[Which definition and taxonomy of incident to use for a French reporting system in primary care settings?]. / Quelle définition et taxonomie utiliser pour un recueil d'événements indésirables associés aux soins primaires en France?

BACKGROUND: There is no widely accepted definition of incident for primary care doctors in France and no taxonomic classification system for epidemiological use. In preparation for a future epidemiological study on primary care incidents in France (the ESPRIT study), this work was designed to identify the definitions and taxonomic classifications used internationally along with the usual methods and results in terms of frequency in the literature. The goal was to determine a French definition and taxonomy. DESIGN: Systematic review of the literature and consensus methods. METHOD: An exhaustive search of epidemiological surveys was performed. A structured grid was used. After having identified the definitions used in the literature, a definition was chosen using the focus groups method. Taxonomies identified in the literature were classified by relationship, architecture, code number, and number of studies published. Subsequently, a consensus among experts, who independently tested these taxonomies on six incidents, was reached for choosing the most appropriate for epidemiological data collection (little information on a large number of cases). RESULTS: Twenty-four papers reporting 17 studies were selected among 139 articles. Five definitions and eight taxonomies were found. The chosen definition of incident was based on the WHO definition "A patient safety incident is an event or circumstance that could have resulted, or did result, in harm to a patient, and whose wish it is not repeated again". The test of incidents resulted in the choice of the TAPS version of the International Taxonomy of Medical Error in Primary Care for a reproducible and internationally recognized codification and the tempos method for its current use in French general practice. DISCUSSION: The definitions, taxonomies, data collection characteristics and frequency of incidents results in the international literature on incidents in primary care are key components for the preparation of an epidemiological survey on incidents in primary care.