A tree-based scan statistic for database studies with time-to-event outcomes

2026-09-29

AI Summary

This study addresses the limitation of existing tree-based scan statistics in handling time-to-event data, which compromises sensitivity in drug safety assessments. To overcome this, survival analysis is introduced into hierarchical adverse event detection for the first time. Building upon proportional hazards and exponential survival models, and incorporating robust asymptotic approximations, permutation tests, and parametric bootstrapping, three novel tree-based scan statistics that integrate event occurrence times are proposed. Simulation experiments and a real-world study involving antidiabetic drugs demonstrate that these methods significantly enhance the statistical power of signal detection. Compared with traditional count-based approaches, the proposed framework captures potential safety signals with greater precision, offering a more effective tool for pharmacovigilance.

Abstract

Tree-based scan statistics (TBSSs) are machine learning methods for disproportionality analyses. They simultaneously scan for thousands of hierarchically related health outcomes to detect potential signals of harm from drugs and vaccines, controlling for multiplicity. TBSSs have been extensively used to mine insurance claims databases to detect potential drug adverse events. Current TBSS implementations do not allow comparative safety evaluations with time-to-event outcomes. Explicitly accounting for event timing in analyses can improve power to detect signals compared to methods that use only event counts. We propose three TBSS methods that explicitly leverage event timing to detect potentially harmful effects of drugs. The first assumes proportional hazard rates for each node of the outcome hierarchy and uses a permutation scheme for inference. The second builds on exponential survival models for the terminal nodes of the hierarchy, assuming constant hazard rates at each node, and uses a parametric bootstrap for inference. The third uses robust asymptotic approximations of the hazard rates in connection with an approximate parametric bootstrap. We compare the proposed methods with standard event count based TBSSs in simulation scenarios. Finally, we present results from a database study comparing two glucose-lowering medications among adults with type 2 diabetes.

Problem

• tree-based scan statistic
• time-to-event outcomes
• comparative safety evaluation
• adverse event detection
• disproportionality analysis

Innovation

• Tree-based scan statistics
• Time-to-event outcomes
• Disproportionality analysis
• Parametric bootstrap
• Comparative safety evaluation