Hospital Academic Status and the Volume-Outcome Association in Postoperative Patients Requiring Intensive Care: Results of a Nationwide Analysis of Intensive Care Units in the United States.

Objective: To determine whether the outcomes of postoperative patients admitted directly to an intensive care unit (ICU) differ based on the academic status of the institution and the total operative volume of the unit. Methods: This was a retrospective analysis using the eICU Collaborative Research Database v2.0, a national database from participating ICUs in the United States. All patients admitted directly to the ICU from the operating room were included. Transfer patients and patients readmitted to the ICU were excluded. Patients were stratified based on admission to an ICU in an academic medical center (AMC) versus non-AMC, and to ICUs with different operative volume experience, after stratification in quartiles (high, medium-high, medium-low, and low volume). Primary outcomes were ICU and hospital mortality. Secondary outcomes included the need for continuous renal replacement therapy (CRRT) during ICU stay, ICU length of stay (LOS), and 30-day ventilator free days. Results: Our analysis included 22,180 unique patients; the majority of which (15,085[68%]) were admitted to ICUs in non-AMCs. Cardiac and vascular procedures were the most common types of procedures performed. Patients admitted to AMCs were more likely to be younger and less likely to be Hispanic or Asian. Multivariable logistic regression indicated no meaningful association between academic status and ICU mortality, hospital mortality, initiation of CRRT, duration of ICU LOS, or 30-day ventilator-free-days. Contrarily, medium-high operative volume units had higher ICU mortality (OR = 1.45, 95%CI = 1.10-1.91, p-value = 0.040), higher hospital mortality (OR = 1.33, 95%CI = 1.07-1.66, p-value = 0.033), longer ICU LOS (Coefficient = 0.23, 95%CI = 0.07-0.39, p-value = 0.038), and fewer 30-day ventilator-free-days (Coefficient = -0.30, 95%CI = -0.48 - -0.13, p-value = 0.015) compared to their high operative volume counterparts. Conclusions: This study found that a volume-outcome association in the management of postoperative patients requiring ICU level of care immediately after a surgical procedure may exist. The academic status of the institution did not affect the outcomes of these patients.

Accurate detection of spontaneous seizures using a generalized linear model with external validation.

OBJECTIVE: Seizure detection is a major facet of electroencephalography (EEG) analysis in neurocritical care, epilepsy diagnosis and management, and the instantiation of novel therapies such as closed-loop stimulation or optogenetic control of seizures. It is also of increased importance in high-throughput, robust, and reproducible pre-clinical research. However, seizure detectors are not widely relied upon in either clinical or research settings due to limited validation. In this study, we create a high-performance seizure-detection approach, validated in multiple data sets, with the intention that such a system could be available to users for multiple purposes. METHODS: We introduce a generalized linear model trained on 141 EEG signal features for classification of seizures in continuous EEG for two data sets. In the first (Focal Epilepsy) data set consisting of 16 rats with focal epilepsy, we collected 1012 spontaneous seizures over 3 months of 24/7 recording. We trained a generalized linear model on the 141 features representing 20 feature classes, including univariate and multivariate, linear and nonlinear, time, and frequency domains. We tested performance on multiple hold-out test data sets. We then used the trained model in a second (Multifocal Epilepsy) data set consisting of 96 rats with 2883 spontaneous multifocal seizures. RESULTS: From the Focal Epilepsy data set, we built a pooled classifier with an Area Under the Receiver Operating Characteristic (AUROC) of 0.995 and leave-one-out classifiers with an AUROC of 0.962. We validated our method within the independently constructed Multifocal Epilepsy data set, resulting in a pooled AUROC of 0.963. We separately validated a model trained exclusively on the Focal Epilepsy data set and tested on the held-out Multifocal Epilepsy data set with an AUROC of 0.890. Latency to detection was under 5 seconds for over 80% of seizures and under 12 seconds for over 99% of seizures. SIGNIFICANCE: This method achieves the highest performance published for seizure detection on multiple independent data sets. This method of seizure detection can be applied to automated EEG analysis pipelines as well as closed loop interventional approaches, and can be especially useful in the setting of research using animals in which there is an increased need for standardization and high-throughput analysis of large number of seizures.

Reduction of COVID-19 Incidence and Nonpharmacologic Interventions: Analysis Using a US County-Level Policy Data Set.

BACKGROUND: Worldwide, nonpharmacologic interventions (NPIs) have been the main tool used to mitigate the COVID-19 pandemic. This includes social distancing measures (closing businesses, closing schools, and quarantining symptomatic persons) and contact tracing (tracking and following exposed individuals). While preliminary research across the globe has shown these policies to be effective, there is currently a lack of information on the effectiveness of NPIs in the United States. OBJECTIVE: The purpose of this study was to create a granular NPI data set at the county level and then analyze the relationship between NPI policies and changes in reported COVID-19 cases. METHODS: Using a standardized crowdsourcing methodology, we collected time-series data on 7 key NPIs for 1320 US counties. RESULTS: This open-source data set is the largest and most comprehensive collection of county NPI policy data and meets the need for higher-resolution COVID-19 policy data. Our analysis revealed a wide variation in county-level policies both within and among states (P<.001). We identified a correlation between workplace closures and lower growth rates of COVID-19 cases (P=.004). We found weak correlations between shelter-in-place enforcement and measures of Democratic local voter proportion (R=0.21) and elected leadership (R=0.22). CONCLUSIONS: This study is the first large-scale NPI analysis at the county level demonstrating a correlation between NPIs and decreased rates of COVID-19. Future work using this data set will explore the relationship between county-level policies and COVID-19 transmission to optimize real-time policy formulation.

Using machine learning to predict the risk of developing hypertensive disorders of pregnancy using a contemporary nulliparous cohort.

Background: Hypertensive disorders of pregnancy (HDP) are significant drivers of maternal and neonatal morbidity and mortality. Current management strategies include early identification and initiation of risk mitigating interventions facilitated by a rules-based checklist. Advanced analytic techniques, such as machine learning, can potentially offer improved and refined predictive capabilities. Objective: To develop and internally validate a machine learning prediction model for hypertensive disorders of pregnancy (HDP) when initiating prenatal care. Study Design: We developed a prediction model using data from the prospective multisite cohort Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-Be (nuMoM2b) among low-risk individuals without a prior history of aspirin utilization for preeclampsia prevention. The primary outcome was the development of HDP. Random forest modeling was utilized to develop predictive models. Recursive feature elimination (RFE) was employed to create a reduced model for each outcome. Area under the curve (AUC), 95% confidence intervals (CI), and calibration curves were utilized to assess discrimination and accuracy. Sensitivity analyses were conducted to compare the sensitivity and specificity of the reduced model compared to existing risk factor-based algorithms. Results: Of 9,124 assessed low risk nulliparous individuals, 21% (n=1,927) developed HDP. The prediction model for HDP had satisfactory discrimination with an AUC of 0.73 (95% CI: 0.70, 0.75). After RFE, a parsimonious reduced model with 30 features was created with an AUC of 0.71 (95% CI: 0.68, 0.74). Variables included in the model after RFE included body mass index at the first study visit, pre-pregnancy weight, first trimester complete blood count results, and maximum systolic blood pressure at the first visit. Calibration curves for all models revealed relatively stable agreement between predicted and observed probabilities. Sensitivity analysis noted superior sensitivity (AUC 0.80 vs 0.65) and specificity (0.65 vs 0.53) of the model compared to traditional risk factor-based algorithms. Conclusion: In cohort of low-risk nulliparous pregnant individuals, a prediction model may accurately predict HDP diagnosis at the time of initiating prenatal care and aid employment of close interval monitoring and prophylactic measures earlier in pregnancy.

Designing an Egocentric Video-Based Dashboard to Report Hand Performance Measures for Outpatient Rehabilitation of Cervical Spinal Cord Injury.

Background: Functional use of the upper extremities (UEs) is a top recovery priority for individuals with cervical spinal cord injury (cSCI), but the inability to monitor recovery at home and limitations in hand function outcome measures impede optimal recovery. Objectives: We developed a framework using wearable cameras to monitor hand use at home and aimed to identify the best way to report information to clinicians. Methods: A dashboard was iteratively developed with clinician (n = 7) input through focus groups and interviews, creating low-fidelity prototypes based on recurring feedback until no new information emerged. Affinity diagramming was used to identify themes and subthemes from interview data. User stories were developed and mapped to specific features to create a high-fidelity prototype. Results: Useful elements identified for a dashboard reporting hand performance included summaries to interpret graphs, a breakdown of hand posture and activity to provide context, video snippets to qualitatively view hand use at home, patient notes to understand patient satisfaction or struggles, and time series graphing of metrics to measure trends over time. Conclusion: Involving end-users in the design process and breaking down user requirements into user stories helped identify necessary interface elements for reporting hand performance metrics to clinicians. Clinicians recognized the dashboard's potential to monitor rehabilitation progress, provide feedback on hand use, and track progress over time. Concerns were raised about the implementation into clinical practice, therefore further inquiry is needed to determine the tool's feasibility and usefulness in clinical practice for individuals with UE impairments.

Development of an Offline, Open-Source, Electronic Health Record System for Refugee Care.

While electronic health records (EHRs) have been shown to be effective in improving patient care in low-resource settings, there are still barriers to implementing them, including adaptability, usability, and sustainability. Taking a user-centered design process we developed the Hikma Health EHR for low resourced clinics caring for displaced populations. This EHR was built using React Native and Typescript that sync to a Python backend repository which is deployed on Google Cloud SQL. To date the Hikma Health EHR has been deployed for 26,000 patients. The positive impacts of the system reported by clinician users are 3-fold: (1) improved continuity of care; (2) improved visualization of clinical data; and (3) improved efficiency, resulting in a higher volume of patients being treated. While further development is needed, our open-source model will allow any organization to modify this system to meet their clinical and administrative needs.