Cross-National Pharmacovigilance Data Pooling Accelerates Drug Safety Signal Detection: Evidence from Change-Point Analysis of Australia’s DAEN and the US FAERS

<b>Aim:</b> To test whether change-point detection (CPD) identifies drug safety signals earlier than standard disproportionality analysis, and whether pooling data from two national databases improves detection timing. <b>Methods:</b> An ensemble of three CPD algorithms — PELT, CUSUM, and Bayesian Online Change Point Detection (BOCPD) — was applied to quarterly drug-specific adverse event reporting proportions in Australia’s DAEN (664,747 reports, 2004-2025). Rolling Proportional Reporting Ratio (PRR) served as comparator. Performance was validated against 32 TGA regulatory actions, 40 curated reference associations, and OMOP/EU-ADR reference standards (111 positive controls, 87 negative controls). Parallel time series were constructed in FAERS for all 32 TGA pairs. The protocol was pre-registered on the Open Science Framework. <b>Results:</b> Twenty of 23 confirmed CPD detections preceded TGA regulatory action (87%), with a median lead time of 12.0 quarters. For established drugs developing new adverse events, CPD outpaced rolling PRR by a median of 5.0 quarters (Wilcoxon p=0.003). PRR was faster for newly marketed drugs. Sensitivity reached 74% (Tier A) and 95% (Tier B); specificity was 89%. Pooling DAEN and FAERS brought detection forward by a median of 8.0 quarters over DAEN alone (p<0.0001), concentrated among weak signals (15.0 vs 4.0 quarters for below- vs above-median EBGM). The detection-to-action gap widened over the study period (Spearman rho=0.78, p<0.0001). <b>Conclusion:</b> CPD complements disproportionality analysis, with the greatest advantage for late-emerging adverse events in established drugs. Cross-national data pooling substantially accelerates detection for weak signals. These results support integrating temporal change-point monitoring and cross-database data sharing into pharmacovigilance operations.