Filling quality of the reports of adverse drug reactions received at the Pharmacovigilance Centre of São Paulo (Brazil): missing information hinders the analysis of suspected associations

Abstract

BACKGROUND: The completeness and accuracy of the reports of suspected adverse drug reactions is important in pharmacovigilance. The aim of the present study was to analyze the quality of the information included in the reports sent to the Pharmacovigilance Centre of São Paulo (Brazil). RESEARCH DESIGN AND METHODS: A sample of 999 reports received from January 2013 to December 2014 was selected. The quality of the filled information was evaluated according to a 'sufficiency' criterion to apply the Karch-Lasagna causality algorithm. RESULTS: There were 820 reports from manufacturers and 179 from health centres. Only 4.4% (44) were fully filled, thus allowing the adequate analysis of the causal relationship between the suspected medication and the adverse event. In 30% of the reports from manufacturers, the information about the critical variables was lacking or incomplete, preventing the adequate evaluation of the report. It was also noted that the reports' poor filling quality was not related with less severity or with old and well-known medicines. CONCLUSIONS: The poor quality of the information included in the reports received by this centre, especially those sent by pharmaceutical manufacturers, hampers the identification of potential safety signals. Measures to improve the quality of the reports must be urgently adopted.