55

Introduction: Spontaneous reporting remains one of the cornerstones of post-marketing drug safety surveillance. One of its main limitations is a lack of completeness.The main aim of this study was to assess the completeness of pharmacovigilance reports sent by general practitioners (GPs) to regional pharmacovigilance centers (RPC) reported in the French pharmacovigilance database (FPVD). Secondary aim was to identify factors associated with complete reports. Method: All adverse drugs reactions (ADRs) sent by GPs in France in 2015 were analyzed. According to information provided in ADR reports (ADR, date of occurrence, clinical description, drugs suspected, etc.), completeness was analyzed from “mandatory” criteria (age, gender, ADR and suspected drug(s)) and “non-mandatory” criteria (medical history, concomitant drugs, symptoms evolution and complementary exams) and classified as “well-documented”, “slightly-documented” or “poorly-documented”. Results: In 2015, the FPVD contained 3,020 ADR reports realized by GPs. Only 16.4% of these reports were classified as “well-documented”, in accordance with study criteria. The most poorly documented items were concomitant drugs (41.4%) and complementary exams (37.4%). An association between a “well-documented” ADR report and its “seriousness” (OR = 3,02 [95% CI 2,44; 3,23], P < 10–3) and elderly compared to adults (OR = 1,76 [95% CI 1,42; 2,18], P < 10–3) or children (OR = 4,59 [95% CI 2,51; 8,39], P < 10–3). Conclusion: Our study shows that only one out of six ADR reports was “well-documented”. It appears to be important to promote pharmacovigilance to improve completeness of ADR reports.