Towards Predicting Length of Stay and Identification of Cohort Risk Factors Using Self-Attention-Based Transformers and Association Mining: COVID-19 as a Phenotype.

Predicting length of stay (LoS) and understanding its underlying factors is essential to minimizing the risk of hospital-acquired conditions, improving financial, operational, and clinical outcomes, and better managing future pandemics. The purpose of this study was to forecast patients' LoS using a deep learning model and to analyze cohorts of risk factors reducing or prolonging LoS. We employed various preprocessing techniques, SMOTE-N to balance data, and a TabTransformer model to forecast LoS. Finally, the Apriori algorithm was applied to analyze cohorts of risk factors influencing hospital LoS. The TabTransformer outperformed the base machine learning models in terms of F1 score (0.92), precision (0.83), recall (0.93), and accuracy (0.73) for the discharged dataset and F1 score (0.84), precision (0.75), recall (0.98), and accuracy (0.77) for the deceased dataset. The association mining algorithm was able to identify significant risk factors/indicators belonging to laboratory, X-ray, and clinical data, such as elevated LDH and D-dimer levels, lymphocyte count, and comorbidities such as hypertension and diabetes. It also reveals what treatments have reduced the symptoms of COVID-19 patients, leading to a reduction in LoS, particularly when no vaccines or medication, such as Paxlovid, were available.

Effectiveness of Telemedicine in Diabetes Management: A Retrospective Study in an Urban Medically Underserved Population Area (UMUPA)

This paper examines the efficacy of telemedicine (TM) technology compared to traditional face-to-face (F2F) visits as an alternative healthcare delivery service for managing diabetes in populations residing in urban medically underserved areas (UMUPAs). Retrospective electronic patient health records (ePHR) with type 2 diabetes mellitus (T2DM) were examined from 1 January 2019 to 30 June 2021. Multiple linear regression models indicated that T2DM patients with uncontrolled diabetes utilizing TM were similar to traditional visits in lowering hemoglobin (HbA1c) levels. The healthcare service type significantly predicted HbA1c % values, as the regression coefficient for TM (vs. F2F) showed a significant negative association (B = −0.339, p < 0.001), suggesting that patients using TM were likely to have 0.34 lower HbA1c % values on average when compared with F2F visits. The regression coefficient for female (vs. male) gender showed a positive association (B = 0.190, p < 0.034), with HbA1c % levels showing that female patients had 0.19 higher HbA1c levels than males. Age (B = −0.026, p < 0.001) was a significant predictor of HbA1c % levels, with 0.026 lower HbA1c % levels for each year's increase in age. Black adults (B = 0.888, p < 0.001), on average, were more likely to have 0.888 higher HbA1c % levels when compared with White adults.

A method for comparing multiple imputation techniques: A case study on the U.S. national COVID cohort collaborative.

Healthcare datasets obtained from Electronic Health Records have proven to be extremely useful for assessing associations between patients' predictors and outcomes of interest. However, these datasets often suffer from missing values in a high proportion of cases, whose removal may introduce severe bias. Several multiple imputation algorithms have been proposed to attempt to recover the missing information under an assumed missingness mechanism. Each algorithm presents strengths and weaknesses, and there is currently no consensus on which multiple imputation algorithm works best in a given scenario. Furthermore, the selection of each algorithm's parameters and data-related modeling choices are also both crucial and challenging. In this paper we propose a novel framework to numerically evaluate strategies for handling missing data in the context of statistical analysis, with a particular focus on multiple imputation techniques. We demonstrate the feasibility of our approach on a large cohort of type-2 diabetes patients provided by the National COVID Cohort Collaborative (N3C) Enclave, where we explored the influence of various patient characteristics on outcomes related to COVID-19. Our analysis included classic multiple imputation techniques as well as simple complete-case Inverse Probability Weighted models. Extensive experiments show that our approach can effectively highlight the most promising and performant missing-data handling strategy for our case study. Moreover, our methodology allowed a better understanding of the behavior of the different models and of how it changed as we modified their parameters. Our method is general and can be applied to different research fields and on datasets containing heterogeneous types.

The Real-World Experiences of Persons With Multiple Sclerosis During the First COVID-19 Lockdown: Application of Natural Language Processing.

BACKGROUND: The increasing availability of "real-world" data in the form of written text holds promise for deepening our understanding of societal and health-related challenges. Textual data constitute a rich source of information, allowing the capture of lived experiences through a broad range of different sources of information (eg, content and emotional tone). Interviews are the "gold standard" for gaining qualitative insights into individual experiences and perspectives. However, conducting interviews on a large scale is not always feasible, and standardized quantitative assessment suitable for large-scale application may miss important information. Surveys that include open-text assessments can combine the advantages of both methods and are well suited for the application of natural language processing (NLP) methods. While innovations in NLP have made large-scale text analysis more accessible, the analysis of real-world textual data is still complex and requires several consecutive steps. OBJECTIVE: We developed and subsequently examined the utility and scientific value of an NLP pipeline for extracting real-world experiences from textual data to provide guidance for applied researchers. METHODS: We applied the NLP pipeline to large-scale textual data collected by the Swiss Multiple Sclerosis (MS) registry. Such textual data constitute an ideal use case for the study of real-world text data. Specifically, we examined 639 text reports on the experienced impact of the first COVID-19 lockdown from the perspectives of persons with MS. The pipeline has been implemented in Python and complemented by analyses of the "Linguistic Inquiry and Word Count" software. It consists of the following 5 interconnected analysis steps: (1) text preprocessing; (2) sentiment analysis; (3) descriptive text analysis; (4) unsupervised learning-topic modeling; and (5) results interpretation and validation. RESULTS: A topic modeling analysis identified the following 4 distinct groups based on the topics participants were mainly concerned with: "contacts/communication;" "social environment;" "work;" and "errands/daily routines." Notably, the sentiment analysis revealed that the "contacts/communication" group was characterized by a pronounced negative emotional tone underlying the text reports. This observed heterogeneity in emotional tonality underlying the reported experiences of the first COVID-19-related lockdown is likely to reflect differences in emotional burden, individual circumstances, and ways of coping with the pandemic, which is in line with previous research on this matter. CONCLUSIONS: This study illustrates the timely and efficient applicability of an NLP pipeline and thereby serves as a precedent for applied researchers. Our study thereby contributes to both the dissemination of NLP techniques in applied health sciences and the identification of previously unknown experiences and burdens of persons with MS during the pandemic, which may be relevant for future treatment.

Developing a highly-reliable learning health system.

Multiple independent frameworks to support continuous improvement have been proposed to guide healthcare organizations. Two of the most visible are High-reliability Health care, (Chassin et al., 2013) which is emphasized by The Joint Commission, and Learning Health Systems, (Institute of Medicine, 2011) highlighted by the National Academy of Medicine. We propose that organizations consider tightly linking these two models, creating a "Highly-reliable Learning Health System." We describe several efforts at our organization that has resulted from this combined model and have helped our organization weather the COVID-19 pandemic. The organizational changes created using this framework will enable our health system to support a culture of quality across our teams and better fulfill our tripartite mission of high-quality care, effective education of trainees, and dissemination of important innovations.

Developing a highly-reliable learning health system.

Multiple independent frameworks to support continuous improvement have been proposed to guide healthcare organizations. Two of the most visible are High-reliability Health care, (Chassin et al., 2013) which is emphasized by The Joint Commission, and Learning Health Systems, (Institute of Medicine, 2011) highlighted by the National Academy of Medicine. We propose that organizations consider tightly linking these two models, creating a "Highly-reliable Learning Health System." We describe several efforts at our organization that has resulted from this combined model and have helped our organization weather the COVID-19 pandemic. The organizational changes created using this framework will enable our health system to support a culture of quality across our teams and better fulfill our tripartite mission of high-quality care, effective education of trainees, and dissemination of important innovations.

Health Information Systems Adaptation and Flexibility in Extreme Situations: The COVID Effect.

In 2020, a pandemic forced the entire world to adapt to a new scenario. The objective of this study was to know how Health Information Systems were adapted driven by the pandemic of COVID. 12 CIOS of healthcare organizations were interviewed and the interviews were classified according to the dimensions of a sociotechnical model: Infrastructure, Clinical Content, Human Computer Interface, People, Workflow and Communication, Organizational Characteristics and Internal Policies, Regulations, and Measurement and Monitoring. Adaptation to the Pandemic involved social, organizational and cultural rather than merely technical aspects in private organizations with mature and stable Health Information Systems.

Combating COVID-19 Using Generative Adversarial Networks and Artificial Intelligence for Medical Images: Scoping Review.

BACKGROUND: Research on the diagnosis of COVID-19 using lung images is limited by the scarcity of imaging data. Generative adversarial networks (GANs) are popular for synthesis and data augmentation. GANs have been explored for data augmentation to enhance the performance of artificial intelligence (AI) methods for the diagnosis of COVID-19 within lung computed tomography (CT) and X-ray images. However, the role of GANs in overcoming data scarcity for COVID-19 is not well understood. OBJECTIVE: This review presents a comprehensive study on the role of GANs in addressing the challenges related to COVID-19 data scarcity and diagnosis. It is the first review that summarizes different GAN methods and lung imaging data sets for COVID-19. It attempts to answer the questions related to applications of GANs, popular GAN architectures, frequently used image modalities, and the availability of source code. METHODS: A search was conducted on 5 databases, namely PubMed, IEEEXplore, Association for Computing Machinery (ACM) Digital Library, Scopus, and Google Scholar. The search was conducted from October 11-13, 2021. The search was conducted using intervention keywords, such as "generative adversarial networks" and "GANs," and application keywords, such as "COVID-19" and "coronavirus." The review was performed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews (PRISMA-ScR) guidelines for systematic and scoping reviews. Only those studies were included that reported GAN-based methods for analyzing chest X-ray images, chest CT images, and chest ultrasound images. Any studies that used deep learning methods but did not use GANs were excluded. No restrictions were imposed on the country of publication, study design, or outcomes. Only those studies that were in English and were published from 2020 to 2022 were included. No studies before 2020 were included. RESULTS: This review included 57 full-text studies that reported the use of GANs for different applications in COVID-19 lung imaging data. Most of the studies (n=42, 74%) used GANs for data augmentation to enhance the performance of AI techniques for COVID-19 diagnosis. Other popular applications of GANs were segmentation of lungs and superresolution of lung images. The cycleGAN and the conditional GAN were the most commonly used architectures, used in 9 studies each. In addition, 29 (51%) studies used chest X-ray images, while 21 (37%) studies used CT images for the training of GANs. For the majority of the studies (n=47, 82%), the experiments were conducted and results were reported using publicly available data. A secondary evaluation of the results by radiologists/clinicians was reported by only 2 (4%) studies. CONCLUSIONS: Studies have shown that GANs have great potential to address the data scarcity challenge for lung images in COVID-19. Data synthesized with GANs have been helpful to improve the training of the convolutional neural network (CNN) models trained for the diagnosis of COVID-19. In addition, GANs have also contributed to enhancing the CNNs' performance through the superresolution of the images and segmentation. This review also identified key limitations of the potential transformation of GAN-based methods in clinical applications.

Disease-Course Adapting Machine Learning Prognostication Models in Elderly Patients Critically Ill With COVID-19: Multicenter Cohort Study With External Validation.

BACKGROUND: The COVID-19 pandemic caused by SARS-CoV-2 is challenging health care systems globally. The disease disproportionately affects the elderly population, both in terms of disease severity and mortality risk. OBJECTIVE: The aim of this study was to evaluate machine learning-based prognostication models for critically ill elderly COVID-19 patients, which dynamically incorporated multifaceted clinical information on evolution of the disease. METHODS: This multicenter cohort study (COVIP study) obtained patient data from 151 intensive care units (ICUs) from 26 countries. Different models based on the Sequential Organ Failure Assessment (SOFA) score, logistic regression (LR), random forest (RF), and extreme gradient boosting (XGB) were derived as baseline models that included admission variables only. We subsequently included clinical events and time-to-event as additional variables to derive the final models using the same algorithms and compared their performance with that of the baseline group. Furthermore, we derived baseline and final models on a European patient cohort, which were externally validated on a non-European cohort that included Asian, African, and US patients. RESULTS: In total, 1432 elderly (≥70 years old) COVID-19-positive patients admitted to an ICU were included for analysis. Of these, 809 (56.49%) patients survived up to 30 days after admission. The average length of stay was 21.6 (SD 18.2) days. Final models that incorporated clinical events and time-to-event information provided superior performance (area under the receiver operating characteristic curve of 0.81; 95% CI 0.804-0.811), with respect to both the baseline models that used admission variables only and conventional ICU prediction models (SOFA score, P<.001). The average precision increased from 0.65 (95% CI 0.650-0.655) to 0.77 (95% CI 0.759-0.770). CONCLUSIONS: Integrating important clinical events and time-to-event information led to a superior accuracy of 30-day mortality prediction compared with models based on the admission information and conventional ICU prediction models. This study shows that machine-learning models provide additional information and may support complex decision-making in critically ill elderly COVID-19 patients. TRIAL REGISTRATION: ClinicalTrials.gov NCT04321265; https://clinicaltrials.gov/ct2/show/NCT04321265.

Tackling the Burden of Electronic Health Record Use Among Physicians in a Mental Health Setting: Physician Engagement Strategy.

The burden associated with using the electronic health record system continues to be a critical issue for physicians and is potentially contributing to physician burnout. At a large academic mental health hospital in Canada, we recently implemented a Physician Engagement Strategy focused on reducing the burden of electronic health record use through close collaboration with clinical leadership, information technology leadership, and physicians. Built on extensive stakeholder consultation, this strategy highlights initiatives that we have implemented (or will be implementing in the near future) under four components: engage, inspire, change, and measure. In this viewpoint paper, we share our process of developing and implementing the Physician Engagement Strategy and discuss the lessons learned and implications of this work.

Use of clinical data to augment healthcare worker contact tracing during the COVID-19 pandemic.

OBJECTIVE: This work examined the secondary use of clinical data from the electronic health record (EHR) for screening our healthcare worker (HCW) population for potential exposures to patients with coronavirus disease 2019 (COVID-19). MATERIALS AND METHODS: We conducted a cross-sectional study at a free-standing, quaternary care pediatric hospital comparing first-degree, patient-HCW pairs identified by the hospital's COVID-19 contact tracing team (CTT) to those identified using EHR clinical event data (EHR Report). The primary outcome was the number of patient-HCW pairs detected by each process. RESULTS: Among 233 patients with COVID-19, our EHR Report identified 4116 patient-HCW pairs, including 2365 (30.0%) of the 7890 pairs detected by the CTT. The EHR Report also revealed 1751 pairs not identified by the CTT. The highest number of patient-HCW pairs per patient was detected in the inpatient care venue. Nurses comprised the most frequently identified HCW role overall. CONCLUSIONS: Automated methods to screen HCWs for potential exposures to patients with COVID-19 using clinical event data from the EHR (1) are likely to improve epidemiological surveillance by contact tracing programs and (2) represent a viable and readily available strategy that should be considered by other institutions.

Virtual primary care in high-income countries during the COVID-19 pandemic: Policy responses and lessons for the future.

BACKGROUND: Telemedicine, once defined merely as the treatment of certain conditions remotely, has now often been supplanted in use by broader terms such as 'virtual care', in recognition of its increasing capability to deliver a diverse range of healthcare services from afar. With the unexpected onset of COVID-19, virtual care (e.g. telephone, video, online) has become essential to facilitating the continuation of primary care globally. Over several short weeks, existing healthcare policies have adapted quickly and empowered clinicians to use digital means to fulfil a wide range of clinical responsibilities, which until then have required face-to-face consultations. OBJECTIVES: This paper aims to explore the virtual care policies and guidance material published during the initial months of the pandemic and examine their potential limitations and impact on transforming the delivery of primary care in high-income countries. METHODS: A rapid review of publicly available national policies guiding the use of virtual care in General Practice was conducted. Documents were included if issued in the first six months of the pandemic (March to August of 2020) and focussed primarily on high-income countries. Documents must have been issued by a national health authority, accreditation body, or professional organisation, and directly refer to the delivery of primary care. RESULTS: We extracted six areas of relevance: primary care transformation during COVID-19, the continued delivery of preventative care, the delivery of acute care, remote triaging, funding & reimbursement, and security standards. CONCLUSION: Virtual care use in primary care saw a transformative change during the pandemic. However, despite the advances in the various governmental guidance offered, much work remains in addressing the shortcomings exposed during COVID-19 and strengthening viable policies to better incorporate novel technologies into the modern primary care clinical environment.

Telehealth as a new care delivery model: The headache provider experience.

OBJECTIVE: To assess telehealth practice for headache visits in the United States. BACKGROUND: The rapid roll out of telehealth during the COVID-19 pandemic impacted headache specialists. METHODS: American Headache Society (AHS) members were emailed an anonymous survey (9/9/20-10/12/20) to complete if they had logged ≥2 months or 50+ headache visits via telehealth. RESULTS: Out of 1348 members, 225 (16.7%) responded. Most were female (59.8%; 113/189). Median age was 47 (interquartile range [IQR] 37-57) (N = 154). The majority were MD/DOs (83.7%; 159/190) or NP/PAs (14.7%; 28/190), and most (65.1%; 123/189) were in academia. Years in practice were 0-3: 28; 4-10: 58; 11-20: 42; 20+: 61. Median number of telehealth visits was 120 (IQR 77.5-250) in the prior 3 months. Respondents were "comfortable/very comfortable" treating via telehealth (a) new patient with a chief complaint of headache (median, IQR 4 [3-5]); (b) follow-up for migraine (median, IQR 5 [5-5]); (c) follow-up for secondary headache (median, IQR 4 [3-4]). About half (51.1%; 97/190) offer urgent telehealth. Beyond being unable to perform procedures, top barriers were conducting parts of the neurologic exam (157/189), absence of vital signs (117/189), and socioeconomic/technologic barriers (91/189). Top positive attributes were patient convenience (185/190), reducing patient travel stress (172/190), patient cost reduction (151/190), flexibility with personal matters (128/190), patient comfort at home (114/190), and patient medications nearby (103/190). Only 21.3% (33/155) of providers said telehealth visit length differed from in-person visits, and 55.3% (105/190) believe that the no-show rate improved. On a 1-5 Likert scale, providers were "interested"/"very interested" in digitally prescribing headache apps (median 4, IQR 3-5) and "interested"/"very interested" in remotely monitoring patient symptoms (median 4, IQR 3-5). CONCLUSIONS: Respondents were comfortable treating patients with migraine via telehealth. They note positive attributes for patients and how access may be improved. Technology innovations (remote vital signs, digitally prescribing headache apps) and remote symptom monitoring are areas of interest and warrant future research.

Can Orthodontic Informatics Combat the Pandemic Pitfalls?

The largest public health crisis of our time, COVID-19 has recklessly squandered many of the channelized healthcare facilities globally with execution of newer guidelines over the standard architectural norms. There has been unparalleled use of smartphones and internet services to bear the major pitfall- social distancing- especially for elective treatment services. This demands a new paradigm shift from offline to online doctor-patient, student-educator, researcher-researcher operations. This articles provides an insight into potential role of orthodontic informatics to provide a combined platform to generate a learning system that routinely collects, correlates, and analyzes data for developing artificial intelligence programs, lab exploratory systems, clinical decision support systems and health-information exchange systems. In order to develop this system, orthodontic analytic communities as start-ups for developing user-friendly programs must be encouraged, where orthodontic informatics itself can be taken up as a didactic career source.

Novel application of telemedicine and an alternate EHR environment for virtual clinical education: A new model for primary care education during the SARS-CoV-2 pandemic.

BACKGROUND: Restrictions to direct patient contact resulting from the Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) pandemic left some medical students near graduation in need of a required critical care medicine (CCM) sub-internship. A group of educators deployed a virtual curriculum utilizing telemedicine and electronic health record (EHR) technologies. METHODS: Nine students participated in a formal curriculum of high-value critical care medicine topics designed to meet the learning objectives of the in-person experience. Students obtained patient histories and directed physical examinations virtually via telemedicine. They followed assigned patients, submitted clinical documentation, and practiced electronic order entry using a non-production EHR copy. At conclusion these students completed the same evaluation used for "in-person" CCM rotations earlier in the year. RESULTS: Students rated the virtual rotation comparably to the traditional rotation in most evaluated criteria. Lower rated areas included "perform minor procedures", "patient counseling", and "interprofessional experiences". Students' narrative responses specifically noted strengths of the "student focus" and the ability to practice in an EHR copy. DISCUSSION: Students and preceptors generally found that the virtual curriculum provided adequate educational opportunities. Certain areas were clearly lacking, as expected. Students felt the dedication of the faculty to the students' educational needs was the most important factor contributing to the success of the program. The results suggest several ways telemedicine and EHR technologies might enhance clinical medical education in the future. CONCLUSION: This methodology was successful in providing elements of a CCM rotation experience. This technology could prove efficacious for primary care rotations where in-person training is not feasible due to the SARS-CoV-2 pandemic.

Leveraging Informatics and Technology to Support Public Health Response: Framework and Illustrations using COVID-19.

OBJECTIVE: To develop a conceptual model and novel, comprehensive framework that encompass the myriad ways informatics and technology can support public health response to a pandemic. METHOD: The conceptual model and framework categorize informatics solutions that could be used by stakeholders (e.g., government, academic institutions, healthcare providers and payers, life science companies, employers, citizens) to address public health challenges across the prepare, respond, and recover phases of a pandemic, building on existing models for public health operations and response. RESULTS: Mapping existing solutions, technology assets, and ideas to the framework helped identify public health informatics solution requirements and gaps in responding to COVID-19 in areas such as applied science, epidemiology, communications, and business continuity. Two examples of technologies used in COVID-19 illustrate novel applications of informatics encompassed by the framework. First, we examine a hub from The Weather Channel, which provides COVID-19 data via interactive maps, trend graphs, and details on case data to individuals and businesses. Second, we examine IBM Watson Assistant for Citizens, an AI-powered virtual agent implemented by healthcare providers and payers, government agencies, and employers to provide information about COVID-19 via digital and telephone-based interaction. DISCUSSION: Early results from these novel informatics solutions have been positive, showing high levels of engagement and added value across stakeholders. CONCLUSION: The framework supports development, application, and evaluation of informatics approaches and technologies in support of public health preparedness, response, and recovery during a pandemic. Effective solutions are critical to success in recovery from COVID-19 and future pandemics.

Predictability of COVID-19 Hospitalizations, Intensive Care Unit Admissions, and Respiratory Assistance in Portugal: Longitudinal Cohort Study.

BACKGROUND: In the face of the current COVID-19 pandemic, the timely prediction of upcoming medical needs for infected individuals enables better and quicker care provision when necessary and management decisions within health care systems. OBJECTIVE: This work aims to predict the medical needs (hospitalizations, intensive care unit admissions, and respiratory assistance) and survivability of individuals testing positive for SARS-CoV-2 infection in Portugal. METHODS: A retrospective cohort of 38,545 infected individuals during 2020 was used. Predictions of medical needs were performed using state-of-the-art machine learning approaches at various stages of a patient's cycle, namely, at testing (prehospitalization), at posthospitalization, and during postintensive care. A thorough optimization of state-of-the-art predictors was undertaken to assess the ability to anticipate medical needs and infection outcomes using demographic and comorbidity variables, as well as dates associated with symptom onset, testing, and hospitalization. RESULTS: For the target cohort, 75% of hospitalization needs could be identified at the time of testing for SARS-CoV-2 infection. Over 60% of respiratory needs could be identified at the time of hospitalization. Both predictions had >50% precision. CONCLUSIONS: The conducted study pinpoints the relevance of the proposed predictive models as good candidates to support medical decisions in the Portuguese population, including both monitoring and in-hospital care decisions. A clinical decision support system is further provided to this end.

Modeling Predictive Age-Dependent and Age-Independent Symptoms and Comorbidities of Patients Seeking Treatment for COVID-19: Model Development and Validation Study.

BACKGROUND: The COVID-19 pandemic continues to ravage and burden hospitals around the world. The epidemic started in Wuhan, China, and was subsequently recognized by the World Health Organization as an international public health emergency and declared a pandemic in March 2020. Since then, the disruptions caused by the COVID-19 pandemic have had an unparalleled effect on all aspects of life. OBJECTIVE: With increasing total hospitalization and intensive care unit admissions, a better understanding of features related to patients with COVID-19 could help health care workers stratify patients based on the risk of developing a more severe case of COVID-19. Using predictive models, we strive to select the features that are most associated with more severe cases of COVID-19. METHODS: Over 3 million participants reported their potential symptoms of COVID-19, along with their comorbidities and demographic information, on a smartphone-based app. Using data from the >10,000 individuals who indicated that they had tested positive for COVID-19 in the United Kingdom, we leveraged the Elastic Net regularized binary classifier to derive the predictors that are most correlated with users having a severe enough case of COVID-19 to seek treatment in a hospital setting. We then analyzed such features in relation to age and other demographics and their longitudinal trend. RESULTS: The most predictive features found include fever, use of immunosuppressant medication, use of a mobility aid, shortness of breath, and severe fatigue. Such features are age-related, and some are disproportionally high in minority populations. CONCLUSIONS: Predictors selected from the predictive models can be used to stratify patients into groups based on how much medical attention they are expected to require. This could help health care workers devote valuable resources to prevent the escalation of the disease in vulnerable populations.

Machine Learning to Predict Mortality and Critical Events in a Cohort of Patients With COVID-19 in New York City: Model Development and Validation.

BACKGROUND: COVID-19 has infected millions of people worldwide and is responsible for several hundred thousand fatalities. The COVID-19 pandemic has necessitated thoughtful resource allocation and early identification of high-risk patients. However, effective methods to meet these needs are lacking. OBJECTIVE: The aims of this study were to analyze the electronic health records (EHRs) of patients who tested positive for COVID-19 and were admitted to hospitals in the Mount Sinai Health System in New York City; to develop machine learning models for making predictions about the hospital course of the patients over clinically meaningful time horizons based on patient characteristics at admission; and to assess the performance of these models at multiple hospitals and time points. METHODS: We used Extreme Gradient Boosting (XGBoost) and baseline comparator models to predict in-hospital mortality and critical events at time windows of 3, 5, 7, and 10 days from admission. Our study population included harmonized EHR data from five hospitals in New York City for 4098 COVID-19-positive patients admitted from March 15 to May 22, 2020. The models were first trained on patients from a single hospital (n=1514) before or on May 1, externally validated on patients from four other hospitals (n=2201) before or on May 1, and prospectively validated on all patients after May 1 (n=383). Finally, we established model interpretability to identify and rank variables that drive model predictions. RESULTS: Upon cross-validation, the XGBoost classifier outperformed baseline models, with an area under the receiver operating characteristic curve (AUC-ROC) for mortality of 0.89 at 3 days, 0.85 at 5 and 7 days, and 0.84 at 10 days. XGBoost also performed well for critical event prediction, with an AUC-ROC of 0.80 at 3 days, 0.79 at 5 days, 0.80 at 7 days, and 0.81 at 10 days. In external validation, XGBoost achieved an AUC-ROC of 0.88 at 3 days, 0.86 at 5 days, 0.86 at 7 days, and 0.84 at 10 days for mortality prediction. Similarly, the unimputed XGBoost model achieved an AUC-ROC of 0.78 at 3 days, 0.79 at 5 days, 0.80 at 7 days, and 0.81 at 10 days. Trends in performance on prospective validation sets were similar. At 7 days, acute kidney injury on admission, elevated LDH, tachypnea, and hyperglycemia were the strongest drivers of critical event prediction, while higher age, anion gap, and C-reactive protein were the strongest drivers of mortality prediction. CONCLUSIONS: We externally and prospectively trained and validated machine learning models for mortality and critical events for patients with COVID-19 at different time horizons. These models identified at-risk patients and uncovered underlying relationships that predicted outcomes.

Clinical informatics during the COVID-19 pandemic: Lessons learned and implications for emergency department and inpatient operations.

In response to a pandemic, hospital leaders can use clinical informatics to aid clinical decision making, virtualizing medical care, coordinating communication, and defining workflow and compliance. Clinical informatics procedures need to be implemented nimbly, with governance measures in place to properly oversee and guide novel patient care pathways, diagnostic and treatment workflows, and provider education and communication. The authors' experience recommends (1) creating flexible order sets that adapt to evolving guidelines that meet needs across specialties, (2) enhancing and supporting inherent telemedicine capability, (3) electronically enabling novel workflows quickly and suspending noncritical administrative or billing functions in the electronic health record, and (4) using communication platforms based on tiered urgency that do not compromise security and privacy.