“My doctor didn't give me half of that privilege”: Incorporating Black Patients’ Lived Experiences in Virtual Patients for Racial Bias Mitigation Training

Research Background and Problem Statement

• Identified Problem: Current "virtual patient" designs for medical training are primarily based on the physician's perspective and fail to adequately reflect the lived experiences of Black patients. This oversight results in virtual patients lacking authenticity and may further reinforce stereotypes (e.g., the perception of Black patients as aggressive).
• Importance of the Problem: Racial bias significantly impacts healthcare quality, reducing patient trust, lowering satisfaction with care, and even contributing to health outcome disparities. Improving physician-patient communication, particularly in scenarios involving racial bias, can help mitigate these issues.
• Research Motivation and Related Work:
  • Studies show that physicians' racial biases often manifest in verbal dominance, unequal nonverbal behaviors (e.g., reduced eye contact), and biased medical decisions.
  • While virtual patients have been used as tools to cultivate empathy and cultural sensitivity, most designs have been physician-centric and have not sufficiently incorporated patient experiences.
  • Current technological advocacy emphasizes directly embedding patients' lived experiences into design to enhance the authenticity of scenarios involving racial bias.

Proposed Solution

• Proposed Solution: Develop a virtual patient prototype named EQUITY to assist physicians in identifying and addressing their biases through racial bias training. EQUITY's design is grounded in the real experiences of Black patients, incorporating key elements such as narratives, virtual patient nonverbal and verbal behaviors, and role-playing to enhance physicians' self-reflection.
• Innovations:
  1. Systematically integrating patients' lived experiences into virtual patient design to authentically reflect the complexities of real-world racial bias.
  2. Employing a participatory design approach, gathering real-life experiences from 33 Black patients across six focus groups to drive iterative prototype development.
  3. Introducing the display of patients' unexpressed emotions and thoughts (via "thought bubbles") to further reveal the impact of bias on patients' inner worlds.
  4. Utilizing role-playing to simulate racial bias scenarios, triggering discomfort and fostering profound self-reflection among physicians.
• Implementation Steps:
  1. Collect real-world experiences of Black patients in racial bias scenarios through focus groups.
  2. Develop the virtual patient prototype, including scenario settings and the design of virtual patient verbal and nonverbal behaviors.
  3. Conduct user testing and iterative improvements of the initial prototype to ensure alignment with patients' lived experiences.
  4. Use "thought bubbles" to display patients' nonverbal emotions and reactions.
  5. Train physicians by having them role-play biased physician roles and complete self-reflection questionnaires based on Gibbs' Reflective Cycle.

Research Outcomes

• Specific Outcomes:
  • The design of EQUITY's virtual patient was validated through focus groups and deemed a faithful representation of Black patients' healthcare experiences.
  • Physician evaluations indicated that EQUITY successfully elicited discomfort ("disruptive experiences") and facilitated deep self-reflection.
  • Qualitative and quantitative studies demonstrated that EQUITY improved physicians' understanding of patients' inner worlds and the impact of bias, while also increasing awareness of unconscious bias.
• Advantages Over Existing Solutions:
  • Current virtual patient designs are physician-centric, whereas EQUITY effectively integrates patients' multidimensional experiences and emotions.
  • Through role-playing and thought bubbles, physicians gain a more comprehensive understanding of patients' emotions and the rationale behind their behaviors.
  • By offering two virtual patient scenarios (confrontational and non-confrontational responses to racial bias), EQUITY helps physicians recognize that non-confrontational behavior does not imply a lack of impact on the patient.
• Experimental or Evaluation Results:
  • Black patients in focus groups found the virtual patient's behaviors and responses closely aligned with their own experiences. Iterative design improvements significantly enhanced consistency with patients' lived experiences.
  • Participating physicians reported that role-playing elicited strong discomfort and helped them recognize the potential impact of racial bias on patient interactions.
  • EQUITY heightened physicians' awareness of their verbal and nonverbal behaviors and their assumptions about patients.
• Limitations and Future Directions:
  • The current study only evaluated EQUITY's impact on physicians' self-reflection, without assessing its potential for long-term behavioral change.
  • While the focus was on Black patients, future work could expand to include the experiences of other marginalized groups to enhance the inclusivity of training.
  • Cross-regional studies could explore the global applicability of EQUITY.

Conclusion

EQUITY innovatively incorporates the real-life experiences of Black patients into virtual patient design, leveraging multidimensional interactions and reflective learning mechanisms to effectively support physicians in identifying and mitigating racial bias. This study highlights the importance of integrating minority group experiences into technology design within the HCI field, offering a scalable participatory design and educational training model that warrants further application and replication in other domains.