Modeling Precision Feedback Knowledge for Healthcare Professional Learning and Quality Improvement

Presentation Time: 03:30 PM - 03:45 PM

Abstract Keywords: Knowledge Representation and Information Modeling, Healthcare Quality, Human-computer Interaction, Education and Training
Primary Track: Foundations
Programmatic Theme: Clinical Informatics

Healthcare providers learn continuously, but better support for provider learning is needed as new biomedical knowledge is produced at an increasing rate alongside widespread use of EHR data for clinical performance measurement. Precision feedback is an approach to improve support for provider learning by prioritizing coaching and appreciation messages based on each message's motivational potential for a specific recipient. We developed a Precision Feedback Knowledge Base as an open resource to support precision feedback systems, containing knowledge models that hold potential as key infrastructure for learning health systems. We describe the design and development of the Precision Feedback Knowledge Base, as well as its key components, including quality measures, feedback message templates, causal pathway models, signal detectors, and prioritization algorithms. Presently, the knowledge base is implemented in a national-scale quality improvement consortium for anesthesia care, to enhance provider feedback email messages.

Speaker(s):
Zach Landis-Lewis, PhD,MLIS
University of Michigan

Author(s):
Yidan Cao, MS, MPP - University of Michigan Medical School; Hana Chung, MS - University of Michigan School of Information; Peter Boisvert, MFA - University of Michigan Medical School; Anjana Deep Renji, MBA - University of Michigan Medical School; Patrick Galante, BSE - University of Michigan Medical School; Ayshwarya Jagadeesan, MSc - University of Michigan Medical School; Farid Seifi, PhD - University of Michigan Medical School; Allison Janda, MD; Nirav Shah, MD - University of Michigan Medical School; Andrew Krumm, PhD - University of Michigan Medical School; Allen Flynn, PharmD, PhD - University of Michigan;

Using Constraint Programming to Optimize Pediatric Infusion Center Scheduling

Presentation Time: 03:45 PM - 04:00 PM

Abstract Keywords: Data Mining, Simulation of Complex Systems, Machine Learning, Pediatrics, Nursing Informatics
Primary Track: Foundations
Programmatic Theme: Clinical Informatics

The CHOP Day Medicine (DM) outpatient infusion center provides more than 50 distinct infusion services, catering to children aged 0 to 16 years and drawing referrals from various specialties. In light of the growing demand for new appointments, the introduction of additional infusion services, and an anticipated increase in visits, establishing an optimal scheduling system is crucial to manage the forthcoming changes while maintaining the same standard of care.

Speaker(s):
Dhineshvikram Krishnamurthy, Master of Science
Childrens Hospital of Philadelphia

Author(s):
Abdul Tariq, PhD - Children's Hospital of Philadelphia; Ekaterina Nekrasova, MPH - Children's Hospital of Philadelphia; Hojjat Salmasian, MD, MPH, PhD, FAMIA - Children's Hospital of Philadelphia;

Orders Without Borders: Ensuring Safe Care During Times of Overflow Hospital Capacity

Presentation Time: 04:00 PM - 04:15 PM

Abstract Keywords: Clinical Decision Support, Healthcare Quality, Patient Safety, Nursing Informatics, Workflow
Primary Track: Applications
Programmatic Theme: Clinical Informatics

This study addresses the critical issue of Emergency Department (ED) boarding, where patients admitted to the hospital await inpatient beds. ED boarding is increasingly common due to hospital capacity challenges and is associated with increased medical errors, treatment delays, and worse patient outcomes. A multidisciplinary workgroup within a large pediatric quaternary health system employed Plan-Do-Study-Act (PDSA) cycles to enhance communication between ED nurses and inpatient clinicians and to streamline the process for inpatient order placement and release for boarding patients. Key interventions included modifying Electronic Health Record (EHR) functionalities to display inpatient provider contact information, creating visual identifiers for patients boarding over two hours, establishing new user access rules for order placement and release, and implementing interruptive alerts with "just-in-time" education. The effectiveness of these interventions was evaluated by tracking the proportion of patients with inpatient orders released prior to inpatient transfer and the time from boarding to order release, comparing pre- and post-intervention periods. Results indicated a significant increase in the proportion of patients with orders released before inpatient admission (from 34% to 64%) and a decrease in median time to order release (from 118 minutes to 85 minutes). These findings demonstrate the success of EHR enhancements and communication tools in facilitating a more efficient transition of care from the ED to inpatient settings, reducing patient wait times in the ED, and highlighting the importance of similar strategies in healthcare systems to improve care coordination and address ED boarding and hospital congestion issues.

Speaker(s):
Julia Yarahuan, MD
Children's Healthcare of Atlanta/Emory University

Author(s):
Julia Yarahuan, MD - Children's Healthcare of Atlanta/Emory University; Brian Curry, MD - Emory University School of Medicine; Megan Fellows, MD - Emory University School of Medicine; Sara Brown, BSN, MPH, RN - Children's Healthcare of Atlanta; Rebekah Carter; Jordan Mitchell, BSN, RN, CPEN - Children's Healthcare of Atlanta; Reena Blanco; Thuy Bui, MD - Children's Healthcare of Atlanta & Pediatric Emergency Medicine Associates, LLC; John Cheng, MD - Pediatric Emergency Medicine Associates, LLC; Jason Aragon, MD - Children's Healthcare of Atlanta; Evan Orenstein, MD - Childrenís Healthcare of Atlanta;

Predicting Home Discharge for Hospitalized Patients Using EHR Data: Methodological Considerations

Presentation Time: 04:15 PM - 04:30 PM

Abstract Keywords: Data Mining, Fairness and elimination of bias, Surveys and Needs Analysis, Transitions of Care
Primary Track: Applications
Programmatic Theme: Clinical Research Informatics

We aim to use patient self-reported social determinants of health (SDOH) data collected in Electronic Health Record (EHR) to predict hospitalized patients’ probability of home discharge. To build a robust prediction model that accommodates missing data, we studied the Bootstrap Imputation-Stability Selection method (BI-SS) and the Stacked Elastic net (SENET) method. Simulation results validated performance of both methods, which yield similar prediction models from the EHR data.

Speaker(s):
Chun Wang, PhD
University of Washington

Author(s):
He Ren, M.S. - University of Washington; Andrea Cheville, MD, MSCE - Mayo Clinic College of Medicine; David Weiss, PhD - University of Minnesota; Gongjun Xu, PhD - University of Michigan; Kathryn Bowles, PhD, RN, FAAN, FACMI - University of Pennsylvania; Tamra Keeney, PhD, PT, DPT, CCS - MGH Institute;

Objects Detection System for ICU Room Using Computer Vision

Presentation Time: 04:30 PM - 04:45 PM

Abstract Keywords: Machine Learning, Knowledge Representation and Information Modeling, Privacy and Security, Simulation of Complex Systems, Patient Safety
Primary Track: Applications
Programmatic Theme: Clinical Informatics

Training deep models tailored for medical applications requires a large amount of data, a challenge that frequently conflicts with patient and medical staff privacy concerns. Our approach addresses this challenge by capturing thousands of images within an intensive care unit (ICU) patient room, employing a diverse dataset to effectively train a deep learning model for ICU object detection.

Speaker(s):
Keivan Nalaie, Doctor of Philosophy
Mayo Clinic

Author(s):
Brian Pickering, MD, FFARCSI - Mayo Clinic; Daniel Diedrich, MD - Mayo Clinic; Vitaly Herasevich, MD, PhD, FCCM, FAMIA - Mayo Clinic; Heidi Lindroth, PhD RN - Mayo Clinic Minnesota;

Discharge Prediction Models for Operational Support: The Case of Multisite Healthcare Systems

Presentation Time: 04:45 PM - 05:00 PM

Abstract Keywords: Machine Learning, Healthcare Quality, Data Mining
Primary Track: Applications
Programmatic Theme: Clinical Informatics

Managing multisite health systems operations is a challenging task. The use of predictive analytics has been proposed as an alternative to improve operational and patient outcomes. However, applications of previous work are limited to patient level long term discharge predictions. Such models are useful to estimate patient level outflows but fail to provide support for capacity management decisions that need to be taken in real time, such as bed placement. In this study, we conduct two experiments using more than 140,000 discharge records from two facilities to evaluate single and multitask models to dynamically predict short term discharge volume. In experiment 1, we study the performance of different machine learning models to predict discharges in the next hour and discharges in the next four hours. Additionally, we compare multitask learning models with single task learning models. In experiment 2, we evaluated the performance of a random forest model to predict the number of discharges from 12:00 PM to 4:00 PM with one to four hours in advance. Results from the numerical experiments suggest that a random forest regressor can significantly outperform a linear regression model in most prediction tasks. In addition, we found that predicting discharges in the next hour is harder relative to discharges in the next four hours and that that accurate forecasts of afternoon discharges can be made using a simple set of explanatory variables even when predicting hours in advance.

Speaker(s):
Fernando Acosta-Perez, B.S.
University of Wisconsin-Madison

Author(s):
Fernando Acosta-Perez, B.S. - University of Wisconsin-Madison; Justin Boutilier, Ph.D. - University of Wisonsin-Madison; Gabriel Zayas-Cabán, Ph.D. - University of Wisconsin-Madison; Sabrina Adelaine, Ph.D. - UW-Health; Frank Liao, PhD - University of Wisconsin, Madison - UW Health; Brian Patterson, MD MPH - University of Wisconsin-Madison;