2.3 Statistical analysis
Disproportionality analysis is a data mining algorithm extensively employed in the global monitoring of ADRs. This algorithm demonstrates a notable level of sensitivity and is capable of reducing numerous biases. Each identified signal derived from this approach represents a statistical association between a specific drug and an AE. In this study, a combination of proportional reporting ration (PRR)[16] and reporting odds ratio (ROR)[17], both recognized for their heightened sensitivity, were utilized for signal mining. The screening criterias for PRR are: N≥3; χ2≥4; PRR≥2, and for ROR are: N≥3; lower limit of 95% CI>1. The signal value obtained from these two algorithms directly represents the strength of statistical correlation between an AE and the drug. AEs satisfied with both PRR and ROR criteria were subjected to further analysis.