Prediction of emergency department patient disposition based on natural language processing of triage notes

Sterling, Nicholas W.; Patzer, Rachel E.; Di, Mengyu; Schrager, Justin D.

doi:10.1016/j.ijmedinf.2019.06.008

Cited by 85 publications

(78 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Nursing Informatics is defined as the "science and practice (that) integrates nursing, its information and knowledge, with the used for real-time bio-terrorism surveillance ( Chapman, Dowling, & Wagner, 2004) , and for epidemiological purposes such as identifying injury ( McKenzie, Scott, Campbell, & McClure, 2010) and drug use patterns (Indig, Copeland, Conigrave, & Rotenko, 2008). These studies have shown that triage narratives can be used in isolation ( Sterling, Patzer, Schrager, 2019) or in combination with ICD codes ( Horng, et al, 2017;Mitchell, Finch, Boufous, & Browne, 2009) for epidemiological research; with some research suggesting that the triage narratives may even be superior to coding or ICD codes for identifying clinical cases (Indig, Copeland, Conigrave, & Rotenko, 2009). Despite the time invested by Canadian triage nurses in recording these data during triage assessments, to date there has been no push for CIHI to include it as part of mandatory data reporting.…”

Section: Informatics Competency As a Catalyst For Change In Emergencymentioning

confidence: 99%

Advancing Emergency Nurses’ Leadership and Practice through Informatics

Picard

Kleib

2020

CJEN

View full text Add to dashboard Cite

Collection of data in healthcare is vitally important to inform healthcare resource planning and monitor effectiveness of care. The Canadian Emergency Department information System and Canadian Triage Acuity Scale are primary tools for collecting such data. Although emergency nurses use these tools to collect significant patient and healthcare data on daily basis, their understanding of the purposes and implications for collecting these data is sub-optimal. Furthermore, emergency nurses’ awareness about informatics, and the limited representation in information and communication technology strategic initiatives and research within Canadian emergency nursing are some barriers preventing nurses from realizing the full benefits of information and communication technology to improve patient and system outcomes, and nursing knowledge development. The National Emergency Nurses Association is well positioned to provide the leadership required to move nurses from being data collectors, to information users by maximizing their potential to advance Canadian emergency nursing practice through informatics.

show abstract

Section: Informatics Competency As a Catalyst For Change In Emergencymentioning

confidence: 99%

Advancing Emergency Nurses’ Leadership and Practice through Informatics

Picard

Kleib

2020

CJEN

View full text Add to dashboard Cite

show abstract

“…Let's mention the work by Lavertu and Altman [9] who designed the Redmed system in order to produce a specific drug lexicon to be used in social media applications. When clinical data are available and accessible to researchers, NLP technics may be applied on all types of textual data: electronic health records for the identification of adverse drug events [8], discharge summaries [10], and more rarely triage notes [11] for performing named entity recognition.…”

Section: Nlp and Application Contextsmentioning

confidence: 99%

“…The first objective, which is present in a large amount of published papers, addresses the improvement of the medical care process. Hence, papers published in 2019 focus on the identification of patients with obesity and several comorbidities from clinical texts [10], the prediction of emergency department patient disposition from triage notes [11], the identification of drug discontinuation events from EHR [12], and the help for monitoring patients in intensive care unit (ICU) [13].…”

Section: Original Issuesmentioning

confidence: 99%

A Year of Papers Using Biomedical Texts:

Grouin

Grabar

2020

Yearb Med Inform

View full text Add to dashboard Cite

Objectives: Analyze papers published in 2019 within the medical natural language processing (NLP) domain in order to select the best works of the field. Methods: We performed an automatic and manual pre-selection of papers to be reviewed and finally selected the best NLP papers of the year. We also propose an analysis of the content of NLP publications in 2019. Results: Three best papers have been selected this year including the generation of synthetic record texts in Chinese, a method to identify contradictions in the literature, and the BioBERT word representation. Conclusions: The year 2019 was very rich and various NLP issues and topics were addressed by research teams. This shows the will and capacity of researchers to move towards robust and reproducible results. Researchers also prove to be creative in addressing original issues with relevant approaches.

show abstract

“…The copyright holder for this preprint this version posted December 4, 2020. ; https://doi.org/10.1101/2020. 12 Table 2. Performance on the validation set for models using individual features (models a-n) and sets of features (models o and p).…”

Section: Model Explainabilitymentioning

confidence: 99%

Using explainable machine learning to identify patients at risk of reattendance at discharge from emergency departments

Chmiel¹,

Azor

Borca

et al. 2020

Preprint

View full text Add to dashboard Cite

Short-term reattendances to emergency departments are a key quality of care indicator. Identifying patients at increased risk of early reattendance could help reduce the number of missed critical illnesses and could reduce avoidable utilization of emergency departments by enabling targeted post-discharge intervention. In this manuscript we present a retrospective, single-centre study where we create and evaluate a machine-learnt classifier trained to identify patients at risk of reattendance within 72 hours of discharge from an emergency department. On a patient hold-out test set, our highest performing classifier obtained an AUROC of 0.749 and an average precision of 0.232; demonstrating that machine-learning algorithms can be used to classify patients, with moderate performance, into low and high-risk groups for reattendance. In parallel to our predictive model we train an explanation model, capable of explaining the predictions of the machine-learnt classifier at an attendance level. These explanations can be used to help understand the decisions a model is making, evaluate biases present in it's decisions and help inform the design of bespoke interventional strategies.

show abstract

Prediction of emergency department patient disposition based on natural language processing of triage notes

Cited by 85 publications

References 17 publications

Advancing Emergency Nurses’ Leadership and Practice through Informatics

Advancing Emergency Nurses’ Leadership and Practice through Informatics

A Year of Papers Using Biomedical Texts:

Using explainable machine learning to identify patients at risk of reattendance at discharge from emergency departments

Contact Info

Product

Resources

About