The construction of power in family medicine bedside teaching: a video observation study

INTRODUCTION: Bedside teaching is essential for helping students develop skills, reasoning and professionalism, and involves the learning triad of student, patient and clinical teacher. Although current rhetoric espouses the sharing of power, the medical workplace is imbued with power asymmetries. Power is context-specific and although previous research has explored some elements of the enactment and resistance of power within bedside teaching, this exploration has been conducted within hospital rather than general practice settings. Furthermore, previous research has employed audio-recorded rather than video-recorded observation and has therefore focused on language and para-language at the expense of non-verbal communication and human-material interaction. METHODS: A qualitative design was adopted employing video- and audio-recorded observations of seven bedside teaching encounters (BTEs), followed by short individual interviews with students, patients and clinical teachers. Thematic and discourse analyses of BTEs were conducted. RESULTS: Power is constructed by students, patients and clinical teachers throughout different BTE activities through the use of linguistic, para-linguistic and non-verbal communication. In terms of language, participants construct power through the use of questions, orders, advice, pronouns and medical/health belief talk. With reference to para-language, participants construct power through the use of interruption and laughter. In terms of non-verbal communication, participants construct power through physical positioning and the possession or control of medical materials such as the stethoscope. CONCLUSIONS: Using this paper as a trigger for discussion, we encourage students and clinical teachers to reflect critically on how their verbal and non-verbal communication constructs power in bedside teaching. Students and clinical teachers need to develop their awareness of what power is, how it can be constructed and shared, and what it means for the student-patient-doctor relationship within bedside teaching.

DeepCare: a deep dynamic memory model for predictive medicine

Personalized predictive medicine necessitates the modeling of patient illness and care processes, which inherently have long-term temporal dependencies. Healthcare observations, recorded in electronic medical records, are episodic and irregular in time. We introduce DeepCare, an end-to-end deep dynamic neural network that reads medical records, stores previous illness history, infers current illness states and predicts future medical outcomes. At the data level, DeepCare represents care episodes as vectors in space, models patient health state trajectories through explicit memory of historical records. Built on Long Short-Term Memory (LSTM), DeepCare introduces time parameterizations to handle irregular timed events by moderating the forgetting and consolidation of memory cells. DeepCare also incorporates medical interventions that change the course of illness and shape future medical risk. Moving up to the health state level, historical and present health states are then aggregated through multiscale temporal pooling, before passing through a neural network that estimates future outcomes. We demonstrate the efficacy of DeepCare for disease progression modeling, intervention recommendation, and future risk prediction. On two important cohorts with heavy social and economic burden -- diabetes and mental health -- the results show improved modeling and risk prediction accuracy.