Negativity and Positivity in the ICU: Exploratory Development of Automated Sentiment Capture in the Electronic Health Record

Critical Care Explorations · 15 authors, 11 centres

AI SUMMARY

FIDELITY 100%

This summary was generated by AI from a single paper. It has not been reviewed by a clinician and is not clinical advice. Verify against the source before acting on it.

AGENT 01 · STRUCTURED SUMMARY

TRACE TO SOURCE

“

TL;DR

This study developed and validated two healthcare-specific sentiment analysis models (a keyword-based lexicon and a deep learning model) to detect provider sentiment in ICU clinical notes, finding they performed substantially better than general-purpose sentiment tools.

Full summary

700 CHARS

The study conducted a retrospective observational analysis of 198,944 clinical notes from 52,997 ICU admissions in the MIMIC-III database, with replication on an external UCSF sample. The authors validated these against manually labeled note fragments. The key result was that both clinical sentiment models demonstrated high correlation (Spearman 0.62–0.84, p < 0.00001) with labeled sentiment, significantly outperforming three general language sentiment algorithms (correlation 0.28–0.46). The implication is that domain-specific tools are necessary for accurate sentiment capture in clinical notes, which may have future applications in prognostication and understanding clinical decision-making.

STUDY TYPE

cohort

SPECIALTY

critical care

SUMMARISED BY

AI pipeline

FIDELITY CHECK

100% · A

ORIGINAL PAPER

PMC10519480

Negativity and Positivity in the ICU: Exploratory Development of Automated Sentiment Capture in the Electronic Health Record

Critical Care Explorations

READ THE FULL PAPER

ABSTRACT

Abstract OBJECTIVES: To develop proof-of-concept algorithms using alternative approaches to capture provider sentiment in ICU notes. DESIGN: Retrospective observational cohort study. SETTING: The Multiparameter Intelligent Monitoring of Intensive Care III (MIMIC-III) and the University of California, San Francisco (UCSF) deidentified notes databases. PATIENTS: Adult (≥18 yr old) patients admitted to the ICU. MEASUREMENTS AND MAIN RESULTS: We developed two sentiment models: 1) a keywords-based approach using a consensus-based clinical sentiment lexicon comprised of 72 positive and 103 negative phrases, including negations and 2) a Decoding-enhanced Bidirectional Encoder Representations from Transformers with disentangled attention-v3-based deep learning model (keywords-independent) trained on clinical sentiment labels. We applied the models to 198,944 notes across 52,997 ICU admissions in the MIMIC-III database. Analyses were replicated on an external sample of patients admitted to a UCSF ICU from 2018 to 2019. We also labeled sentiment in 1,493 note fragments and compared the predictive accuracy of our tools to three popular sentiment classifiers. Clinical sentiment terms were found in 99% of patient visits across 88% of notes. Our two sentiment tools were substantially more predictive (Spearman correlations of 0.62–0.84, p values < 0.00001) of labeled sentiment compared with general language algorithms (0.28–0.46). CONCLUSION: Our exploratory healthcare-specific sentiment models can more accurately detect positivity and negativity in clinical notes compared with general sentiment tools not designed for clinical usage. Keywords: computer-assisted decision-making, critical care, critical care outcomes, natural language processing, sentiment analysis

CITATION EXPORT

Chris J Kennedy, Catherine Chiu, Allyson Cook Chapman et al.. Negativity and Positivity in the ICU: Exploratory Development of Automated Sentiment Capture in the Electronic Health Record. Critical Care Explorations. (2023). https://doi.org/10.1097/CCE.0000000000000960

AUTHORS

15 · 11 CENTRES

Chris J Kennedy, Catherine Chiu, Allyson Cook Chapman, Oksana Gologorskaya, Hassan Farhan, Mary Han, MacGregor Hodgson, Daniel Lazzareschi, Deepshikha Ashana, Sei Lee, Alexander K Smith, Edie Espejo, John Boscardin, Romain Pirracchio, Julien Cobert

1 Department of Psychiatry, Harvard Medical School, Boston, MA.
2 Center for Precision Psychiatry, Massachusetts General Hospital, Boston, MA.
3 Department of Anesthesia and Perioperative Care, University of California San Francisco, San Francisco, CA.
4 Critical Care and Palliative Medicine, Department of Internal Medicine, University of California San Francisco, San Francisco, CA.
5 Department of Surgery, University of California San Francisco, San Francisco, CA.
6 Bakar Computational Health Sciences Institute, University of California San Francisco, San Francisco, CA.
7 Department of Anesthesiology, Perioperative and Pain Management, Stanford University, Stanford, CA.
8 Division of Pulmonary, Allergy, and Critical Care Medicine, Duke University, Durham, NC.
9 Division of Geriatrics, Department of Medicine, University of California San Francisco, San Francisco, CA.
10 Geriatrics, Palliative, and Extended Care, Veterans Affairs Medical Center, San Francisco, CA.
11 Department of Anesthesia, Anesthesia Service, San Francisco VA Health Care System, San Francisco, CA.