SHM Converge Scientific Abstract ePoster Gallery-Design of a Prediction Algorithm to Escalate Post-Discharge Adverse Events

Pdf Summary

This project focuses on improving detection of adverse events (AEs) after hospital discharge, especially in patients with multiple chronic conditions. AEs are common during care transitions and often result in unexpected healthcare use such as emergency department visits or readmissions. Although new or worsening symptoms are early warning signs, systematic symptom monitoring is often missing.<br /><br />The team is developing a prediction model that combines electronic health record (EHR) data with responses to the 10-item Global Health PRO questionnaire. The goal is to use this model in a digital health app that can monitor symptoms in real time, identify patients at higher risk, and escalate concerns to clinicians earlier.<br /><br />Their design process includes five steps: enrolling hospitalized patients, collecting PRO data at admission, reviewing charts to identify adverse events within 30 days after discharge, creating journey maps for high-risk cases, and selecting EHR-based input variables for the algorithm. Potential inputs include self-rated health, discharge preparedness, readmission risk scores, symptom reports, appointments/encounters, medications, demographics, and social determinants of health.<br /><br />Among 101 adjudicated cases, 8 patients (7.9%) experienced at least one preventable or ameliorable AE within 14 days of discharge, and these events were linked to unplanned utilization. In these cases, common symptoms included shortness of breath and swollen legs; some patients rated their overall health as fair or poor. Journey mapping helped identify when symptoms and healthcare contacts occurred, informing the best timing for questionnaires and escalation.<br /><br />The project’s conclusion is that electronic PRO monitoring could be transformative for post-discharge care by helping patients understand their risk, receive tailored self-care guidance, and seek help sooner. Next steps include confirming the final data elements with stakeholders and aligning them with USCDI standards to maximize interoperability.

Asset Subtitle

Anuj K. Dalal

Meta Tag

Author List Anuj K. Dalal, Jorge A. Rodriguez, Kaitlyn A. Konieczny, Marie Leeson, Pamela Garabedian, Robert S. Rudin, Savanna Plombon

Category Innovations

Concept Adverse Event

Concept Prediction Algorithm

Concept Care Transition

Concept Post-discharge Healthcare Utilization

Concept Patient-Reported Outcome

Distinguished Non-Finalist

Presenter Organization Brigham and Women's Hospital

Presenter Organization Harvard Medical School

Presenting Author Anuj K. Dalal

Track Transitions of Care

Keywords

adverse events

post-discharge care

patient-reported outcomes

electronic health records

symptom monitoring

care transitions

readmission risk

digital health app

multiple chronic conditions

USCDI interoperability

Adverse Event

Prediction Algorithm

Care Transition

Post-discharge Healthcare Utilization

Patient-Reported Outcome