Dyadic Trust Measurements Applied to Assessment in a Clinical Learning Setting

Abstract
Trust between members of medical teams is key to the effective delivery of patient care. In teaching hospitals, trust between supervisors and trainees establishes roles within which trainees can develop skills and competencies safely via a collaborative process. This training involves frequent exchanges between different pairings of supervisors-trainee dyads. The degree of trust established within a dyad is unique, depending both on each individual’s latent traits, and on the interaction between them. To measure these dyadic properties, we propose a Dyadic Item Response Theory (dIRT) model for measuring interactions of pairs of individuals when the responses to items represent the actions (or behaviors, perceptions, etc.) of each individual (actor) made within the context of a dyad formed with another individual (partner). The dIRT model generalizes both Item Response Theory (IRT) models for measurement and the Social Relations Model (SRM) for dyadic data. The responses of an actor when paired with a partner are modeled as a function of not only the actor's inclination to act (i.e. tendency to trust) and the partner's tendency to elicit that action (i.e. trustworthiness), but also the unique relationship of the pair, represented by two directional, possibly correlated, interaction latent variables. Generalizations are discussed, such as accommodating triads or larger groups. Estimation of the dIRT model is performed using Markov-chain Monte Carlo implemented in Stan, making it straightforward to extend the model in various ways. We show how the basic dIRT model can be extended to accommodate latent regressions and joint modeling of dyadic data and a distal outcome. We apply our proposed approach to speed-dating data and find new evidence of pairwise interactions between participants, describing a mutual attraction that is inadequately characterized by individual properties alone. Additional applications include collaborative problem solving, professional relationships, and other co-constructed latent traits. We conclude by discussing and inviting your feedback on our ongoing development of constructs for measuring bidirectional trust in supervisor-trainee dyads, derived from our qualitative work in clinical settings.

Stephanie Tsoi is a first year pediatrics resident at UCSF Benioff Children's Hospital. She graduated from UCLA Medical School in 2018 and did her undergraduate degree in Human Biology at UCSD. After residency, she plans to work with the pediatric population in the inpatient setting. Outside of clinical medicine, her interests include quality improvement of the hospital system, healthcare policy and advocacy, and the teaching process of medical students and residents through academic medicine.
Brian C Gin is an assistant clinical professor in the division of Pediatric Hospital Medicine and associate program director for the Pediatric Hospital Medicine Fellowship at the UCSF School of Medicine. He did his MA in Education and BS/PhD in Chemistry at UC Berkeley, and his MD at UCSF. Aside from his role as a pediatric hospitalist attending, his research interests include statistical modeling of emergent phenomena, examining the role of trust in care delivery and medical education, and supporting the wellbeing of clinical learners.

Date: 
Tuesday, April 16, 2019 - 2:00pm
Building: 
2121 Berkeley Way West
Room: 
1217