Predictive Analytics and Real-Time Data Integration for Resilient Prescribing: An AI-Powered Approach to Mitigate Drug Shortages and Control Costs

• Tafhimul Islam , Sarjil Rahman

  College of Information Engineering, Yangzhou University, China, Department of Electrical and Electronics Engineering, University of Asia Pacific, Dhaka, Bangladesh

  Author

Outpatient prescribing is increasingly disrupted by medication shortages that raise costs, delay therapy, and erode

adherence. This paper proposes a supply chain–aware approach that blends shortage prediction, evidence-based

substitution, and Real-Time Benefit Tools (RTBT) embedded in e-prescribing workflows. Using a mixed-methods

design, we modeled quarterly shortage alerts and compared a predictive model against heuristic baselines; we also

evaluated RTBT’s impact on patient out-of-pocket expenses and adherence under shortage-aware substitution. Results

indicate the predictive model improved precision/recall over heuristics, RTBT reduced median out-of-pocket costs,

and substitution during shortages increased adherence compared with no substitution. We argue that coupling

predictive signals with cost transparency and safe alternatives enables resilient, equitable outpatient medication

access.

• PDF

Copyright (c) 2025 Tafhimul Islam , Sarjil Rahman (Author)

This work is licensed under a Creative Commons Attribution 4.0 International License.