Recruitment and retention strategies for improving representation in clinical research: A meta-synthesis

The objective was to identify innovative strategies that may increase recruitment and/or retention of groups less represented in chronic disease clinical research. A systematic review was conducted. Inclusion criteria were: (a) NIH-defined racial and ethnic minority groups and clinical research; (b) evidence-based, clinical research recruitment and/or retention strategies involving the leading causes of mortality and morbidity in the United States; (c) conducted in the United States; and (d) qualitative design. Data exploring the strategies were extracted and thematically analyzed. Twenty-seven studies were included. Studies focused on cancer (70%), recruitment (93%), and perspectives from clinicians (63%). The most referenced strategies were education (44%), communication (48%), and community-based participatory research (63%). Critical themes include empowerment, transparency, trust, and sustainability. Strategies must prioritize the community and be implemented sustainably, where cultural humility and community-based participatory research are foundational.

Methods
We adhered to and adapted the Enhancing Transparency in Reporting the Synthesis of Qualitative Research (ENTREQ) guidelines, Enhancing the Quality and Transparency of Health Research (EQUATOR) guidelines, and The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist to report this completed systematic review and meta-synthesis.

Eligibility criteria

Briefly, we included studies with qualitative or mixed methods research designs, were conducted in the U.S., published in English or Spanish in a peer-reviewed journal between 2009 and 2024, and that demonstrated evidence-based recruitment and/or retention strategies for clinical research focused on the leading causes of morbidity and mortality in the U.S. as determined by the Centers for Disease Control and Prevention (CDC). Studies spanning ten years from 2009 were initially selected; this was later updated to include fifteen years from 2009 to reflect the increasing significance and importance of this work during this period. Eligible studies targeted ethnic and racial minorities defined by the NIH. Studies included all ages and used the NIH definition of clinical research.

Information sources and search strategy

The literature search strategy was developed in collaboration with the review team and trained biomedical librarians (NT and AL) at the National Institutes of Health (NIH). The search strategy was created using a combination of text words and the controlled vocabulary terms in the following databases: (PubMed (MeSH) Medical Subject Headings, Embase - EMTREE, and CINAHL subject headings. The search was refined using an iterative process and finalized by the review team members and librarians. For each search strategy, the search terms included these text words and controlled vocabulary when available: underrepresented, minority, racial and ethnic groups, clinical research, and disparities. The following databases were searched: PubMed (National Library of Medicine), Embase (Elsevier), CINAHL Plus (Cumulative Index to Nursing and Allied Health Literature - EBSCOhost), and Web of Science Core Collection (Clarivate Analytics). The following limits were applied using the filters available in each database. The search was limited to human studies only and was limited to studies conducted in the United States. The final search strategy can be found in the S2 Appendix.

Selection process

Covidence systematic review software (Veritas Health Innovation, Melbourne, Australia; www.covidence.org) imported studies and automatically excluded duplicates. All stages of the screening, data extraction, and quality assessment were independently conducted by members of the review team (CJP, JMG, AK, MW, LA, and JGG). The review team was composed of six members.

The review team first screened titles and abstracts to identify studies that met the inclusion criteria. Next, the full texts of studies included during the title and abstract were screened using the same eligibility criteria. Each article was screened by two reviewers and conflicts between reviewers were resolved by consensus discussion with the review team.

Data collection process & data items

Data from each included study were collected by two reviewers using Covidence. The following outcomes of interest were extracted: focus on recruitment, retention, or both; and a description of the evidence-based strategies. We extracted data on study characteristics, including year of publication and condition of interest, including subtype for cancer. Additionally, we extracted characteristics including race and ethnicity, sex assigned at birth if applicable, geography (urban or rural), and role in clinical research (e.g., participant, clinician (i.e., medical or research staff), community leader, etc.).

Study risk of bias assessment

For the quality assessment, we evaluated the following domains: (a) role of the researcher; (b) sampling method; (c) data collection method; and (d) analysis method, which were identified as all criteria met or criteria partially met. We followed the adapted guidelines and conceptual domains of the Critical Appraisal Skills Programme (CASP) quality assessment tool to assess the quality of studies. Two reviewers assessed the risk of bias for each included study and resolved disagreements by consensus discussion with the review team.

Synthesis methods

A thematic synthesis was operationalized for the data analysis, where we analyzed the findings and developed inductive and deductive codes using qualitative synthesis methodologies and established guidelines. The thematic synthesis utilized an iterative process grounded in qualitative thematic analysis methodologies. We initially developed a deductive coding scheme, focusing on direct meaning and content that highlighted evidence-based strategies and direct quotations from study participants. The team discussed and created the codebook, and then each code was defined. Discrepancies or additional deductive codes were added and discussed by the team for consensus. Each study was coded independently by reviewers. Inductive codes were later added to describe high-level interpretation and themes. Both deductive and inductive codes existed in our codebook using this iterative process. Further analysis was conducted where strategies and themes were summarized into a conceptual model emphasizing key elements for the recruitment and retention of racial and ethnic groups historically underrepresented in clinical research.