OBJECTIVE: People who experience marginalization, including Black, Indigenous, People of Color (BIPOC) and Lesbian, Gay, Bisexual, Transgender, Queer, Plus (ie, all other marginalized genders and sexual orientations) people (LGBTQ+) experience discrimination during healthcare interactions, which negatively impacts patient-provider communication and care. Yet, scarce research examines the lived experience of unfair treatment among patients from marginalized groups to guide patient-centered tools that improve healthcare equity. MATERIALS AND METHODS: We interviewed 25 BIPOC and/or LGBTQ+ people about their experiences of unfair treatment and discrimination when visiting healthcare providers. Through thematic analysis, we describe participants immediate reactions and longer-term consequences of those experiences. RESULTS: We identified 4 ways that participants reacted to discrimination in the moment: Fighting, Fleeing, Excusing, and Working Around Bias. Long-term consequences reflect 6 ways they coped: Delaying or Avoiding Care, Changing Healthcare Providers, Self-prescribing, Covering Behaviors, Experiencing Health Complications, and Mistrusting Healthcare Institutions. DISCUSSION: By describing how patients react to experiences of unfair treatment and discrimination, our findings enhance the understanding of health disparities as patients cope and struggle to speak out.To combat these problems, we identify 3 future directions for informatics interventions that improve provider behavior, support patient advocacy, and address power dynamics in healthcare. CONCLUSIONS: BIPOC and LGBTQ+ patients perspectives on navigating unfair treatment and discrimination in healthcare offers critical insight into their experiences and long-term consequences of those experiences. Understanding the circumstances and consequences of unfair treatment, discrimination, and the impact of bias through this patient-centered lens is crucial to inform informatics technologies that promote health equity.

Bias toward historically marginalized patients affects patient-provider interactions and can lead to lower quality of care and poor health outcomes for patients who are Black, Indigenous, People of Color (BIPOC) and Lesbian, Gay, Bisexual, Transgender and Gender Diverse (LGBTQ+). We gathered experiences with biased healthcare interactions and suggested solutions from 25 BIPOC and LGBTQ+ people. Through qualitative thematic analysis of interviews, we identified ten themes. Eight themes reflect the experience of bias: Transactional Care, Power Inequity, Communication Casualties, Bias-Embedded Medicine, System-level problems, Bigotry in Disguise, Fight or Flight, and The Aftermath. The remaining two themes reflect strategies for improving those experiences: Solutions and Good Experiences. Characterizing these themes and their interconnections is crucial to design effective informatics solutions that can address biases operating in clinical interactions with BIPOC and LGBTQ+ patients, improve the quality of patient-provider interactions, and ultimately promote health equity.

Implicit biases may negatively influence healthcare providers behaviors toward patients from historically marginalized communities, impacting providers communication style, clinical decision-making, and delivery of quality care. Existing interventions to mitigate negative experiences of implicit biases are primarily designed to increase recognition and management of stereotypes and prejudices through provider-facing tools and resources. However, there is a gap in understanding and designing interventions from patient perspectives. We conducted seven participatory co-design workshops with 32 Black, Indigenous, People of Color (BIPOC), Lesbian, Gay, Bisexual, Transgender, Queer/Questioning (LGBTQ+), and Queer, Transgender, Black, Indigenous, People of Color (QTBIPOC) individuals to design patient-centered interventions that help them address and recover from provider implicit biases in primary care. Participants designed four types of solutions: accountability measures, real-time correction, patient enablement tools, and provider resources. These informatics interventions extend the research on implicit biases in healthcare through inclusion of valuable, firsthand patient perspectives and experiences.

OBJECTIVES: Implicit bias perpetuates health care inequities and manifests in patient-provider interactions, particularly nonverbal social cues like dominance. We investigated the use of artificial intelligence (AI) for automated communication assessment and feedback during primary care visits to raise clinician awareness of bias in patient interactions. MATERIALS AND METHODS: (1) Assessed the technical performance of our AI models by building a machine-learning pipeline that automatically detects social signals in patient-provider interactions from 145 primary care visits. (2) Engaged 24 clinicians to design usable AI-generated communication feedback for their workflow. (3) Evaluated the impact of our AI-based approach in a prospective cohort of 108 primary care visits. RESULTS: Findings demonstrate the feasibility of AI models to identify social signals, such as dominance, warmth, engagement, and interactivity, in nonverbal patient-provider communication. Although engaged clinicians preferred feedback delivered in personalized dashboards, they found nonverbal cues difficult to interpret, motivating social signals as an alternative feedback mechanism. Impact evaluation demonstrated fairness in all AI models with better generalizability of provider dominance, provider engagement, and patient warmth. Stronger clinician implicit race bias was associated with less provider dominance and warmth. Although clinicians expressed overall interest in our AI approach, they recommended improvements to enhance acceptability, feasibility, and implementation in telehealth and medical education contexts. DISCUSSION AND CONCLUSION: Findings demonstrate promise for AI-driven communication assessment and feedback systems focused on social signals. Future work should improve the performance of this approach, personalize models, and contextualize feedback, and investigate system implementation in educational workflows. This work exemplifies a systematic, multistage approach for evaluating AI tools designed to raise clinician awareness of implicit bias and promote patient-centered, equitable health care interactions.