Defensive Responses to Implicit Association Tests and Bias Awareness in an Implicit Bias Mitigation Training

Janice Sabin,1 Eric Mick,2 Ethan Eisdorfer,3 Majid Yazdani,4 Maria M Garcia,4 Janet Fraser Hale,5 Jill Terrien,5 Geraldine Puerto,2 Vennesa Duodu,2 Valerie Zolezzi-Wyndham,6 Joshua Rumbut,2 Joanne Calista,7 Olga Valdman,3 Stacy Potts,3 Jeroan Allison,2 Michele Pugnaire,3 Jennifer Tjia2 1Department of Biomedical Informatics and Medical Education, University of Washington, Seattle, WA, USA; 2Department of Population and Quantitative Health Sciences, UMass Chan Medical School, Worcester, MA, USA; 3Department of Family Medicine, UMass Chan Medical School, Worcester, MA, USA; 4Department of Medicine, UMass Chan Medical School, Worcester, MA, USA; 5Tan Chingfen Graduate School of Nursing, UMass Chan Medical School, Worcester, MA, USA; 6Promoting Good, Worcester, MA, USA; 7Center for Health Impact, Worcester, MA, USACorrespondence: Jennifer Tjia, Department of Population and Quantitative Health Sciences, UMass Chan Medical School, 55 N Lake Avenue, AS6-2065, Worcester, MA, 01605, USA, Tel +1 774-455-3538, Email jennnifer.tjia@umassmed.eduBackground: Implicit bias education that utilizes the Implicit Association Test (IAT) to raise self-awareness of bias can induce defensiveness.Objective: To describe clinical learners’ bias awareness, self-perceptions of bias relative to colleagues (better-than-average), implicit and explicit biases and defensive response to the IATs.Design: Cross-sectional study.Participants: Internal medicine and family medicine residents, and Doctor of Nursing Practice students at a public medical and nursing school affiliated with a disproportionate share hospital and who completed an implicit bias recognition and mitigation educational program (including didactics, IATs, and communication skills training and practice with standardized patients) in 2018– 2019.Main Measures: We measured implicit and explicit attitudes and stereotypes, reactions to IAT results (defensive or not defensive), better-than-average perceptions, bias awareness and participants’ characteristics. We examined associations between defensive responses to the IAT and participant characteristics, self-reported explicit biases, bias awareness within self, society, and healthcare, and IAT scores.Key Results: Of N=61 respondents, 57% were female and 59% White. We found moderate implicit bias favoring White people versus Black people, weak bias favoring White people versus Hispanic/Latinx people and moderate bias favoring White people on both race and ethnicity medical compliance stereotype IATs. Participants demonstrated awareness of bias in society and healthcare, but not within self. Eighteen percent were defensive regarding their IAT results. Perceptions of own bias (self) were always of their having less bias than their colleagues, and they were better-than-average. There were no statistically significant associations between IAT scores and participant demographics and no interaction effect between implicit bias, defensiveness and better-than-average scores.Conclusion: Clinical learners hold moderate implicit biases, believe they have less bias than others, and almost 1-in-5 have a defensive response to IAT feedback. It is important to design implicit bias educational interventions to include reflection on personal bias and provide a safe environment to minimize defensiveness.Keywords: implicit bias, defensiveness, implicit association test, educational program, healthcare bias