Quality of adverse drug reaction (QADRA) reports: an algorithm to appraise the efficiency of spontaneous reporting systems in pharmacovigilance

Aim: To design and validate an algorithm for the multidimensional evaluation of the quality of adverse drug reaction (ADR) case reports (QADRA). Subject and methods: One hundred fifty-three patients randomly and retrospectively selected from 15,906 records included in the Italian database of spontaneous ADR reports throughout 2009. Each report was evaluated by two panels of experts blinded to one another as well as by the algorithm developed in the present study. Each case was classified taking three parameters into consideration: plausibility, notoriety and clinical relevance. Results: The two panels assessed that 21.6 % of reports were of "high" quality. When applying the QADRA algorithm (score range 0-15), its median value was 6 (4-7, 25 and 75 percentiles, respectively). The area under the receiver operator characteristics (ROC) curve, which assesses the ability of the risk score to predict the report quality, was 0.93 (95 % CI: 0.88-0.97). Herein, the cut-off points ≤5, 6 or 7 and ≥8 indicated the best balance between sensitivity and specificity, and they could be used to categorize the reports as being of 'high', 'intermediate' and 'low' quality (AUC = 0.87; 95 % CI: 0.80-0.92), respectively. Conclusion: The QADRA algorithm performs as a reliable and complete tool for assessing the quality of ADR reports. Several potential applications of this algorithm should be investigated in the future, both for scientific purposes and healthcare system management. © 2013 Springer-Verlag Berlin Heidelberg.