SCH: Wearable Augmented Prediction of Burnout in Nurses: A Synergy of Engineering, Bioethics, Nursing

Project: Research project

Project Details


Program Director/Principal Investigator: Athreya, Arjun, Prasanna PROJECT DESCRIPTION I. MOTIVATION: A REAL-WORLD CASE STUDY Registered nurse (RN) Armstrong is in the middle of her 12- hour shift on a busy medical-surgical floor in an urban tertiary- care hospital. Today, she is in charge of 8 post-surgical patients, 4 of whome are post-surgical. Two RNs on the unit are on sick leave, leaving them short staffed. As her shift progresses, RN Armstrong finds herself becoming increasingly irritated with the one post-surgical patient who keeps requesting more pain medication. β€œIt has only been 20 minutes since the last dose,” she said to herself. She checks the guidelines again to quell her own anxiety and perhaps also her sense of guilt at actually feeling angry at her patient who simply wanted relief from her incisional pain. After paging the surgical team with a request for more pain Workplace Resources Demands r.t:::::> _ ~ Job ~ Lifestyle ~ ~~ Adverse Consequences of Burnout in Nurses Medical Patient Mental distress Errors satisfaction in Nurses ~ Figure 1: Burnout in nurses and adverse consequences. medication, she is loudly reprimanded for bothering them during an emergency. She strategically avoids the patient until two hours have passed and she administers the pain medication. Both the patient and her husband are unsatisfied with the care. At the end of her shift she realizes that she missed an order to remove a catheter from another surgical patient – increasing the odds of Catheter Associated Urinary Tract Infection, which RN Armstrong has never knowingly caused. She finds herself not caring, emotionally drained, and unempathetic towards her patients. In recent months, she has no longer felt rewarded by her work and, given the persistent high job demands, has doubted her ability to give optimal care. These thoughts occupy her leasure time and disturb her sleep. Was she depressed? Probably not, although she feels fatigued. Was she angry? Yes, but she had been angry after a shift before. Has job demand (e.g., patient acuity) exceeded job resources (e.g., staffing) long enough to cause her to burnout? Maybe, as RN Armstrong is emotionally drained from her work, contemplating quiting, and, at times, not effectively dealing with patients' problems (illustrated in Figure 1). She calls in sick for the next several shifts, thus worsening the RN shortage on the unit. Even in the absence of the COVID pandemic, 35-45% of RNs experience burnout1-9. RNs and hospital administrators are often unaware of impending burnout and mitigation strategies are not implemented until errors or catastrophes occur in the workplace (e.g., patient death). Hence, there is an urgent need to predict impending burnout by routinely monitoring individual (e.g., physiological, psychological) and workplace (e.g., patient acuity) factors. II. VISION and AIMS This project's vision (illustrated in Figure 2) is to develop a technology to predict burnout in RNs (as a case study) by combining workplace, psychological, and physiological factors, and exploring the barriers to adopting such a technology. This effort focuses on the following aims: Aim1. To create a unique, open-access, de-identified dataset that transforms the science of burnout internationally and informs the interaction of continuous physiological measures (measured from smart watches) and repeated (quarterly) psychological (measured using validated rating scales) and work-related factors (administrative databases) for predicting burnout (Aim 2) in RNs at Mayo Clinic's Florida (Cohorts-A&B) and Rochester (Cohort-C) sites. Aim 2. To develop an analytical framework combining probabilistic graphical models (PGMs) and multitask learning (MTL) to derive interpretable predictions of burnout. PGMs addresses the challenge of inherent stochasticity of burnout manifestation across individuals, and MTL will identify common burnout factors predictive of burnout risks (high, medium and low). Predictability established using Cohort-A will be validated in Cohorts-B&C. Aim 3. Explore barriers (bioethics and administrative) to adopting burnout prediction technologies by assessing perspectives of RNs, nurse supervisors and hospital administrators. OMB No. 0925-0001/0002 (Rev. 03/2020 Approved Through 02/28/2023) Page 70 Continuation Format Page
Effective start/end date4/11/22 β†’ 1/31/23


  • National Institute of Nursing Research: $299,999.00


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.