish were then translated into English for analysis. After transcription, we reviewed all transcripts against the audio recordings for clarity and accuracy.
We conducted a directed content analysis guided by the Consolidation Framework for Implementation Research (CFIR). 41 CFIR is a comprehensive framework that incorporates 39 constructs across five broad domains thought to influence implementation success: [1] intervention characteristics, [2] inner setting, [3] outer setting, [4] characteristics of the individuals, and [5] implementation process. 41  Once coding was complete, we used rapid qualitative data analysis methods to produce expedient and actionable results. [42][43][44] To complete rapid analysis, we followed the following steps: created analysis summaries for each transcript; rated analysis summaries with positive/negative and weak/strong ratings to determine the influence of each summary on participation behaviours; aggregated summary analyses in each thematic area; and generated final themes using the aggregated summaries. At each step of the rapid analysis process, we selected illustrative quotes to ensure conceptual clarity was maintained. Consistent with the initial coding, two study team members

| Characteristics of the study population
A total of 38 individuals participated in the current study; of these, 36 were interviewed in English and two in Spanish. Among study participants, 53% identified as Black/African American and 29% identified as White. Just under 25% additionally identified as Latino.
The mean age of participants was 51.6 (SD 18.4) years, with a range of 18-83. Just over 60% of the sample was female (Table 1).

| Overview of themes
We identified six major themes regarding research participation

| DISCUSSION
We employed qualitative methods and the CIFR framework to identify factors that influenced decisions to participate in research studies from the perspective of diverse safety-net hospital patients.
One notable finding is that the concerns about research among patients in a safety-net setting are similar to those of research participants more generally. 45,46 Our participants reported barriers to participation, such as securing transportation to a study site and commitment to a study of long duration. However, participants indicated that providing financial compensation or other support for participation, such as reimbursement for transportation, made participation easier and increased their willingness to participate.
Participants indicated additional facilitators to include recruitment in familiar settings, in-person recruitment, and increasing factors of convenience, such as allowing for remote participation. While catering to individual preferences and creating opportunities for high-quality interaction was found to increase the likelihood of participation, the current findings highlight that a critical ingredient for study recruitment and retention is trust. Lastly, in regard to obtaining informed consent, participants reported this process to be challenging and recalled little detail, feeling that upon agreement for participation, they had less than a full understanding of study details.
Our findings are consistent with previous research which found that minority populations experience increased challenges that limit their ability to participate in research studies. 27 Our results are also consistent with reported facilitators that improve willingness to participate, such as utilizing familiar settings, in-person recruitment, and capitalizing on convenience, such as with remote participation. 15