Evaluating clinician emotion regulation during a serious illness conversation in oncology using multimodal assessment: A pilot study

ABSTRACT

Background: Clinician emotion regulation (ER), self and the patient's, likely moderates successful prognostic discussions with patients, yet challenges around its evaluation limit its investigation. We performed a pilot study to develop and assess an experimental framework that uses multimodal assessment (self-report, observer, and biometric data) to measure clinician ER during a simulated, serious illness conversation (SIC). Method(s) We developed our experimental framework in four steps: 1) drafted a patient case and assessment framework;2) refined the data collection strategy using a multidisciplinary research team;3) trained our actor;and 4) iteratively piloted the case to optimize data collection. For the assessment, we conducted a cross-sectional, case series pilot study with physicians trained in SIC to assess its feasibility and acceptability, defined a priori as an enrollment rate >60% of approached clinicians, > 90% completion rate of survey items, < 20% missing data from wearable heart rate variability (HRV) sensors. To characterize clinician ER strategies, we analyzed the visit dialogue, physician interviews performed while watching the visit back, and physician SIC documentation generated post visit. We used a hybrid approach of inductive and deductive coding and theme development based on preexisting ER theory. Result(s) The development phase yielded two major modifications 1) abandoned use of AppleWatch since it did not provide continuous HRV measurement;and 2) used telehealth with video given context of COVID-19 pandemic. We approached 12 physicians and 11 enrolled, of which 5 were female and 10 white, 5 were medical oncologists, and 6 specialty palliative care physicians. All participants completed all study survey items. The results of our three HRV sensors were as follows Empatica E4 and Polar H10 met our pre-specified HRV collection in all 11 resting tasks and SIC encounters, and the Scoche R24 the benchmark in 7/11 resting tasks and 4/11 of simulated encounters. Preliminary qualitative analysis suggests investigators can characterize clinician use of intrapersonal and interpersonal ER strategies. Conclusion(s) The use of multimodal assessment of clinician ER in a simulated, telehealth SIC visit was acceptable and feasible.