Remote Physiologic Monitoring of Resident Wellness and Burnout
Remote Physiologic Monitoring of Resident Wellness and Burnout
Resident wellness and physician burnout are under the spotlight more and more as data begins to show that there is a point of diminishing return on the number of hours in training. In 2003, resident work hours were restricted to less than 80 hours per week averaged over 4 weeks. This change was implemented in response to the robust body of evidence that increased work hours leads to decreased sleep, which in turn leads to medical errors and depression. These factors directly and indirectly lead to worse outcomes for patients. In residency, it is difficult objectively to assess when residents are beginning to experience burnout and depression. The investigators propose a study to determine whether tracking of certain heart rate parameters (resting heart rate and heart rate variability) as well as sleep can correlate to subjective assessment of resident wellness, burnout and depression. The investigators will also compare these measures to biomarkers of stress, such as salivary cortisol. The results of this study may lead to improved understanding of what truly causes burnout and may be an eventual target for intervention to help improve short- and long-term outcomes for resident physicians as well as their patients.
Sleep deprivation contributes to workplace burnout, a psychological work-related syndrome characterized by depersonalization, emotional exhaustion and feelings of decreased personal accomplishment [Montgomery, 2019]. Medical residency training is associated with decreased sleep and exercise as well as an increase in burnout, which may also be associated with depression [Kamblach, 2019]. Resident wellness has become a focal point of many residency programs in order to prevent depression and long-term physician burnout. Many previous studies tracking sleep have used self-reporting, which institutes a certain level of bias, and some newer technologies such as FitBit tracking have become more prevalent [Case, 2015; de Zambotti, 2018]. Real-time physiologic metric tracking, such as resting heart rate (RHR) and heart rate variability (HRV), in addition to accurate sleep tracking, could provide a far more accurate and objective assessment of resident wellness [Sekiguchi, 2019]. These metrics have not been compared directly to subjective assessments of wellness, burnout and depression, thus their true value in this realm is unknown [Mendelsohn, 2019; Kamblach, 2018]. However, having an objective assessment of resident wellness, stratified by specific rotation, could help identify, develop, and institute interventions to prevent burnout and depression and improve resident well-being.
Previous studies have attempted to make an association between sleep hours, duty hours, exercise and wellness, burnout, depression; however, they have used primitive forms of physiologic tracking (i.e. counting steps as a surrogate for exercise and self-reporting of sleep), which is likely why the results have been relatively inconclusive [Mendelsohn, 2019; Kamblach, 2018; Poonja, 2018; Basner, 2017; Marek, 2019]. A systematic review and meta-analysis of studies attempting to identify factors associated with greater resident well-being showed that increased sleep and time away from work were the strongest influencers of improved resident wellness [Raj, 2016]. Objective, real-time assessment of sleep may identify a stronger association and the addition of RHR and HRV to this analysis could further validate subjective assessment of wellness.
HRV, or the fluctuation in the time intervals between adjacent heart beats, has never before been used to track resident well-being but it is an established metric for prediction and management of disease states such as heart failure [Jimenez-Morgan, 2017; Goessl, 2017, Shaffer, 2017; Bullinga; 2005; Tsuji, 1996]. HRV has been shown to predict mortality in Heart Failure with reduced Ejection Fraction (HFrEF) and new cardiac events (angina, myocardial infarction, coronary artery disease-related death, or HF) in the Framingham study, and it also correlates with improved hemodynamics in response to beta-blocker therapy for HF [Bullinga; 2005; Tsuji, 1996].
The investigators propose to use the WHOOP strap 3.0 for remote monitoring of residents to determine a relationship between its measured data (RHR, HRV, and sleep duration) and wellness using literature-validated surveys (Maslach Burnout Inventory, Mini-ReZ survey, Physician Well Being Index, Patient Health Questionnaire-9) [Montgomery, 2019; Linzer, 2016; Olson 2019; Kroenke, 2001; Levis, 2019]. There is no published literature or known ongoing studies investigating this relationship Recent studies have, however, validated the WHOOP device for sleep tracking and determined its efficacy to be nearly identical to that of the gold standard of polysomnography (PSG) [Berryhill, 2020]. This study also showed that the precision of HRV measurements using the wearable WHOOP device had less than 10% error when compared to continuous ECG monitoring, as part of PSG.
There is an established relationship between HRV and anticipated stress, quantified by salivary cortisol levels, yet there has not been studies linking salivary cortisol as a marker of stress, to subjective assessments in physicians nor against data from wearable devices. Biomarkers of stress (salivary cortisol and alpha-amylase) will compared at baseline and on different rotation considered to be associated with varying levels of stress (i.e. outpatient clinic and inpatient consult services versus the intensive care unit (ICU) setting) [Dickerson, 2004; Petrakova, 2015]. Saliva samples provided by subjects will allow the investigators to validate the WHOOP device as a novel tool to measure stress by allowing the team to assess the association between HRV and other device metrics and objective stress-based analytes found in saliva (e.g., cortisol and alpha amylase). These results will be correlated with each other and with work hours via duty logging to determine whether specific rotations in medical residency have better or worse objective and subjective metrics; these results will also be correlated to baseline (according to baseline characteristics survey).
Inclusion Criteria:
Exclusion Criteria: