IS IT POSSIBLE TO ACCURATELY IDENTIFY RARE DISEASES USING NATIONAL HOSPITAL DISCHARGE DATABASES?
Author(s)
Bergot E1, de Léotoing L2, Bendjenana H3, Tournier C2, Vainchtock A2, Nachbaur G3, Humbert M4
1CHU Côté de Nacre, Caen, France, 2HEVA, LYON, France, 3GSK, Marly-le-Roi, France, 4Hôpital de Bicêtre, Le Kremlin Bicêtre, France
OBJECTIVES: Generation of accurate data in a timely manner is often challenging especially for rare diseases. Data from administrative discharge databases, as an alternative to prospective studies, may be important and value but has methodological limitations. This study (HO-15-16391, funded by GSK) aimed at evaluating the feasibility of describing the hospital burden of Pulmonary Arterial Hypertension (PAH) in France using the French National Hospital discharge database (PMSI). METHODS: An exploratory analysis was performed from the PMSI databases. All 2013 hospital stays with ICD-10 codes for PAH (I27.0, I27.2) whether principal or associated diagnosis were extracted. Only incident adult patients were selected (not hospitalized with PAH in 2011-2012) and followed during 1 year. RESULTS: CONCLUSIONS: The PMSI database is an accessible and relevant source of data, especially for diseases mainly managed in hospital. Its exhaustiveness allows to overcome the low frequency of events that is specific to rare diseases and thus to gain a representative picture of the hospital burden of disease. Nevertheless, clinical information is limited and requires assumptions validated by medical experts.
Conference/Value in Health Info
2017-11, ISPOR Europe 2017, Glasgow, Scotland
Value in Health, Vol. 20, No. 9 (October 2017)
Code
PRM60
Topic
Real World Data & Information Systems
Topic Subcategory
Reproducibility & Replicability
Disease
Cardiovascular Disorders