Presentation (CTI)

OBJECTIVES: The SNDS is one of the leading claim databases in Europe, covering the entire French population and patient pathways. Despite its advantages, evidence generation using the SNDS is constrained by the limited ability to select specific populations due to the lack of detailed ICD-10 coding, and potential coding delay. The use of NLP to analyze patients' electronic health records (EHR) to identify these gaps could significantly improve SNDS studies without requiring time-consuming and costly human review. The aim of this study was to evaluate the limitations of the SNDS for studying a specific subpopulation such as HER2+ metastatic breast cancer (BC).
METHODS: Retrospective analysis on EHR and SNDS in BC patients treated in 3 general hospitals and 1 university hospital between January 1st, 2017, and December 31st, 2021. The comparison of HER2+ metastatic BC patient counts in EHR (using REALLI) and SNDS was performed. The delay between metastasis diagnosis dates in both sources was described.
RESULTS: Between January 2017 and December 2021, 3214 BC patients were included. Metastatic status was identified for 1045 and 913 patients in EHR and SNDS respectively. HER2+ status was identified for 67 and 48 patients in EHR and SNDS respectively (HER2+ status was only performed on 749 patients as HER2+ patients were not treated in the university hospital). This represents an underestimation of 13% and 28% of metastatic and HER2+ status among BC patients in the SNDS, respectively. Among the 671 patients with a metastatic diagnosis date in both EHR and SNDS, 91 (12%) had a coding delay, including 31 (5%) with a delay of more than 12 months
CONCLUSIONS: Leveraging EHRs using REALLI has highlighted the underrepresentation of patients with HER2+ metastatic BC and delays in coding in the SNDS. Further studies involving various medical centers are necessary to validate the results, considering variability in coding practices.