The UnBIASED Project Overview
The UnBIASED Project is investigating a new approach to address hidden healthcare bias – based on patients’ race, gender, sexual orientation – by improving patient-doctor communication in primary care. This approach monitors body language for signs of bias and provides feedback to raise awareness of patients and doctors for opportunities to adjust their communication style. This project is being led by Dr. Andrea Hartzler at the University of Washington and Dr. Nadir Weibel at the University of California, San Diego.
-
I worked as an Undergraduate Research Assistant from January 2020 to March 2022, where my key roles were:
Qualitatively analyzing generative user interviews with patients and primary care providers to understand the pain points and needs of these groups
Creating evaluative concept testing session protocols to design a provider-facing tool
Synthesizing results into manuscripts for publication and presentations that were published and presented at the American Medical Informatics Association Annual Proceedings
-
We had a massive team across the University of Washington and the University of California, San Diego. I work directly with Dr. Andrea Hartzler, Dr. Wanda Pratt, Dr. Janice Sabin, Reggie Casanova-Perez, Calvin Apodaca, Erin Beneteau, Connie Yang, Lisa Dirks, Naba Rizvi, Deepthi Mohanraj, and Harshini Ramaswamy. You can meet the rest of the team here.
-
My key accomplishments on the UnBIASED Project are:
Giving an invited talk at the American Medical Association 2021 Annual Proceedings on the need for the consideration of intersectionality in medical informatics solutions
Having a podium abstract titled “Toward Patient-Centered Informatics Solutions: The Role of Intersectionality” published in the American Medical Association 2021 Annual Proceedings
Being the third author on a paper titled “Broken down by bias: Healthcare biases experienced by BIPOC and LGBTQ+ patients” that was published in the American Medical Association 2021 Annual Proceedings
Giving an invited talk at the University of Washington’s Undergraduate Research Symposium on the experiences of bias faced by BIPOC and LGBTQ+ people in healthcare
Being a co-author on five other manuscripts that are under review at internationally recognized journals and publications
-
The key skills and methodologies I used during this project are:
Semi-Structured Generative User Interviews
Thematic Qualitative Analysis
Protocol Development
Participant Recruitment
Concept Testing
Scientific Writing
Presentation of Findings
The Details
The UnBIASED Project is a five-year grant with three aims; 1) develop and validate a Social Signal Processing (SSP) model to detect hidden bias in patient-provider communication, 2) design SSP feedback that effectively conveys hidden bias to patients and providers, and 3) evaluate the efficacy of SSP technology in controlled and real-world settings. I worked on Aim 2.
Patient & Provider Interviews
We conducted 25 semi-structured interviews with patients and 21 interviews with providers. Our patient participants self-identified as being Black, Indigenous, Persons of Color (BIPOC), and/or Lesbian, Gay, Bisexual, Transgender, or Queer (LGBTQ+), while our provider participants identified as serving patients who are part of these groups.
Our goal for patient interviews was to explore their experiences of bias in healthcare settings. During the first half of the interview, we asked participants to “Tell us about a time when you had an interaction with a doctor where you felt not heard, disrespected, or made uncomfortable?” and the second half of the interview asked them “If you had the power to change the experience you described, what would you change?”
Provider interviews followed a similar format, asking first “Can you tell us about your observations or a colleague's observations of patients' and providers’ experiences of racism, sexism, or other ‘isms’ related to bias and unequal treatment associated with race, ethnicity, gender, or sexual orientation?” followed by, “Given those experiences, we would like your help to brainstorm ways that communication between patients and providers could be better in the future.”
Provider Concept Testing
Based on the problems identified by patients and providers and their proposed solutions, we began generating three different low-fidelity prototypes that would serve as a provider-facing solution to help providers overcome potential implicit bias that would impact their communication with patients.
The three prototypes are 1) Guided Reflection, 2) Patient Dashboard, and 3) Real-Time Nudges. These prototypes reflected different timelines providers and patients suggested for giving feedback to providers about their communication, as well as different potential modalities.
Following the development of these prototypes, we have been concept testing them with providers to collect feedback about improvements that can be made and potential workflow or institutional barriers that would prevent them from implementing solutions like these into their clinical practice.
Accomplishments
AMIA Invited Talk
I gave an invited talk at the American Medical Association 2021 Annual Proceedings on the need for the consideration of intersectionality in medical informatics solutions. The link to my presentation is above.
AMIA Podium Abstract
I first-authored a podium abstract titled “Toward Patient-Centered Informatics Solutions: The Role of Intersectionality” published in the American Medical Association 2021 Annual Proceedings.
AMIA Full Paper
I was the third author on a paper titled “Broken down by bias: Healthcare biases experienced by BIPOC and LGBTQ+ patients” that was published in the American Medical Association 2021 Annual Proceedings.
UW Undergraduate Research Symposium
I gave an invited talk at the University of Washington’s Undergraduate Research Symposium on the experiences of bias faced by BIPOC and LGBTQ+ people in healthcare. The link to my presentation is above.