PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
November 27, 2025Journal of Medical Internet ResearchOpen Access

Detection of Antithrombotic-Related Bleeding in Older Inpatients: Multicenter Retrospective Study Using Structured and Unstructured Electronic Health Record Data

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Bleeding complications are a major contributor to adverse drug events in older inpatients receiving antithrombotic agents, making timely and accurate detection essential for drug safety surveillance and clinical risk management.

Does an integrated approach combining structured data algorithms and natural language processing improve the detection of hemorrhagic events in older hospitalized patients treated with antithrombotic agents?

Population

36,039 inpatient stays of patients 65 years or older receiving antithrombotic agents across 3 Swiss university hospitals

Comparison

Structured data algorithms, NLP, and their combination vs a manually reviewed gold standard

Design

Multicenter retrospective validation study

Discussion

Loading...

Member takes

Overview

May enhance bleeding surveillance in antithrombotic-treated inpatients; leaves open prospective validation before EMR integration.

Structured PICO

Does an integrated approach combining structured data algorithms and natural language processing improve the detection of hemorrhagic events in older hospitalized patients treated with antithrombotic agents?

P
Population
Patients 65 years or older who received at least one antithrombotic agent and were hospitalized between January 2015 and December 2016 (n=36,039 inpatient stays for internal validation, n=24,054 stays for external validation).
I
Intervention
Automated algorithms combining structured data-based rule models (SDA) and a natural language processing (NLP) approach applied to electronic medical records.
C
Comparator
Manually reviewed gold standard of 754 electronic medical records.
O
Outcome
Detection of major bleeding (MB) and clinically relevant nonmajor bleeding (CRNMB) events.safety

Combining structured data algorithms with natural language processing enhances the automated detection of antithrombotic-related bleeding events in older hospitalized patients, which can improve drug safety surveillance.

Cite This Study

A 2025 study studied this question.

synapsesocial.com/papers/6a8c7e10de8837d52d130acfhttps://doi.org/10.2196/77809
View Full Paper
Ask AI
Bookmark
Share