John Libbey Eurotext

Environnement, Risques & Santé

MENU

Epidemiologic utility of gastroenteritis indicators derived from French health insurance data Volume 5, issue 5, Septembre-Octobre 2006

Figures

See all figures

Authors
Institut de veille sanitaire (InVS), 14, rue du Val d’Osne, 94415 Saint-Maurice cedex, Laboratoire de pharmacie galénique, ADEN EA3234, 22, bd Gambetta, 76183 Rouen
  • Key words: data interpretation, statistical, epidemiologic measurements, gastroenteritis, France, prescriptions, drug
  • Page(s) : 373-82
  • Published in: 2006

The computerized system established by the French national health insurance fund in 1998 to process reimbursement claims has made it possible to set up a database of prescription drug purchases. This study seeks to assess the epidemiologic value of some indicators of daily incidence of acute gastroenteritis, derived from these data. We used 2 datasets, one of routine information for 14 urban and rural areas (2002-2003) and the other from a region where a waterborne outbreak of gastroenteritis took place (Isère, 2002). We studied the behavior of the gastroenteritis incidence indicators, focusing on spatial variations and time patterns. We also compared them with indicators from independent sources, including the French "Sentinelle" network of general practitioners and the retrospective cohort study conducted for the outbreak investigation. We finally chose 2 indicators from the drug reimbursement data, each based on a separate case definition. The first was more specific (annual incidence rate of 79 cases per 1000) and was consistent with data from both independent sources (incidence rate levels, seasonal patterns and impact of the outbreak). The incidence rate for the second definition, which included milder diseases and less specific symptoms, was twice that of the first. Data from the French national health insurance database cover almost the entire French population and are specifically appropriate for studying the epidemiology of waterborne gastroenteritis, including time series studies and outbreak investigations.