Utilizing temporal pattern of adverse event reports to identify potential late-onset adverse events

Objectives: Through the use of FDA adverse event reporting system (FAERS) dataset, this study analyzes the pattern of time-to-event (TTE) for drugs and adverse events, and suggest ways to identify candidate late-onset events for monitoring. Methods: The duration between administration date of the drug and the onset of adverse events was explored with using FAERS data from 2012–2021. The fold change of proportional reporting ratios or reporting odds ratios were calculated to identify enriched events in the later period and to suggest the late-onset events for further monitoring. To compare the findings, we used the claims database of the Korean National Health Insurance Service (NHIS). Results: A total of 1,426,781 reports were included. The median TTE was 10 days (interquartile range [IQR]: 0–98 days), with 11.5% (n = 164,093) reporting events that occurred at least one year after administration. TTE and fold change analysis captured historical cases of late-onset events, while generating an additional less-explored list of events. The results for tumor necrosis factor (TNF) inhibitors were compared using the NHIS dataset. Conclusion: Our study provides a comprehensive analysis of the FAERS dataset, focusing on TTE data. Periodic summarization of reports would be helpful in monitoring the late-onset events.