Detecting adverse high-order drug combinations from individual case safety reports using computational statistics on disproportionality measures
Abstract
Adverse drug reactions linked to the intake of drug combinations are a critical concern in pharmacovigilance, particularly as the controlled environment of clinical trials often lacks the scale and diversity to detect rare events involving multiple medications. While spontaneous reporting systems provide the necessary breadth for post-market surveillance, identifying overrepresented drug cocktails within such large-scale data remains a significant computational challenge. This study introduces a computational framework for the detection of drug cocktails associated with adverse events, leveraging disproportionality analysis on individual case safety reports. By integrating the Anatomical Therapeutic Chemical classification, the framework extends beyond individual drugs to capture hierarchical pharmacological relationships, enabling exploration of the space of drug combinations beyond pairwise analysis. To address biases inherent in existing disproportionality measures, we employ a hypergeometric risk metric, while a Markov Chain Monte Carlo algorithm provides robust empirical p-value estimation for the risk associated with cocktails. A genetic algorithm further facilitates efficient identification of high-risk drug cocktails. A post-treatment step based on penalized logistic regression allows distinguishing true pharmacological interactions from combined individual effects for cocktails of any size. Validation on synthetic and FDA Adverse Event Reporting System data demonstrates the method’s efficacy in detecting established drugs and drug combinations associated with myopathy-related adverse events. Implemented as an R package, this framework offers a reproducible, scalable tool for post-market drug safety surveillance.
Related
- Analysis of Polynomial Threshold Functions on Random Regular Graphs: Computational Complexity of Detecting Noisy Random Lifts
- Doubly robust win statistics for hierarchical survival endpoints in individual- and cluster-randomized trials
- On wall-crossing invariant combinations of Welschinger numbers for rational surfaces with a conic bundle
- Boolean Combinations of Graphs
- Parallel Searching Method for Verifying Finiteness of Prime Numbers Presented as Specific Linear Combinations of Factorials
BibTeX
@article{bangard2026detecting,
title = {{Detecting adverse high-order drug combinations from individual case safety reports using computational statistics on disproportionality measures}},
author = {Jules Bangard and Einar Holsbø and Kristian Svendsen and Vittorio Perduca and Etienne Birmelé},
journal = {Computo},
year = {2026},
month = jul,
doi = {10.57750/7e54-2z98},
eprint = {2504.00646},
archivePrefix = {arXiv},
url = {https://hal.science/hal-05012416v2},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Detecting adverse high-order drug combinations from individual case safety reports using computational statistics on disproportionality measures
AU - Jules Bangard
AU - Einar Holsbø
AU - Kristian Svendsen
AU - Vittorio Perduca
AU - Etienne Birmelé
PY - 2026
DA - 2026/07/15
JO - Computo
DO - 10.57750/7e54-2z98
UR - https://hal.science/hal-05012416v2
AB - Adverse drug reactions linked to the intake of drug combinations are a critical concern in pharmacovigilance, particularly as the controlled environment of clinical trials often lacks the scale and diversity to detect rare events involving multiple medications. While spontaneous reporting systems provide the necessary breadth for post-market surveillance, identifying overrepresented drug cocktails within such large-scale data remains a significant computational challenge. This study introduces a computational framework for the detection of drug cocktails associated with adverse events, leveraging disproportionality analysis on individual case safety reports. By integrating the Anatomical Therapeutic Chemical classification, the framework extends beyond individual drugs to capture hierarchical pharmacological relationships, enabling exploration of the space of drug combinations beyond pairwise analysis. To address biases inherent in existing disproportionality measures, we employ a hypergeometric risk metric, while a Markov Chain Monte Carlo algorithm provides robust empirical p-value estimation for the risk associated with cocktails. A genetic algorithm further facilitates efficient identification of high-risk drug cocktails. A post-treatment step based on penalized logistic regression allows distinguishing true pharmacological interactions from combined individual effects for cocktails of any size. Validation on synthetic and FDA Adverse Event Reporting System data demonstrates the method’s efficacy in detecting established drugs and drug combinations associated with myopathy-related adverse events. Implemented as an R package, this framework offers a reproducible, scalable tool for post-market drug safety surveillance.
SN - 2824-7795
ER -