Making Science Computable Using Evidence-Based Medicine on Fast Healthcare Interoperability Resources: Standards Development Project.

BACKGROUND: Evidence-based medicine (EBM) has the potential to improve health outcomes, but EBM has not been widely integrated into the systems used for research or clinical decision-making. There has not been a scalable and reusable computer-readable standard for distributing research results and synthesized evidence among creators, implementers, and the ultimate users of that evidence. Evidence that is more rapidly updated, synthesized, disseminated, and implemented would improve both the delivery of EBM and evidence-based health care policy. OBJECTIVE: This study aimed to introduce the EBM on Fast Healthcare Interoperability Resources (FHIR) project (EBMonFHIR), which is extending the methods and infrastructure of Health Level Seven (HL7) FHIR to provide an interoperability standard for the electronic exchange of health-related scientific knowledge. METHODS: As an ongoing process, the project creates and refines FHIR resources to represent evidence from clinical studies and syntheses of those studies and develops tools to assist with the creation and visualization of FHIR resources. RESULTS: The EBMonFHIR project created FHIR resources (ie, ArtifactAssessment, Citation, Evidence, EvidenceReport, and EvidenceVariable) for representing evidence. The COVID-19 Knowledge Accelerator (COKA) project, now Health Evidence Knowledge Accelerator (HEvKA), took this work further and created FHIR resources that express EvidenceReport, Citation, and ArtifactAssessment concepts. The group is (1) continually refining FHIR resources to support the representation of EBM; (2) developing controlled terminology related to EBM (ie, study design, statistic type, statistical model, and risk of bias); and (3) developing tools to facilitate the visualization and data entry of EBM information into FHIR resources, including human-readable interfaces and JSON viewers. CONCLUSIONS: EBMonFHIR resources in conjunction with other FHIR resources can support relaying EBM components in a manner that is interoperable and consumable by downstream tools and health information technology systems to support the users of evidence.

Impact of the COVID-19 Pandemic on Emergency Department Encounters in a Major Metropolitan Area.

BACKGROUND: The end of 2019 marked the emergence of the COVID-19 pandemic. Public avoidance of health care facilities, including the emergency department (ED), has been noted during prior pandemics. OBJECTIVE: This study described pandemic-related changes in adult and pediatric ED presentations, acuity, and hospitalization rates during the pandemic in a major metropolitan area. METHODS: The study was a cross-sectional analysis of ED visits occurring before and during the pandemic. Sites collected daily ED patient census; monthly ED patient acuity, as the Emergency Severity Index (ESI) score; and disposition. Prepandemic ED visits occurring from January 1, 2019 through December 31, 2019 were compared with ED visits occurring during the pandemic from January 1, 2020 through March 31, 2021. The change in prepandemic and pandemic ED volume was found using 7-day moving average of proportions. RESULTS: The study enrolled 83.8% of the total ED encounters. Pandemic adult and pediatric visit volume decreased to as low as 44.7% (95% CI 43.1-46.3%; p < 0.001) and 22.1% (95% CI 19.3-26.0%; p < 0.001), respectively, of prepandemic volumes. There was also a relative increase in adult and pediatric acuity (ESI level 1-3) and the admission percentage for adult (20.3% vs. 22.9%; p < 0.01) and pediatric (5.1% vs. 5.6%; p < 0.01) populations. CONCLUSIONS: Total adult and pediatric encounters were reduced significantly across a major metropolitan area. Patient acuity and hospitalization rates were relatively increased. The development of strategies for predicting ED avoidance will be important in future pandemics.

WindowSHAP: An efficient framework for explaining time-series classifiers based on Shapley values.

Unpacking and comprehending how black-box machine learning algorithms (such as deep learning models) make decisions has been a persistent challenge for researchers and end-users. Explaining time-series predictive models is useful for clinical applications with high stakes to understand the behavior of prediction models, e.g., to determine how different variables and time points influence the clinical outcome. However, existing approaches to explain such models are frequently unique to architectures and data where the features do not have a time-varying component. In this paper, we introduce WindowSHAP, a model-agnostic framework for explaining time-series classifiers using Shapley values. We intend for WindowSHAP to mitigate the computational complexity of calculating Shapley values for long time-series data as well as improve the quality of explanations. WindowSHAP is based on partitioning a sequence into time windows. Under this framework, we present three distinct algorithms of Stationary, Sliding and Dynamic WindowSHAP, each evaluated against baseline approaches, KernelSHAP and TimeSHAP, using perturbation and sequence analyses metrics. We applied our framework to clinical time-series data from both a specialized clinical domain (Traumatic Brain Injury - TBI) as well as a broad clinical domain (critical care medicine). The experimental results demonstrate that, based on the two quantitative metrics, our framework is superior at explaining clinical time-series classifiers, while also reducing the complexity of computations. We show that for time-series data with 120 time steps (hours), merging 10 adjacent time points can reduce the CPU time of WindowSHAP by 80 % compared to KernelSHAP. We also show that our Dynamic WindowSHAP algorithm focuses more on the most important time steps and provides more understandable explanations. As a result, WindowSHAP not only accelerates the calculation of Shapley values for time-series data, but also delivers more understandable explanations with higher quality.

A self-supervised learning-based approach to clustering multivariate time-series data with missing values (SLAC-Time): An application to TBI phenotyping.

Self-supervised learning approaches provide a promising direction for clustering multivariate time-series data. However, real-world time-series data often include missing values, and the existing approaches require imputing missing values before clustering, which may cause extensive computations and noise and result in invalid interpretations. To address these challenges, we present a Self-supervised Learning-based Approach to Clustering multivariate Time-series data with missing values (SLAC-Time). SLAC-Time is a Transformer-based clustering method that uses time-series forecasting as a proxy task for leveraging unlabeled data and learning more robust time-series representations. This method jointly learns the neural network parameters and the cluster assignments of the learned representations. It iteratively clusters the learned representations with the K-means method and then utilizes the subsequent cluster assignments as pseudo-labels to update the model parameters. To evaluate our proposed approach, we applied it to clustering and phenotyping Traumatic Brain Injury (TBI) patients in the Transforming Research and Clinical Knowledge in Traumatic Brain Injury (TRACK-TBI) study. Clinical data associated with TBI patients are often measured over time and represented as time-series variables characterized by missing values and irregular time intervals. Our experiments demonstrate that SLAC-Time outperforms the baseline K-means clustering algorithm in terms of silhouette coefficient, Calinski Harabasz index, Dunn index, and Davies Bouldin index. We identified three TBI phenotypes that are distinct from one another in terms of clinically significant variables as well as clinical outcomes, including the Extended Glasgow Outcome Scale (GOSE) score, Intensive Care Unit (ICU) length of stay, and mortality rate. The experiments show that the TBI phenotypes identified by SLAC-Time can be potentially used for developing targeted clinical trials and therapeutic strategies.

Characteristics and outcomes of 627 044 COVID-19 patients living with and without obesity in the United States, Spain, and the United Kingdom.

BACKGROUND: A detailed characterization of patients with COVID-19 living with obesity has not yet been undertaken. We aimed to describe and compare the demographics, medical conditions, and outcomes of COVID-19 patients living with obesity (PLWO) to those of patients living without obesity. METHODS: We conducted a cohort study based on outpatient/inpatient care and claims data from January to June 2020 from Spain, the UK, and the US. We used six databases standardized to the OMOP common data model. We defined two non-mutually exclusive cohorts of patients diagnosed and/or hospitalized with COVID-19; patients were followed from index date to 30 days or death. We report the frequency of demographics, prior medical conditions, and 30-days outcomes (hospitalization, events, and death) by obesity status. RESULTS: We included 627 044 (Spain: 122 058, UK: 2336, and US: 502 650) diagnosed and 160 013 (Spain: 18 197, US: 141 816) hospitalized patients with COVID-19. The prevalence of obesity was higher among patients hospitalized (39.9%, 95%CI: 39.8-40.0) than among those diagnosed with COVID-19 (33.1%; 95%CI: 33.0-33.2). In both cohorts, PLWO were more often female. Hospitalized PLWO were younger than patients without obesity. Overall, COVID-19 PLWO were more likely to have prior medical conditions, present with cardiovascular and respiratory events during hospitalization, or require intensive services compared to COVID-19 patients without obesity. CONCLUSION: We show that PLWO differ from patients without obesity in a wide range of medical conditions and present with more severe forms of COVID-19, with higher hospitalization rates and intensive services requirements. These findings can help guiding preventive strategies of COVID-19 infection and complications and generating hypotheses for causal inference studies.

Translational bioinformatics in mental health: open access data sources and computational biomarker discovery.

Mental illness is increasingly recognized as both a significant cost to society and a significant area of opportunity for biological breakthrough. As -omics and imaging technologies enable researchers to probe molecular and physiological underpinnings of multiple diseases, opportunities arise to explore the biological basis for behavioral health and disease. From individual investigators to large international consortia, researchers have generated rich data sets in the area of mental health, including genomic, transcriptomic, metabolomic, proteomic, clinical and imaging resources. General data repositories such as the Gene Expression Omnibus (GEO) and Database of Genotypes and Phenotypes (dbGaP) and mental health (MH)-specific initiatives, such as the Psychiatric Genomics Consortium, MH Research Network and PsychENCODE represent a wealth of information yet to be gleaned. At the same time, novel approaches to integrate and analyze data sets are enabling important discoveries in the area of mental and behavioral health. This review will discuss and catalog into an organizing framework the increasingly diverse set of MH data resources available, using schizophrenia as a focus area, and will describe novel and integrative approaches to molecular biomarker discovery that make use of mental health data.

Risk of depression, suicide and psychosis with hydroxychloroquine treatment for rheumatoid arthritis: a multinational network cohort study.

OBJECTIVES: Concern has been raised in the rheumatology community regarding recent regulatory warnings that HCQ used in the coronavirus disease 2019 pandemic could cause acute psychiatric events. We aimed to study whether there is risk of incident depression, suicidal ideation or psychosis associated with HCQ as used for RA. METHODS: We performed a new-user cohort study using claims and electronic medical records from 10 sources and 3 countries (Germany, UK and USA). RA patients ≥18 years of age and initiating HCQ were compared with those initiating SSZ (active comparator) and followed up in the short (30 days) and long term (on treatment). Study outcomes included depression, suicide/suicidal ideation and hospitalization for psychosis. Propensity score stratification and calibration using negative control outcomes were used to address confounding. Cox models were fitted to estimate database-specific calibrated hazard ratios (HRs), with estimates pooled where I2 <40%. RESULTS: A total of 918 144 and 290 383 users of HCQ and SSZ, respectively, were included. No consistent risk of psychiatric events was observed with short-term HCQ (compared with SSZ) use, with meta-analytic HRs of 0.96 (95% CI 0.79, 1.16) for depression, 0.94 (95% CI 0.49, 1.77) for suicide/suicidal ideation and 1.03 (95% CI 0.66, 1.60) for psychosis. No consistent long-term risk was seen, with meta-analytic HRs of 0.94 (95% CI 0.71, 1.26) for depression, 0.77 (95% CI 0.56, 1.07) for suicide/suicidal ideation and 0.99 (95% CI 0.72, 1.35) for psychosis. CONCLUSION: HCQ as used to treat RA does not appear to increase the risk of depression, suicide/suicidal ideation or psychosis compared with SSZ. No effects were seen in the short or long term. Use at a higher dose or for different indications needs further investigation. TRIAL REGISTRATION: Registered with EU PAS (reference no. EUPAS34497; http://www.encepp.eu/encepp/viewResource.htm? id=34498). The full study protocol and analysis source code can be found at https://github.com/ohdsi-studies/Covid19EstimationHydroxychloroquine2.

COVID-19 in patients with autoimmune diseases: characteristics and outcomes in a multinational network of cohorts across three countries.

OBJECTIVE: Patients with autoimmune diseases were advised to shield to avoid coronavirus disease 2019 (COVID-19), but information on their prognosis is lacking. We characterized 30-day outcomes and mortality after hospitalization with COVID-19 among patients with prevalent autoimmune diseases, and compared outcomes after hospital admissions among similar patients with seasonal influenza. METHODS: A multinational network cohort study was conducted using electronic health records data from Columbia University Irving Medical Center [USA, Optum (USA), Department of Veterans Affairs (USA), Information System for Research in Primary Care-Hospitalization Linked Data (Spain) and claims data from IQVIA Open Claims (USA) and Health Insurance and Review Assessment (South Korea). All patients with prevalent autoimmune diseases, diagnosed and/or hospitalized between January and June 2020 with COVID-19, and similar patients hospitalized with influenza in 2017-18 were included. Outcomes were death and complications within 30 days of hospitalization. RESULTS: We studied 133 589 patients diagnosed and 48 418 hospitalized with COVID-19 with prevalent autoimmune diseases. Most patients were female, aged ≥50 years with previous comorbidities. The prevalence of hypertension (45.5-93.2%), chronic kidney disease (14.0-52.7%) and heart disease (29.0-83.8%) was higher in hospitalized vs diagnosed patients with COVID-19. Compared with 70 660 hospitalized with influenza, those admitted with COVID-19 had more respiratory complications including pneumonia and acute respiratory distress syndrome, and higher 30-day mortality (2.2-4.3% vs 6.32-24.6%). CONCLUSION: Compared with influenza, COVID-19 is a more severe disease, leading to more complications and higher mortality.

Making science computable: Developing code systems for statistics, study design, and risk of bias.

The COVID-19 crisis led a group of scientific and informatics experts to accelerate development of an infrastructure for electronic data exchange for the identification, processing, and reporting of scientific findings. The Fast Healthcare Interoperability Resources (FHIR®) standard which is overcoming the interoperability problems in health information exchange was extended to evidence-based medicine (EBM) knowledge with the EBMonFHIR project. A 13-step Code System Development Protocol was created in September 2020 to support global development of terminologies for exchange of scientific evidence. For Step 1, we assembled expert working groups with 55 people from 26 countries by October 2020. For Step 2, we identified 23 commonly used tools and systems for which the first version of code systems will be developed. For Step 3, a total of 368 non-redundant concepts were drafted to become display terms for four code systems (Statistic Type, Statistic Model, Study Design, Risk of Bias). Steps 4 through 13 will guide ongoing development and maintenance of these terminologies for scientific exchange. When completed, the code systems will facilitate identifying, processing, and reporting research results and the reliability of those results. More efficient and detailed scientific communication will reduce cost and burden and improve health outcomes, quality of life, and patient, caregiver, and healthcare professional satisfaction. We hope the achievements reached thus far will outlive COVID-19 and provide an infrastructure to make science computable for future generations. Anyone may join the effort at https://www.gps.health/covid19_knowledge_accelerator.html.

Applying Computational Ethnography to Examine Nurses' Workflow Within Electronic Health Records.

OBJECTIVE: The aim of this study was to describe computational ethnography as a contemporary and supplemental methodology in EHR workflow analysis and the relevance of this method to nursing research. METHODS: We explore the use of audit logs as a computational ethnographic data source and the utility of data mining techniques, including sequential pattern mining (SPM) and Markov chain analysis (MCA), to analyze nurses' workflow within the EHRs. SPM extracts frequent patterns in a given transactional database (e.g., audit logs from the record). MCA is a stochastic process that models a sequence of states and allows for calculating the probability of moving from one state to the next. These methods can help uncover nurses' global navigational patterns (i.e., how nurses navigate within the record) and enable robust workflow analyses. RESULTS: We demonstrate hypothetical examples from SPM and MCA, such as (a) the most frequent sequential pattern of nurses' workflow when navigating the EHR using SPM and (b) transition probability from one record screen to the next using MCA. These examples demonstrate new methods to address the inflexibility of current approaches used to examine nursing EHR workflow. DISCUSSION: Within a clinical context, the use of computational ethnographic data and data mining techniques can inform the optimization of the EHR. Results from these analyses can be used to supplement the data needed in redesigning the EHR, such as organizing and combining features within a screen or predicting future navigation to improve the record that nurses use.

The bidirectional gut-brain-microbiota axis as a potential nexus between traumatic brain injury, inflammation, and disease.

As head injuries and their sequelae have become an increasingly salient matter of public health, experts in the field have made great progress elucidating the biological processes occurring within the brain at the moment of injury and throughout the recovery thereafter. Given the extraordinary rate at which our collective knowledge of neurotrauma has grown, new insights may be revealed by examining the existing literature across disciplines with a new perspective. This article will aim to expand the scope of this rapidly evolving field of research beyond the confines of the central nervous system (CNS). Specifically, we will examine the extent to which the bidirectional influence of the gut-brain axis modulates the complex biological processes occurring at the time of traumatic brain injury (TBI) and over the days, months, and years that follow. In addition to local enteric signals originating in the gut, it is well accepted that gastrointestinal (GI) physiology is highly regulated by innervation from the CNS. Conversely, emerging data suggests that the function and health of the CNS is modulated by the interaction between 1) neurotransmitters, immune signaling, hormones, and neuropeptides produced in the gut, 2) the composition of the gut microbiota, and 3) integrity of the intestinal wall serving as a barrier to the external environment. Specific to TBI, existing pre-clinical data indicates that head injuries can cause structural and functional damage to the GI tract, but research directly investigating the neuronal consequences of this intestinal damage is lacking. Despite this void, the proposed mechanisms emanating from a damaged gut are closely implicated in the inflammatory processes known to promote neuropathology in the brain following TBI, which suggests the gut-brain axis may be a therapeutic target to reduce the risk of Chronic Traumatic Encephalopathy and other neurodegenerative diseases following TBI. To better appreciate how various peripheral influences are implicated in the health of the CNS following TBI, this paper will also review the secondary biological injury mechanisms and the dynamic pathophysiological response to neurotrauma. Together, this review article will attempt to connect the dots to reveal novel insights into the bidirectional influence of the gut-brain axis and propose a conceptual model relevant to the recovery from TBI and subsequent risk for future neurological conditions.

Discovery of Generalizable TBI Phenotypes Using Multivariate Time-Series Clustering.

Traumatic Brain Injury (TBI) presents a broad spectrum of clinical presentations and outcomes due to its inherent heterogeneity, leading to diverse recovery trajectories and varied therapeutic responses. While many studies have delved into TBI phenotyping for distinct patient populations, identifying TBI phenotypes that consistently generalize across various settings and populations remains a critical research gap. Our research addresses this by employing multivariate time-series clustering to unveil TBI's dynamic intricates. Utilizing a self-supervised learning-based approach to clustering multivariate time-Series data with missing values (SLAC-Time), we analyzed both the research-centric TRACK-TBI and the real-world MIMIC-IV datasets. Remarkably, the optimal hyperparameters of SLAC-Time and the ideal number of clusters remained consistent across these datasets, underscoring SLAC-Time's stability across heterogeneous datasets. Our analysis revealed three generalizable TBI phenotypes (α, ß, and γ), each exhibiting distinct non-temporal features during emergency department visits, and temporal feature profiles throughout ICU stays. Specifically, phenotype α represents mild TBI with a remarkably consistent clinical presentation. In contrast, phenotype ß signifies severe TBI with diverse clinical manifestations, and phenotype γ represents a moderate TBI profile in terms of severity and clinical diversity. Age is a significant determinant of TBI outcomes, with older cohorts recording higher mortality rates. Importantly, while certain features varied by age, the core characteristics of TBI manifestations tied to each phenotype remain consistent across diverse populations.

Acute Respiratory Failure From Early Pandemic COVID-19: Noninvasive Respiratory Support vs Mechanical Ventilation.

BACKGROUND: The optimal strategy for initial respiratory support in patients with respiratory failure associated with COVID-19 is unclear, and the initial strategy may affect outcomes. RESEARCH QUESTION: Which initial respiratory support strategy is associated with improved outcomes in patients with COVID-19 with acute respiratory failure? STUDY DESIGN AND METHODS: All patients with COVID-19 requiring respiratory support and admitted to a large health care network were eligible for inclusion. We compared patients treated initially with noninvasive respiratory support (NIRS; noninvasive positive pressure ventilation by facemask or high-flow nasal oxygen) with patients treated initially with invasive mechanical ventilation (IMV). The primary outcome was time to in-hospital death analyzed using an inverse probability of treatment weighted Cox model adjusted for potential confounders. Secondary outcomes included unweighted and weighted assessments of mortality, lengths of stay (ICU and hospital), and time to intubation. RESULTS: Nearly one-half of the 2,354 patients (47%) who met inclusion criteria received IMV first, and 53% received initial NIRS. Overall, in-hospital mortality was 38% (37% for IMV and 39% for NIRS). Initial NIRS was associated with an increased hazard of death compared with initial IMV (hazard ratio, 1.42; 95% CI, 1.03-1.94), but also an increased hazard of leaving the hospital sooner that waned with time (noninvasive support by time interaction: hazard ratio, 0.97; 95% CI, 0.95-0.98). INTERPRETATION: Patients with COVID-19 with acute hypoxemic respiratory failure initially treated with NIRS showed an increased hazard of in-hospital death.

Assessment of Prediction Tasks and Time Window Selection in Temporal Modeling of Electronic Health Record Data: a Systematic Review.

Temporal electronic health record (EHR) data are often preferred for clinical prediction tasks because they offer more complete representations of a patient's pathophysiology than static data. A challenge when working with temporal EHR data is problem formulation, which includes defining the time windows of interest and the prediction task. Our objective was to conduct a systematic review that assessed the definition and reporting of concepts relevant to temporal clinical prediction tasks. We searched PubMed® and IEEE Xplore® databases for studies from January 1, 2010 applying machine learning models to EHR data for patient outcome prediction. Publications applying time-series methods were selected for further review. We identified 92 studies and summarized them by clinical context and definition and reporting of the prediction problem. For the time windows of interest, 12 studies did not discuss window lengths, 57 used a single set of window lengths, and 23 evaluated the relationship between window length and model performance. We also found that 72 studies had appropriate reporting of the prediction task. However, evaluation of prediction problem formulation for temporal EHR data was complicated by heterogeneity in assessing and reporting of these concepts. Even among studies modeling similar clinical outcomes, there were variations in terminology used to describe the prediction problem, rationale for window lengths, and determination of the outcome of interest. As temporal modeling using EHR data expands, minimal reporting standards should include time-series specific concerns to promote rigor and reproducibility in future studies and facilitate model implementation in clinical settings. Supplementary Information: The online version contains supplementary material available at 10.1007/s41666-023-00143-4.

Advances in Clinical Decision Support Systems: Contributions from the 2022 Literature.

OBJECTIVE: To summarize significant research contributions published in 2022 in the field of clinical decision support (CDS) systems and select the best papers for the Decision Support section of the International Medical Informatics Association (IMIA) Yearbook 2023. METHODS: A renewed search query for identifying CDS scholarship was developed using Medical Subject Headings (MeSH) terms and related keywords. The query was executed in PubMed in January 2023. The search results were reviewed in three stages by two reviewers: title-based triaging, followed by abstract screening, and then full text review. The resulting articles were sent for external review to identity best paper candidates. RESULTS: A total of 1,939 articles related to CDS were retrieved. Of these, 11 articles were selected as candidates for best papers. The general themes of the final three best papers are (1) reducing documentation burden through in-line guidance for clinical notes, (2) clinician engagement for continuous improvement of CDS, and (3) mitigating healthcare-related carbon emissions using scalable and accessible CDS, respectively. CONCLUSION: The field of clinical decision support remains highly active and dynamic, with innovative contributions to a range of clinical domains from primary to acute care. Interoperability issues, documentation burden, clinician acceptance, and the need for effective integration into existing healthcare workflows are among the prominent challenges and areas of interest faced by CDS implementation efforts.

Managed critical care: impact of remote decision-making on patient outcomes.

OBJECTIVES: Tele-intensive care unit (tele-ICU) use has become increasingly common as an extension of bedside care for critically ill patients. The objective of this work was to illustrate the degree of tele-ICU involvement in critical care processes and evaluate the impact of tele-ICU decision-making authority. STUDY DESIGN: Previous studies examining tele-ICU impact on patient outcomes do not sufficiently account for the extent of decision-making authority between remote and bedside providers. In this study, we examine patient outcomes with respect to different levels of remote intervention. METHODS: Analysis and summary statistics were generated to characterize demographics and patient outcomes across different levels of tele-ICU intervention for 82,049 critically ill patients. Multivariate logistic regression was used to evaluate odds of mortality, readmission, and likelihood of patients being assigned to a particular remote intervention category. RESULTS: Managing (vs consulting) physician type influenced the level of remote intervention (adjusted odds ratio [AOR], 2.42). A higher level of tele-ICU intervention was a significant factor for patient mortality (AOR, 1.25). Female sex (AOR, 1.05), illness severity (AOR, 1.01), and higher tele-ICU intervention level (AOR, 1.13) increased odds of ICU readmission, whereas length of stay in number of days (AOR, 0.93) and consulting (vs managing) physician type (AOR, 0.79) decreased readmission odds. CONCLUSIONS: This study's findings suggest that higher levels of tele-ICU intervention do not negatively affect patient outcomes. Our results are a step toward understanding tele-ICU impact on patient outcomes by accounting for extent of decision-making authority, and they suggest that the level of remote intervention may reflect patient severity. Further research using more granular data is needed to better understand assignment of intervention category and how variable levels of authority affect clinical decision-making in tele-ICU settings.

Computable Phenotypes for Post-acute sequelae of SARS-CoV-2: A National COVID Cohort Collaborative Analysis.

Post-acute sequelae of SARS-CoV-2 (PASC) is an increasingly recognized yet incompletely understood public health concern. Several studies have examined various ways to phenotype PASC to better characterize this heterogeneous condition. However, many gaps in PASC phenotyping research exist, including a lack of the following: 1) standardized definitions for PASC based on symptomatology; 2) generalizable and reproducible phenotyping heuristics and meta-heuristics; and 3) phenotypes based on both COVID-19 severity and symptom duration. In this study, we defined computable phenotypes (or heuristics) and meta-heuristics for PASC phenotypes based on COVID-19 severity and symptom duration. We also developed a symptom profile for PASC based on a common data standard. We identified four phenotypes based on COVID-19 severity (mild vs. moderate/severe) and duration of PASC symptoms (subacute vs. chronic). The symptoms groups with the highest frequency among phenotypes were cardiovascular and neuropsychiatric with each phenotype characterized by a different set of symptoms.

Predicting Failure of Noninvasive Respiratory Support Using Deep Recurrent Learning.

BACKGROUND: Noninvasive respiratory support (NRS) is increasingly used to support patients with acute respiratory failure. However, noninvasive support failure may worsen outcomes compared to primary support with invasive mechanical ventilation. Therefore, there is a need to identify patients where NRS is failing so that treatment can be reassessed and adjusted. The objective of this study was to develop and evaluate 3 recurrent neural network (RNN) models to predict NRS failure. METHODS: This was a cross-sectional observational study to evaluate the ability of deep RNN models (long short-term memory [LSTM], gated recurrent unit [GRU]), and GRU with trainable decay) to predict failure of NRS. Data were extracted from electronic health records from all adult (≥ 18 y) patient records requiring any type of oxygen therapy or mechanical ventilation between November 1, 2013-September 30, 2020, across 46 ICUs in the Southwest United States in a single health care network. Input variables for each model included serum chloride, creatinine, albumin, breathing frequency, heart rate, SpO2 , FIO2 , arterial oxygen saturation (SaO2 ), and 2 measurements each (point-of-care and laboratory measurement) of PaO2 and partial pressure of arterial oxygen from an arterial blood gas. RESULTS: Time series data from electronic health records were available for 22,075 subjects. The highest accuracy and area under the receiver operating characteristic curve were for the LSTM model (94.04% and 0.9636, respectively). Accurate predictions were made 12 h after ICU admission, and performance remained high well in advance of NRS failure. CONCLUSIONS: RNN models using routinely collected time series data can accurately predict NRS failure well before intubation. This lead time may provide an opportunity to intervene to optimize patient outcomes.

Identifying TBI Physiological States by Clustering Multivariate Clinical Time-Series Data.

Determining clinically relevant physiological states from multivariate time-series data with missing values is essential for providing appropriate treatment for acute conditions such as Traumatic Brain Injury (TBI), respiratory failure, and heart failure. Utilizing non-temporal clustering or data imputation and aggregation techniques may lead to loss of valuable information and biased analyses. In our study, we apply the SLAC-Time algorithm, an innovative self-supervision-based approach that maintains data integrity by avoiding imputation or aggregation, offering a more useful representation of acute patient states. By using SLAC-Time to cluster data in a large research dataset, we identified three distinct TBI physiological states and their specific feature profiles. We employed various clustering evaluation metrics and incorporated input from a clinical domain expert to validate and interpret the identified physiological states. Further, we discovered how specific clinical events and interventions can influence patient states and state transitions.

Comparison of Force During the Endotracheal Intubation of Commercial Simulation Manikins.

Background  Medical simulation allows clinicians to safely practice the procedural skill of endotracheal intubation. Applied force to oropharyngeal structures increases the risk of patient harm, and video laryngoscopy (VL) requires less force to obtain a glottic view. It is unknown how much force is required to obtain a glottic view using commercially available simulation manikins and if variability exists. This study compares laryngoscopy force for a modified Cormack-Lehane (CL) grade I view in both normal and difficult airway scenarios between three commercially available simulation manikins. Methods Experienced clinicians (≥2 years experience) were recruited to participate from critical care, emergency medicine, and anesthesia specialties. A C-MAC size 3 VL blade was equipped with five force resistor reading (FSR) sensors (four concave surfaces, one convex), measuring resistance (Ohms) in response to applied pressure (1-100 Newtons). The study occurred in a university simulation lab. Using a randomized sequence, 49 physicians performed intubations on three manikins (Laerdal SimMan 3GPlus, Gaumard Hal S3201, CAE Apollo) in normal and difficult airway scenarios. The outcomes were sensor mean pressure, peak force, and CL grade. Summary statistics were calculated. Generalized estimating equations (GEEs) conducted for both scenarios assessed changes in pressure measured in three manikins while accounting for correlated responses of individuals assigned in random order. Paired t-test assessed for the in-manikin difference between scenarios. STATA/BE v17 (R) was used for analysis; results interpreted at type I error alpha is 0.05.  Results Participants included 49 experienced clinicians. Mean years' experience was 4(±6.6); median prior intubations were 80 (IQR 50-400). Mean individual sensor pressure varied within scenarios depending on manikin (p<0.001). Higher mean forces were used in difficult scenarios (603.4±128.9, 611.1±101.4, 467.5±72.4 FSR) than normal (462.5±121.9, 596.0±90.5, 290.6±63.2 FSR) for each manikin (p<0.001). All manikins required more peak force in the difficult scenario (p<0.03). The highest mean forces (Laerdal, CAE, difficult scenario) were associated with the higher frequency of grade 2A views (p<0.001). The Gaumard manikin was rated most realistic in terms of force required to intubate. Conclusion Commercially available high-fidelity manikins had significant variability in laryngoscopy force in both normal and difficult airway scenarios. In difficult airway scenarios, significant variability existed in CL grade between manikin brands. Experienced clinicians rated Gaumard Hal as the most realistic force applied during endotracheal intubation.