Objective: To bring temporal information into association rule mining on complex pharmaceutical transaction data, so that co-prescribing rules and their evolution over time can both be identified. Methods: Transaction records are reorganized into time-stamped drug-set sequences at a selected Anatomical Therapeutic Chemical (ATC) abstraction level. Temporal association rules with interval constraints are extracted using a residual-leverage measure that shrinks toward the null hypothesis. Each rule is represented as a trajectory of support, confidence and leverage across sliding windows and clustered into five evolution modes: stable, emerging, decaying, periodic and abrupt. Results: On a provincial procurement and insurance settlement dataset, the method compresses the rule space while preserving coverage. It also exposes emerging polypharmacy combinations earlier than cross-sectional screening. Conclusion: It provides an interpretable instrument for rational drug use surveillance, chronic disease monitoring, drug interaction signal detection and insurance fund risk warning.
References
[1] Ageno, A., Català, N., Pons, M. (2023) Acquisition of temporal patterns from electronic health records: an application to multimorbid patients. BMC Medical Informatics and Decision Making, 23(1), 189.
[2] Feng, Y., Wang, K., Fan, J., Wu, X., Li, T., Yang, Z. (2024) Mindfulness intervention, homogeneous medical concept, and concentrated solution nursing for colorectal cancer patients: a retrospective study. BMC Cancer, 24(1), 1055.
[3] Estri, H., Strasser, Z. H., Klann, J. G., McCoy, T. H., Jr., Wagholikar, K. B., Vasey, S., Castro, V. M., Murphy, M. K. E., Murphy, S. N. (2020) Transitive sequencing medical records for mining predictive and interpretable temporal representations. Patterns, 1(4), 1-11.
[4] Assadi Shalmani, M., Khani, M., Taleban, A., Yi, Z., Fink, J. T., Weber, C. E., Luo, J. (2026) Sequential pattern transformer (SPT): a generative and interpretable framework for predicting disease trajectories. Neural Computing and Applications, 38(3), 28.
[5] Wu, P., Nelson, S. D., Zhao, J., Stone Jr, C. A., Feng, Q., Chen, Q., Wei, W. Q. (2021) DDIWAS: High-throughput electronic health record-based screening of drug-drug interactions. Journal of the American Medical Informatics Association, 28(7), 1421-1430.
[6] Mansour, A. M., Khazalah, F., Obeidat, M. A. (2022) Suspected adverse drug reaction detection using association rules mining and fuzzy sets. Przegląd Elektrotechniczny, 98.
[7] Noguchi, Y., Ueno, A., Otsubo, M., Katsuno, H., Sugita, I., Kanematsu, Y., Yoshida, A., Esaki, H., Tachi, T., Teramachi, H. (2018) A new search method using association rule mining for drug-drug interaction based on spontaneous report system. Frontiers in Pharmacology, 9, 197.
[8] Cai, R., Liu, M., Hu, Y., Melton, B. L., Matheny, M. E., Xu, H., Waitman, L. R. (2017) Identification of adverse drug-drug interactions through causal association rule discovery from spontaneous adverse event reports. Artificial Intelligence in Medicine, 76, 7-15.
[9] He, Y., Sun, J., Tan, X. (2025) Performance of apriori algorithm for detecting drug-drug interactions from spontaneous reporting systems. Mathematics, 13(11), 1710.
[10] Hussein, A. H., Tarek, L., Maged, S., Safy, M. (2026) A comparative study of association rule mining algorithms for enhanced drug-drug interaction and adverse drug re-action discovery. Journal of Al-Azhar University Engineering Sector, 21(79), 1179-1189.
[11] Flothow, A., Novelli, A., Sundmacher, L. (2023) Analytical methods for identifying sequences of utilization in health data: a scoping review. BMC Medical Research Methodology, 23(1), 212.
Share and Cite
Chen, H. (2026) Temporal Information Enhanced Association Rule Discovery and Evolution Pattern Identification in Complex Pharmaceutical Transaction Data. Journal of Disease and Public Health, 2(3), 46-51. https://doi.org/10.71052/jdph/FHIN5235