Related Experiment Videos
Evaluation of an algorithm to identify incident breast cancer cases using DRGs data
1Medical Information, Epidemiology and Biostatistics, Hôpital Nord, Place Pauchet, Amiens University Hospital, 80 054 Amiens Cedex 1, France. ganry.olivier@chu-amiens.fr
Summary
The French hospital database (Programme de médicalisation des systèmes d'information) shows high accuracy for identifying breast cancer cases. This cancer data source offers potential for incidence assessment, especially when including secondary diagnoses.
Area of Science:
- Oncology
- Health Informatics
- Public Health
Background:
- Hospital databases can be cost-effective, timely, and nationally representative for cancer information.
- The French hospital database, Programme de médicalisation des systèmes d'information (PMSI), adapted from Diagnosis Related Group (DRG) classification, was evaluated.
Purpose of the Study:
- To examine the utility of the PMSI database as an independent source for identifying incident breast cancer cases.
- To assess the sensitivity, specificity, and positive predictive value (PPV) of the PMSI database against a cancer registry.
Main Methods:
- Identified women diagnosed with breast cancer in the PMSI database from public hospitals in the Somme area, France (1998).
- Matched PMSI data with women in the Somme cancer registry diagnosed in the same year.
- Utilized an algorithm to detect cancer-related diagnoses and procedures within the PMSI database.
Main Results:
- PMSI demonstrated 85% sensitivity, 99.9% specificity, and 97% PPV for breast cancer as a principal diagnosis.
- Sensitivity increased by 9% for secondary diagnoses, though PPV decreased to 78%.
Conclusions:
- The PMSI database shows significant potential for assessing breast cancer incidence due to high sensitivity and specificity.
- Further studies are needed to confirm these findings before widespread use, particularly in regions lacking cancer registries.