Tackling healthcare providers bias: a systematic review of interventions with implications for inclusive clinical research

Maria Grammoustianou, Stefano Maccarone, Emma Gillanders, Gwenn Menvielle, Marie Preau, Stepheline Ginguene, Sibille Everhard, Cyrille Delpierre, Francis Guillemin, Ines Vaz-Luis and Maria Alice Franzoi

Summary
Background Diverse population inclusion in clinical research is essential for generating high-quality, generalizable data that promotes health equity. However, clinical research often underrepresents elderly individuals, minoritised ethnicities, and marginalized socio-demographic groups. Implicit and explicit biases among healthcare providers can create barriers to equitable participation in clinical trials, affecting patient care, clinical outcomes, and provider-patient trust. Bias-focused interventions have been reported, yet methodological inconsistencies and limited generalizability of studies hinder their effectiveness. This systematic review synthesizes global evidence on interventions designed to address healthcare provider biases, identifying intervention components, trends,
outcomes assessed and gaps in the current literature.
Methods Following PRISMA guidelines, a systematic review of English-language peer-reviewed studies published between 2004 and June 2025 was conducted. The review included both quantitative and qualitative studies assessing interventions targeting healthcare provider biases or cultural competency. A comprehensive search of
Scopus, Cochrane, and PubMed MEDLINE databases identified 110 studies. Study quality, risk of bias, and outcomes were assessed using the RoB-2 tool for randomized trials and ROBINS-I for non-randomized studies (PROSPERO ID: CRD42024515985).
Findings Among the 103 studies assessed for efficacy, 75.7% reported positive outcomes. The majority of studies (69%)
were conducted in the U.S. Most interventions combined educational and experiential methods, with the majority
being brief (up to 3 h). An increase in the use of digital delivery methods was observed in the past 5 years, with 45% of studies incorporating digital components, particularly after 2020. Key outcomes included explicit bias (53.6%), cultural competency (31.2%), and implicit bias (31.8%), measured mainly by heterogeneous and non-standardized instruments. Notably, only 38 studies (34.5%) assessed actual healthcare provider behavior, and 30 studies (27.2%)
evaluated long-term efficacy. High risk of bias and methodological challenges impacted the quality of evidence.
Interpretation This review highlights positive short-term effects of interventions on implicit bias and cultural competence but suggests future research could benefit from more rigorous trial designs, standardized metrics for measuring efficacy, and actual provider behavior. It also emphasizes the need for more implementation research to identify barriers and facilitators for scaling interventions to ensure their sustainability and broader
impact on diversity, equity, and inclusion in healthcare and research.
Funding This project is funded by WeShare, DIVERSIFY and SIRIC Epicure projects. WeShare program is supported by the French State, managed by the Agence Nationale de la Recherche under the Investissements Program integrated into France 2030, with reference number: ANR-21-ESRE-0017. DIVERSIFY is funded by La Ligue Contre
le Cancer. SIRIC Epicure is funded by INCA-DGOS-Inserm-ITMO Cancer 18002.