Current estimates suggest that there are between 1 and 3 million rare disease patients in France. Nevertheless, data on rare diseases is still too scattered in various databases and is heterogeneous, which hinders research progress. Many avenues remain to be explored in order to better understand these diseases and thus to better diagnose them, to propose adapted and targeted treatments for each patient, to limit the medical costs, etc. The DROMOS project aims to provide a detailed description of the actual care of rare disease patients for the most frequent diagnoses on a national scale. This knowledge will help to improve the care pathways of affected people and to adapt the care offer.

Because rare diagnoses are not accurately coded in medico-admistrative database, this projet will use the National Data Bank for Rare Diseases linked to the French National Health Insurance data. This matching will make it possible to describe for the first time the care of rare disease patients on a national scale for a very large number of disease. In this presentation, I will discuss two main challenges to achieve such a goal, i.e. firstly available methods for linkage with the French National Health Insurance data and possible resulting biases and secondly methods to model clinical course from patients followed-up over different age periods and their limits.