Improving patient flow during infectious disease outbreaks using machine learning for real-time prediction of patient readiness for discharge.

BACKGROUND: Delays in patient flow and a shortage of hospital beds are commonplace in hospitals during periods of increased infection incidence, such as seasonal influenza and the COVID-19 pandemic. The objective of this study was to develop and evaluate the efficacy of machine learning methods at identifying and ranking the real-time readiness of individual patients for discharge, with the goal of improving patient flow within hospitals during periods of crisis. METHODS AND PERFORMANCE: Electronic Health Record data from Oxford University Hospitals was used to train independent models to classify and rank patients' real-time readiness for discharge within 24 hours, for patient subsets according to the nature of their admission (planned or emergency) and the number of days elapsed since their admission. A strategy for the use of the models' inference is proposed, by which the model makes predictions for all patients in hospital and ranks them in order of likelihood of discharge within the following 24 hours. The 20% of patients with the highest ranking are considered as candidates for discharge and would therefore expect to have a further screening by a clinician to confirm whether they are ready for discharge or not. Performance was evaluated in terms of positive predictive value (PPV), i.e., the proportion of these patients who would have been correctly deemed as 'ready for discharge' after having the second screening by a clinician. Performance was high for patients on their first day of admission (PPV = 0.96/0.94 for planned/emergency patients respectively) but dropped for patients further into a longer admission (PPV = 0.66/0.71 for planned/emergency patients still in hospital after 7 days). CONCLUSION: We demonstrate the efficacy of machine learning methods at making operationally focused, next-day discharge readiness predictions for all individual patients in hospital at any given moment and propose a strategy for their use within a decision-support tool during crisis periods.

Increased time from physiological derangement to critical care admission associates with mortality.

BACKGROUND: Rapid response systems aim to achieve a timely response to the deteriorating patient; however, the existing literature varies on whether timing of escalation directly affects patient outcomes. Prior studies have been limited to using 'decision to admit' to critical care, or arrival in the emergency department as 'time zero', rather than the onset of physiological deterioration. The aim of this study is to establish if duration of abnormal physiology prior to critical care admission ['Score to Door' (STD) time] impacts on patient outcomes. METHODS: A retrospective cross-sectional analysis of data from pooled electronic medical records from a multi-site academic hospital was performed. All unplanned adult admissions to critical care from the ward with persistent physiological derangement [defined as sustained high National Early Warning Score (NEWS) > / = 7 that did not decrease below 5] were eligible for inclusion. The primary outcome was critical care mortality. Secondary outcomes were length of critical care admission and hospital mortality. The impact of STD time was adjusted for patient factors (demographics, sickness severity, frailty, and co-morbidity) and logistic factors (timing of high NEWS, and out of hours status) utilising logistic and linear regression models. RESULTS: Six hundred and thirty-two patients were included over the 4-year study period, 16.3% died in critical care. STD time demonstrated a small but significant association with critical care mortality [adjusted odds ratio of 1.02 (95% CI 1.0-1.04, p = 0.01)]. It was also associated with hospital mortality (adjusted OR 1.02, 95% CI 1.0-1.04, p = 0.026), and critical care length of stay. Each hour from onset of physiological derangement increased critical care length of stay by 1.2%. STD time was influenced by the initial NEWS, but not by logistic factors such as out-of-hours status, or pre-existing patient factors such as co-morbidity or frailty. CONCLUSION: In a strictly defined population of high NEWS patients, the time from onset of sustained physiological derangement to critical care admission was associated with increased critical care and hospital mortality. If corroborated in further studies, this cohort definition could be utilised alongside the 'Score to Door' concept as a clinical indicator within rapid response systems.

Supporting the Quadruple Aim Using Simulation and Human Factors During COVID-19 Care.

The health care sector has made radical changes to hospital operations and care delivery in response to the coronavirus disease (COVID-19) pandemic. This article examines pragmatic applications of simulation and human factors to support the Quadruple Aim of health system performance during the COVID-19 era. First, patient safety is enhanced through development and testing of new technologies, equipment, and protocols using laboratory-based and in situ simulation. Second, population health is strengthened through virtual platforms that deliver telehealth and remote simulation that ensure readiness for personnel to deploy to new clinical units. Third, prevention of lost revenue occurs through usability testing of equipment and computer-based simulations to predict system performance and resilience. Finally, simulation supports health worker wellness and satisfaction by identifying optimal work conditions that maximize productivity while protecting staff through preparedness training. Leveraging simulation and human factors will support a resilient and sustainable response to the pandemic in a transformed health care landscape.

Humanizing the Morbidity and Mortality Conference.

Morbidity and mortality conferences (MMCs) are a long-held legacy institution in academic medicine that enable medical providers and hospital administrators to learn from systemic and individual errors, thereby leading to improved medical care. Originally, this forum had 1 major role-education. The MMC evolved and a second key role was added: quality improvement. In the wake of the 2020 COVID-19 pandemic, a second evolution-one that will humanize the MMC-is required. The pandemic emphasizes the need to use MMCs not only as a place to discuss errors but also as a place for medical providers to reflect on lives lost. The authors' review of the literature regarding MMCs indicates that most studies focus on enabling MMCs to become a forum for quality improvement, while none have emphasized the need to humanize MMCs to decrease medical provider burnout and improve patient satisfaction. Permitting clinicians to be human on the job requires restructuring the MMC to provide a space for reflection and, ultimately, defining a new purpose and charge for the MMC. The authors have 3 main recommendations. First, principles of humanism such as compassion, empathy, and respect, in particular, should be incorporated into traditional MMCs. Second, shorter gatherings devoted to giving clinicians the opportunity to focus on their humanity should be arranged. Third, an MMC focused entirely on the human aspects of medical care should be periodically arranged to provide an outlet for storytelling, artistic expression, and reflection. Humanizing the MMC-a core symposium in clinical medicine worldwide-could be the first step in revitalizing the spirit at the heart of medicine, one dedicated to health and healing. This spirit, which has been eroding as the field of medicine becomes increasingly corporate in structure and mission, is as essential during peaceful times in health care as during a pandemic.