⚡ Quick Summary
A recent study developed the Emergency Department Dementia Algorithm (EDDA), utilizing machine learning to identify older adults (65+) at risk of dementia during emergency department visits. The algorithm demonstrated a remarkable AUROC of 0.85 in the test set, highlighting its potential to enhance clinical decision-making and patient outcomes.
🔍 Key Details
- 📊 Dataset: 759,665 emergency department visits
- 🧩 Features used: Electronic health record data
- ⚙️ Technology: Machine learning models (XGBoost, Random Forest, LASSO)
- 🏆 Performance: AUROC of 0.85 in test set, 0.93 in validation set
🔑 Key Takeaways
- 🧠 EDDA is designed to detect dementia in older adults during emergency visits.
- 💡 Machine learning techniques were employed to analyze extensive electronic health records.
- 📈 Positive-unlabeled learning significantly improved the algorithm’s performance.
- 🤝 Moderate agreement (kappa = 0.50) was found between EDDA and clinician diagnoses.
- 🔍 17% of patients identified as EDDA-positive had undiagnosed probable dementia.
- 🏥 Real-time implementation of EDDA could enhance patient safety and care transitions.
- 🌍 Study conducted across multiple sites within Yale New Haven Health from 2014 to 2022.
📚 Background
Dementia poses a significant challenge in emergency care, often going undetected due to the acute nature of visits. Traditional assessment methods may overlook cognitive impairments, leading to inadequate care. The integration of artificial intelligence in clinical settings offers a promising avenue for improving dementia detection, ultimately enhancing patient safety and care coordination.
🗒️ Study
This study was conducted as a multisite retrospective analysis of 759,665 emergency department visits from 2014 to 2022 at Yale New Haven Health. Researchers aimed to develop the Emergency Department Dementia Algorithm (EDDA) to identify older adults at risk of dementia, utilizing electronic health record data to train various machine learning models, including XGBoost, Random Forest, and LASSO.
📈 Results
The EDDA achieved an impressive AUROC of 0.85 in the test set and an even higher 0.93 in the validation set. The use of positive-unlabeled learning techniques significantly enhanced the algorithm’s performance. Furthermore, the study revealed that 17% of patients identified as EDDA-positive had previously undiagnosed probable dementia, underscoring the algorithm’s potential impact on clinical practice.
🌍 Impact and Implications
The findings from this study could revolutionize how dementia is detected in emergency settings. By implementing the EDDA, healthcare providers can improve patient outcomes and streamline care transitions. The algorithm’s design focuses on balancing detection accuracy with ease of implementation, making it a valuable tool for clinicians in busy emergency departments.
🔮 Conclusion
The development of the Emergency Department Dementia Algorithm (EDDA) represents a significant advancement in the use of machine learning for dementia detection. With its high performance and potential for real-time application, EDDA could transform emergency care for older adults, ensuring timely and appropriate interventions. Continued research and implementation of such technologies are essential for enhancing healthcare delivery.
💬 Your comments
What are your thoughts on the integration of artificial intelligence in dementia detection? We would love to hear your insights! 💬 Leave your comments below or connect with us on social media:
Detection of emergency department patients at risk of dementia through artificial intelligence.
Abstract
INTRODUCTION: The study aimed to develop and validate the Emergency Department Dementia Algorithm (EDDA) to detect dementia among older adults (65+) and support clinical decision-making in the emergency department (ED).
METHODS: In a multisite retrospective study of 759,665 ED visits, electronic health record data from Yale New Haven Health (2014-2022) were used to train three supervised and semi-unsupervised positive-unlabeled machine learning models (XGBoost, Random Forest, LASSO). A separate test set of 400 ED encounters underwent adjudicated chart review for validation.
RESULTS: EDDA achieved an area under the receiver-operating characteristic curve (AUROC) of 0.85 in the test set and 0.93 in the validation set. Positive-unlabeled learning improved performance. Agreement between EDDA and clinician-adjudicated dementia diagnoses was moderate (kappa = 0.50), with 17% of EDDA-positive patients having undiagnosed probable dementia.
DISCUSSION: EDDA enhances dementia detection in the ED, with potential for real-time implementation to improve patient outcomes and care transitions.
HIGHLIGHTS: Developed a machine learning algorithm using electronic health record data to detect dementia in the emergency department (ED). Algorithm designed to balance detection accuracy with ease of ED implementation. Parsimonious model with limited but predictive variables selected for rapid ED use. Focused on real-time application, optimizing ED workflows, and clinician support. Aims to enhance ED dementia detection, patient safety, and care coordination.
Author: [‘Cohen I’, ‘Taylor RA’, ‘Xue H’, ‘Faustino IV’, ‘Festa N’, ‘Brandt C’, ‘Gao E’, ‘Han L’, ‘Khasnavis S’, ‘Lai JM’, ‘Mecca AP’, ‘Sapre AV’, ‘Young J’, ‘Zanchelli M’, ‘Hwang U’]
Journal: Alzheimers Dement
Citation: Cohen I, et al. Detection of emergency department patients at risk of dementia through artificial intelligence. Detection of emergency department patients at risk of dementia through artificial intelligence. 2025; 21:e70334. doi: 10.1002/alz.70334