Disease-related adverse events following non-live vaccines: investigation of a newly described reporting bias through the analysis of the WHO Global ICSR Database, VigiBase.

Due to a vaccine-specific reporting bias, it has been recently suggested that some non-live vaccines may be preferentially reported as the suspected cause of the disease or symptoms/signs related to the disease they should prevent. The aim of this study was to analyse the WHO Global Individual Case Safety Report (ICSR) Database, VigiBase, in order to explore this newly described reporting bias in greater detail and to verify whether it can generate potentially misleading signals of disproportionate reporting (SDRs). Vaccines reports entered into VigiBase between 1990 and 2011 were extracted. Twelve non-combined non-live vaccines were chosen for analysis. Two distinct groups of MedDRA preferred terms (PTs) per vaccine were selected on the basis of the disease-specificity. Twenty-four vaccine-events pairs were obtained and referred as "misleading combinations". A descriptive analysis of the reports retrieved was performed. To verify whether a specific group of selected PTs was disproportionally reported in association with the vaccine representing the respective "misleading combination", the reporting odds ratio (ROR) was calculated for each of the 24 vaccine-event pairs retrieved considering all the other vaccines as the background. A total of 627,165 reports were analysed. Among ICSRs containing a "misleading combination", healthcare professionals were the most frequently noted (17%), though reporter type was unknown in 72% of the remaining reports. The reporting rate distribution of the 24 groups of PTs per vaccine included in the study showed that in 16 cases the highest rate was reached by the vaccine representing the respective "misleading combination". The application of the ROR yielded 21 SDRs out of 24 combinations tested. Findings from this study support the existence of a vaccine-specific reporting bias. Since this phenomenon may result in potentially misleading SDRs, professionals involved in vaccine safety surveillance should also consider this bias during the validation of such disproportional associations.