Table 1.
Comparison of penalties of the original vigiGrade completeness score and penalties of the automated Vigi4Eudra-score for line listings from EudraVigilance.
Fig 1.
Flowchart: identification of reports for all four datasets.
Legend: n = number of reports; SMQ: Standardized MedDRA Queries are pre-defined collections of MedDRA codes (in this case potential ADRs) corresponding to a specific diagnosis or condition (in this case Anaphylaxis) [34].
Table 2.
Distribution of values of the Vigi4Eudra-score in the four exemplary datasets.
Fig 2.
Violin plot and box plot of all key figures for the values of the Vigi4Eudra-score.
Legend: Violin plot showing the distribution of the values of the Vigi4Eudra-score for each dataset. Grey dots and numbers indicate the mean value. White numbers indicate key values of the 25% quantile, median, and 75% quantile displayed by the box plot. Small black dots indicate the value(s) of each report. White dots indicate outliers. The red dotted line shows the threshold of well-documented reports ≥0.8. The numbers and percentual shares for all reports depicted above the threshold fulfilling the well-documented criterion of each dataset are indicated in red.
Fig 3.
Differences between the automatically calculated values of the Vigi4Eudra-score and the values of the respective reference.
Legend Fig 3: Bar chart showing the distribution of the differences between the values of the Vigi4Eudra-score and the vigiGrade completeness score. Above each bar, the total number of reports included and its percentual share within the respective dataset is indicated.