Description

An onsite recording of this session will be included in the Symposium OnDemand offering. 

Multi-state Modeling of Pressure Injury Staging Transition Trajectories

Presentation Time: 01:45 PM - 02:00 PM

Abstract Keywords: Disease Models, Nursing Informatics, Informatics Implementation, Precision Medicine, Data Mining
Primary Track: Applications
Programmatic Theme: Clinical Informatics

This study was conducted to evaluate the time-sensitive progression trajectory of pressure injury stages based on real-world electronic health record (EHR) datasets. Clinical databases within the Mass General Brigham (MGB) Healthcare system was used as data source. Both pressure injury anatomical locations and staging values were obtained through EHR flowsheets. Our results suggested that early intervention, especially for patients with stage 1 can be a very important strategy to prevent severe pressure injury.

Speaker(s):
Wenyu Song, PhD
Brigham and Women's Hospital, Harvard Medical School

Author(s):
Wenyu Song, PhD - Brigham and Women's Hospital, Harvard Medical School; Min Jeoung Kang, PhD - Brigham and Women's Hospital/ Harvard Medical School; Luwei Liu, MBI - Brigham and Women’s Hospital; Graham Lowenthal, BA - Brigham and Women's Hospital; Veysel Baris, Nurse - Dokuz Eylul University; Sandy Cho, MPH IN-BC - Newton-Wellesley Hospital; Diane Carroll, PhD - Massachusetts General Hospital; Debra Furlong, RN-BC - Brigham and Women’s Hospital; Wadia Gilles-Fowler, RN - Brigham and Women’s Hospital; Luciana Goncalves, PhD - Brigham and Women’s Hospital; Beth Melanson, RN - Brigham and Women’s Hospital; Lori Morrow, RN - Salem Hospital; Jacqueline Massaro; Tanya Martel, DNP - Brigham and Women’s Hospital; Paula Wolski, MSN, RN, NI-BC - Brigham and Womens Faulkner Hospital; Linying Zhang, PhD - Washington University in St. Louis; PATRICIA C DYKES, PhD, MA, RN - Brigham and Women's Hospital;

Nurses’ Visual Attention in EHR Nursing Summaries through Eye-Tracking Study

Presentation Time: 02:00 PM - 02:15 PM

Abstract Keywords: Workflow, Human-computer Interaction, Usability, Documentation Burden
Primary Track: Applications

This study assessed how nurses allocate their visual attention when reading EHR nursing summaries and examined with information volume. Conducted with 33 nurses from a university hospital using eye-tracking simulations, findings revealed a predominant focus on "Orders" and "Sidebar" information across patient acuity levels, due in part to its large volume of information. Our results highlight the need for EHR nursing summary redesign by removing less important information-types.

Speaker(s):
Suhyun Park, PhD, RN
UTHealth Houston Cizik School of Nursing

Author(s):
Jenna Marquard, PhD - University of Minnesota; Robin Austin - University of Minnesota, School of Nursing; Christie Martin, PhD, MPH, RN-BC, LHIT-HP - University of Minnesota School of Nursing; David Pieczkiewicz, PhD - University of Minnesota; Connie Delaney, PhD - University of Minnesota, School of Nursing;

Preemptive Forecasting of Symptom Escalation in Cancer Patients Undergoing Chemotherapy

Presentation Time: 02:15 PM - 02:30 PM

Abstract Keywords: Nursing Informatics, Machine Learning, Chronic Care Management, Bioinformatics
Primary Track: Applications
Programmatic Theme: Clinical Informatics

This study evaluates the utility of machine learning (ML) algorithms in early forecasting of total symptom score changes from daily self-reports of 339 chemotherapy patients. The dataset comprised 12 specific symptoms, with severity and distress for each symptom rated on a 1 to 10 scale, generating a "total symptom score" ranging from 0 to 230. To address the challenge of an unbalanced original dataset, where Class I (score change ≥ 5) and Class II (score change < 5) were unevenly represented, we created a balanced dataset specifically for model training.
Using the MATLAB® Classification Learner application, we investigated nine ML models with various classifiers. The objective was to predict the total symptom score change based on the preceding 3 to 5 days' symptom data. Models were trained on the balanced dataset to mitigate the original imbalance's impact, with comparative evaluations also conducted on the unbalanced data to assess performance differences.
The analysis revealed that certain classifiers, delivered optimal performance on the unbalanced dataset, with an accuracy rate peaking at 82%. Yet, these models tended to frequently misclassify Class I as Class II. In contrast, the Ensemble algorithm equipped with the RUSBoost classifier demonstrated exceptional skill in accurately classifying both classes on both datasets, achieving accuracies of 59%, 59.3%, and 59.4% for data from 3, 4, and 5 days prior, respectively. Notably, these figures slightly improved to 61.16%, 58.41%, and 60.05% upon utilizing the balanced dataset for training.

Speaker(s):
AREF SMILEY, Assistant Professor/PhD
The University of Utah

Author(s):
Joseph Finkelstein, MD, PhD - University of Utah; Aref Smiley, Assistant Professor/PhD - The University of Utah; Christina Echeverria, MA - College of Nursing, The University of Utah; Kathi Mooney, PhD, RN, FAAN - College of Nursing, The University of Utah;

Nursing Workload and Overcrowding: Patient Safety Role in Emergency Department

Presentation Time: 02:30 PM - 02:45 PM

Abstract Keywords: Nursing Informatics, Data Mining, Patient Safety, Workflow, Change Management
Primary Track: Applications
Programmatic Theme: Clinical Informatics

This study evaluates nursing workload and the National Overcrowding Score (NEDOCS) in emergency departments (ED) by analyzing electronic health records. It finds weak correlation between workload and NEDOCS (r=0.346) demonstrating their distinct roles in the ED, but a significant link between antibiotic administration for pneumonia patients, NEDOCS (r=0.823), and workload (r=0.952). Highlighting the importance of nurse workload management, it suggests this focus can improve patient safety and care quality in challenging ED settings.

Speaker(s):
Junhyuk Seo, Registered Nurse
Samsung Medical Center

Author(s):
Junhyuk Seo, Registered Nurse - Samsung Medical Center; Junsang Yoo - SAIHST; Wonchul Cha - Samsung Medical Center;

Using a Healthcare Process Modeling Approach to Understand Electronic Health Records-based Pressure Injury Data and to Support Development of a Standardized Pressure Injury Phenotyping Pipeline

Presentation Time: 02:45 PM - 03:00 PM

Abstract Keywords: Nursing Informatics, Informatics Implementation, Healthcare Quality, Workflow, Data Mining
Working Group: Nursing Informatics Working Group
Primary Track: Applications
Programmatic Theme: Clinical Informatics

The complexity of health care processes present significant challenges for using Electronic Health Records (EHR) data to build high fidelity phenotypes. This study leverages a healthcare process modeling (HPM) approach to enable understanding of EHR-based pressure injury (PrI) data patterns needed for building a standardized PrI phenotyping pipeline. The PrI HPM was developed and validated using mixed methods, including exploratory sequential design, through interdisciplinary collaboration among clinical experts, data scientists, database analysts, and informaticians. The qualitative analysis identified the dynamics between PrI care and the associated clinical documentation processes. The quantitative analysis identified inherent challenges and limitations of the PrI data. The PrI HPM includes three moderating factors: system configuration, hospital policy, and nurse's individual workflow. We further incorporated the HPM into the PrI phenotype development process to address phenotyping challenges. Moreover, we suggested a set of standardizable recommendations to address PrI phenotyping challenges.

Speaker(s):
Luwei Liu, Master of Biomedical Informatics
Brigham and Women's Hospital

Author(s):
Luwei Liu, Master of Biomedical Informatics - Brigham and Women's Hospital; Min Jeoung Kang, PhD - Brigham and Women's Hospital/ Harvard Medical School; Michael Sainlaire - Brigham and Women's Health; Graham Lowenthal, BA - Brigham and Women's Hospital; Tanya Martel, DNP, FNP-BC. CWOCN - Center for Nursing Excellence, Brigham and Women’s Hospital; Sandy Cho, MPH IN-BC - Newton-Wellesley Hospital; Debra Furlong, MS, RN-BC - department of Nursing Services, Brigham and Women’s Hospital; Wadia Gilles-Fowler, BSN, RN, CWOCN - Center for Nursing Excellence, Brigham and Women’s Hospital; Luciana Goncalves, PhD - Brigham and Women’s Hospital; Lisa Herlihy, MSN, RN - Salem Hospital; Veysel Baris, RN, PHD - Brigham and Women’s Hospital; Jacqueline Massaro; Jacqueline Massaro, MSN, RN - Brigham and Women’s Hospital; Beth Melanson, MS, RN, ACNS-BC, CWOCN, CCRN - Brigham and Women’s Hospital; Lori Morrow, RN, CWOCN - Salem Hospital; Paula Wolski, MSN, RN, NI-BC - Brigham and Womens Faulkner Hospital; Wenyu Song, PhD - Brigham and Women's Hospital, Harvard Medical School; Patricia Dykes, PhD, RN - Brigham and Women’s Hospital, Harvard Medical School;

A Human Factors Approach to Designing for Human-AI Teaming: The Case of an Emergency Department-based Clinical Decision Support Tool to Prevent Community Falls of Older Adults

Presentation Time: 03:00 PM - 03:15 PM

Abstract Keywords: Clinical Decision Support, Human-computer Interaction, Machine Learning, Governance of Artificial Intelligence, User-centered Design Methods
Primary Track: Applications
Programmatic Theme: Clinical Informatics

Human Factors and Cognitive Systems Engineering (HF/CSE) has long been successfully applied in safety-critical industries such as nuclear power and aviation to achieve remarkable augmentation of human operators. To demonstrate the application of HF/CSE in healthcare, we present our work developing an emergency department-based clinical decision support tool to prevent future falls of older adults. We share specific HF/CSE principles leveraged in design and implementation and reflect on the implications for designing for Human-AI teaming.

Speaker(s):
Hanna Barton, PhD
University of Wisconsin-Madison

Author(s):
Hanna Barton, PhD - University of Wisconsin-Madison; Erkin Otles; Apoorva Maru, B.S./B.A. - University of Wisconsin-Madison; Olivia Lin, N/A - University of Wisconsin-Madison; Lydia H. Malen, N/A - University of Wisconsin - Madison; Margaret A. Leaf, MS - UW Health; Daniel Hekman, MS; Douglas A. Wiegmann, PhD - University of Wisconsin - Madison; Brian Patterson, MD MPH - University of Wisconsin-Madison;