Abstract:Background: The key to effective stroke management is timely diagnosis and triage. Machine learning (ML) methods developed to assist in detecting stroke have focused on interpreting detailed clinical data such as clinical notes and diagnostic imaging results. However, such information may not be readily available when patients are initially triaged, particularly in rural and underserved communities. Objective: This study aimed to develop an ML stroke prediction algorithm based on data widely available at the time of patients' hospital presentations and assess the added value of social determinants of health (SDoH) in stroke prediction. Methods: We conducted a retrospective study of the emergency department and hospitalization records from 2012 to 2014 from all the acute care hospitals in the state of Florida, merged with the SDoH data from the American Community Survey. A case-control design was adopted to construct stroke and stroke mimic cohorts. We compared the algorithm performance and feature importance measures of the ML models (ie, gradient boosting machine and random forest) with those of the logistic regression model based on 3 sets of predictors. To provide insights into the prediction and ultimately assist care providers in decision-making, we used TreeSHAP for tree-based ML models to explain the stroke prediction. Results: Our analysis included 143,203 hospital visits of unique patients, and it was confirmed based on the principal diagnosis at discharge that 73% (n=104,662) of these patients had a stroke. The approach proposed in this study has high sensitivity and is particularly effective at reducing the misdiagnosis of dangerous stroke chameleons (false-negative rate <4%). ML classifiers consistently outperformed the benchmark logistic regression in all 3 input combinations. We found significant consistency across the models in the features that explain their performance. The most important features are age, the number of chronic conditions on admission, and primary payer (eg, Medicare or private insurance). Although both the individual- and community-level SDoH features helped improve the predictive performance of the models, the inclusion of the individual-level SDoH features led to a much larger improvement (area under the receiver operating characteristic curve increased from 0.694 to 0.823) than the inclusion of the community-level SDoH features (area under the receiver operating characteristic curve increased from 0.823 to 0.829). Conclusions: Using data widely available at the time of patients' hospital presentations, we developed a stroke prediction model with high sensitivity and reasonable specificity. The prediction algorithm uses variables that are routinely collected by providers and payers and might be useful in underresourced hospitals with limited availability of sensitive diagnostic tools or incomplete data-gathering capabilities.

Abstract TP119: Predicting Posterior Circulation Stroke During Emergency Triage: Dizzy-7 Project

Choice of Intravenous Thrombolysis Therapy in Patients with Mild Stroke Complaining of Acute Dizziness.

Abstract TP53: Above Par on the Back Nine: Improving Triage of Posterior Circulation Stroke

Abstract TP58: Stroke Recognition Task Force: Increasing Early Activation of Inpatient Stroke Codes

A Dynamic Nomogram to Predict the Risk of Stroke in Emergency Department Patients With Acute Dizziness

Abstract WP38: Nurse-Driven Emergency Evaluation Improves Timely Acute Stroke Response

Abstract WP45: A Collaborative Approach to Improving Inpatient Stroke Code Time Target Metrics

The TriAGe+ Score for Vertigo or Dizziness: A Diagnostic Model for Stroke in the Emergency Department

Abstract TP62: Stroke Triage Checklist Decreases Door to Treatment

Abstract TP74: Risk Assessment Of The Acute Stroke Diagnostic Process

Abstract W P210: Stroke Alert Notification by Emergency Department or Emergency Medical Service Which One Shortened Door to Needle Time?

Characteristics and risk of stroke in emergency department patients with acute dizziness

Abstract TP67: Know Your Numbers: Refining the Stroke Alert Process for Hospitalized Patients

A Two-Step Approach Using the National Health Institutes of Health Stroke Scale Assessed by Paramedics to Enhance Prehospital Stroke Detection: A Case Report and Concept Proposal

Abstract 83: Early Notification Using a Novel Three-Tiered Code Stroke System Reduces Neurointerventional Team Activation Times

Abstract TP61: 15 Minutes Can Save Lives: An EMS Direct to CT Protocol for Stroke Patients in the Emergency Department

Prevalence and Determinants of Prehospital Impression of Stroke in Ischemic Stroke Patients

A Machine Learning Approach to Support Urgent Stroke Triage Using Administrative Data and Social Determinants of Health at Hospital Presentation: Retrospective Study

Clinical use of the manchester triage system in patients with dizziness – An observational study in the emergency department

Abstract MP15: Validation of Phenotyping Algorithms for Stroke from Electronic Health Records Using Natural Language Processing

Abstract TP64: Nursing Led Intervention Improves Thrombolytic Administration Time by Optimizing Emergency Department Code Stroke Pathways