Disproportionality Analysis of Drug-AE Pairs

A data-mining procedure using a reporting odds ratio (ROR) method (Min et al., 2018 (link); Moreland-Head et al., 2021 (link)) was introduced to investigate the disproportionality in reporting ratio caused by interested drug–AE pairs compared with a random drug–AE pair, which were then evaluated in tandem with a Bayesian confidence propagation neural network (BCPNN) method (Bate et al., 1998 (link)), thereby deducing the association between the target drug and event by a prior possibility. The association between diabetes and AEs was also investigated. Drug–AE pairs that could generate stronger signals than the same AEs paired with diabetes were screened out and demonstrated with a heatmap. Osteomyelitis was picked as the major candidate before lower extremity amputation. Data processing was conducted with RStudio 4.1.2, using a logistic regression model. For ROR, a signal was determined as the count of drug–AE pairs greater than 3, plus the value of the ROR higher than 1, and the lower limit of the 95% confidence interval (95% CI) exceeding 1. For BCPNN, a signal was defined as the value of the lower limit of information component (IC₀₂₅) exceeding 0; specifically, an IC₀₂₅ value between 0 and 1.5 was defined as a weak signal, an IC₀₂₅ value between 1.5 and 3 was considered as a medium signal, and an IC₀₂₅ value > 3 was considered as a strong signal.