Related Experiment Video
Updated: Nov 24, 2025
Inverse Probability of Treatment Weighting Propensity Score using the Military Health System Data Repository and National Death Index
Published on: January 8, 2020
Validation of novel identification algorithms for major adverse cardiovascular events in a Japanese claims database
Daisuke Shima1, Yoichi Ii2, Shingo Higa1
1Pfizer Japan Inc., Tokyo, Japan.
Insights
Developing algorithms using ICD-10 codes, drug prescriptions, and procedures accurately identifies major adverse cardiovascular events (MACEs) in Japanese claims data. This improves prediction of critical health outcomes from large datasets.
Area of Science:
- Cardiology
- Health Informatics
- Data Science
Background:
- Predicting rare, life-threatening clinical events from large claims databases is challenging due to low incidence and data complexity.
- Accurate identification of major adverse cardiovascular events (MACEs) is crucial for clinical outcome prediction and research.
- Existing methods relying solely on diagnostic codes may lack precision in claims data analysis.
Purpose of the Study:
- To develop and validate novel algorithms for identifying MACEs, including acute myocardial infarction and stroke, using Japanese healthcare claims data.
- To assess the incremental value of incorporating drug prescription and medical procedure data alongside diagnostic codes (ICD-10) for MACEs detection.
- To improve the accuracy and reliability of MACEs identification in administrative healthcare databases.
Main Methods:
- Algorithms were developed using International Classification of Diseases, Tenth Revision (ICD-10) codes, drug prescriptions, and medical procedures from Japanese claims data (Jichi Medical University Hospital, Oct 2012-Sep 2014).
- A random sample of 100 potential acute myocardial infarction cases and 200 potential stroke cases were extracted based on ICD-10 codes.
- An independent committee validated diagnoses, and algorithm performance was assessed using positive predictive values (PPVs) and apparent sensitivities.
Main Results:
- Initial algorithms based solely on ICD-10 diagnosis showed low PPVs for acute myocardial infarction (81.6%), ischemic stroke (31.0%), and hemorrhagic stroke (45.5%).
- Incorporating drug prescription and procedure data significantly elevated PPVs for all MACEs.
- Algorithms combining ICD-10 codes with event-specific prescription and procedure data achieved PPVs ranging from 70%-98% with apparent sensitivities over 50%.
Conclusions:
- Algorithms integrating ICD-10 diagnoses with specific drug prescription and medical procedure data demonstrate validity for identifying MACEs in Japanese claims databases.
- This multi-modal data approach enhances the accuracy of MACEs detection compared to using diagnostic codes alone.
- The validated algorithms offer a reliable method for predicting clinical outcomes from large-scale administrative healthcare data.
Abstract:
Predicting clinical outcomes can be difficult, particularly for life-threatening events with a low incidence that require numerous clinical cases. Our aim was to develop and validate novel algorithms to identify major adverse cardiovascular events (MACEs) from claims databases. We developed algorithms based on the data available in the claims database International Classification of Diseases, Tenth Revision (ICD-10), drug prescriptions, and medical procedures. We also employed data from the claims database of Jichi Medical University Hospital, Japan, for the period between October 2012 and September 2014. In total, we randomly extracted 100 potential acute myocardial infarction cases and 200 potential stroke cases (ischemic and hemorrhagic stroke were analyzed separately) based on ICD-10 diagnosis. An independent committee reviewed the corresponding clinical data to provide definitive diagnoses for the extracted cases. We then assessed the algorithms' accuracy using positive predictive values (PPVs) and apparent sensitivities. The PPVs of acute myocardial infarction, ischemic stroke, and hemorrhagic stroke were low only by diagnosis (81.6% [95% CI 72.5-88.7]; 31.0% [95% CI 22.8-40.3]; and 45.5% [95% CI 34.1-57.2], respectively); however, the PPVs were elevated after adding the prescription and procedure data (87.0% [95% CI 78.3-93.1]; 44.4% [95% CI 32.7-56.6]; and 46.1% [95% CI 34.5-57.9], respectively). When we added event-specific prescription and procedure data to the algorithms, the PPVs for each event increased to 70%-98%, with apparent sensitivities exceeding 50%. Algorithms that rely on ICD-10 diagnosis in combination with data on specific drugs and medical procedures appear to be valid for identifying MACEs in Japanese claims databases.
More Related Videos
06:16Signal Acquisition, Score Interpretation, and Economics of a Non-Invasive Point-of-Care Test for Coronary Artery Disease
Published on: August 9, 2024
05:47Evidence-based Knowledge Synthesis and Hypothesis Validation: Navigating Biomedical Knowledge Bases via Explainable AI and Agentic Systems
Published on: June 13, 2025
Related Concept Videos
01:30Acute Coronary Syndrome III: Diagnostic Studies
01:18Cardiovascular Drugs: Classification based on Therapeutic Indications
01:26Coronary Artery Disease IV: Preventive Measures
01:20Blood Studies for Cardiovascular System I: Cardiac Biomarkers
The essential diagnostic tools for detecting myocardial necrosis and monitoring individuals suspected of having acute coronary syndrome (ACS) include:
Troponins
Troponins, particularly cardiac troponins I and T, are the most precise and sensitive markers of myocardial injury. They are detectable within 4-6 hours of myocardial injury and remain...
01:21Peripheral Arterial Disease II: Clinical Manifestations and Diagnostic Evaluation
01:19Pharmacovigilance
This process, termed pharmacovigilance, aims to detect, evaluate, and minimize harmful effects related to medication use. The data collection for pharmacovigilance depends on spontaneous reporting systems, where healthcare professionals or patients voluntarily report suspected ADRs.
In some cases, there...