Abstract Number: Esoc2026a2101 Development and Validation of a Care Pathway Quality Indicator for Stroke Using the French National Health Data System: Immediate Post-Stroke Referral to Rehabilitation Deliqua-CNV Project

Abstract Background and aims Stroke remains a leading cause of long-term disability worldwide with vital and functional outcomes dependent on timely access to post-acute rehabilitation. In France, access to rehabilitation care remains insufficient and is marked by pronounced socio-geographical disparities. This study aimed to develop and validate a national healthcare pathway quality indicator measuring immediate referral to specialized rehabilitation services following acute stroke hospitalization, using the French National Health Data System. Methods Indicator components were defined based on stroke management guidelines and input from a multidisciplinary expert group. Validation involved three analyses: comparison with a gold standard dataset (Nouvelle-Aquitaine Stroke Observatory ObA2, period 2019-2021, 18 hospitals) across all subpopulations studied (socioeconomic and clinical patient characteristics, hospitalization and hospital features, COVID-19 period), assessment of inter-hospital variability, and evaluation of temporal trends. Results Among 10,089 stroke patients of the study sample, 25.9% (95% CI: 25.1–26.8) were immediately referred to rehabilitation services, 24.0% (95% CI: 23.1–24.8) after ischemic stroke and 43.9% (95% CI: 40.8–47.0) after hemorrhagic stroke. The indicator showed high performance in comparison with the gold standard (PPV=0.93, NPV=0.99, sensitivity=0.99, specificity=0.98), consistent across stroke types and subpopulations. Referral practices varied significantly between hospitals, while no temporal changes were observed during the COVID-19 period. Conclusions This indicator demonstrates strong potential for nationwide implementation to monitor post-stroke rehabilitation referral trends and detect geographical disparities. This study highlights the value of combining large-scale health administrative data with clinical registries for validation and supports improvement of stroke care pathways. Conflict of interest FADOUA EL HAFA. NOTHING TO DISCLOSE