Follow-up of Elderly Subjects With Falls Recruited in the Geriatric Departments of Several French Hospitals

Falls are the leading cause of accidental death among the elderly, with nearly 8,000 deaths per year in France, including more than 950 deaths on average per year in the Northwest interregion between 2006 and 2013. The incidence of falls increases with age. Falls are often the consequence of multiple factors: extrinsic and intrinsic. The WHO classifies these risk factors into 4 groups: environmental, socio-economic, behavioral and biological. The best way to act to reduce falls is to identify the risk and severity factors (risk of fracture, risk of loss of autonomy, risk of death) in order to correct those that are modifiable. On a national scale, the prevention of falls has been considered for many years as a major issue in health prevention \[INPES 2005\]. The management of elderly patients who fall has been the subject of recommendations for good practice by the French National Authority for Health (HAS) in 2009. These recommendations include the need to look for signs of geriatric severity of falls. The SNDS National System of healthcare data brings together and links the main national health databases in France. Linking data from the PMSI (Programme de Médicalisation des Systèmes d'Information) and the National Inter-regime Information System of the French National Health Insurance with data from death certificates via the SNDS opens up unique perspectives for improving the quality of follow-up of these patients. Crossing data from the Fall assessment clinic (clinical data) and SNDS data will allow data completeness and will improve our knowledge of the care consumption of elderly fallers. In particular, we wish to determine the risk factors for unplanned hospitalizations of patients with falls. This project aims to set up a common multicenter cohort of patients assessed during the day hospital for multidisciplinary fall assessment in 12 geriatric services: the university hospitals of Lille, Amiens-Picardie, Rouen, Caen, Tours, Strasbourg, Angers, Dijon, th