Prescription Analysis in the Digital Era: Comparing Artificial Intelligence-Based Versus Manual Approaches

Abstract

Background Prescription auditing using the World Health Organization (WHO) core prescribing indicators is a standard method to evaluate rational drug use. With advances in artificial intelligence (AI), AI-based tools may offer faster approaches for prescription analysis; however, comparative evidence with manual analysis remains limited. The objective of this study was to compare AI-based prescription analysis with manual prescription analysis using WHO core prescribing indicators. Methodology An observational study was conducted at a tertiary care teaching hospital over a period of three months. A total of 308 outpatient prescriptions were collected from the Medicine, Obstetrics and Gynecology, and Orthopedics outpatient departments. Prescriptions were analyzed manually according to WHO core prescribing indicators. AI-based analysis was performed using Google Gemini Pro (Google LLC, Mountain View, California). Results from both approaches were compared using a paired t-test and the chi-square test. Results AI-based prescription analysis produced results largely comparable to manual analysis for most WHO prescribing indicators. No significant differences were observed between AI and manual analysis for most prescribing indicators. However, a significant difference was noted in the percentage of medicines prescribed from the essential medicines list in the Medicine and Obstetrics and Gynecology departments. Conclusion AI-based prescription analysis demonstrated results broadly comparable to conventional manual analysis for most WHO core prescribing indicators. These findings suggest that AI-assisted tools may serve as a feasible supportive approach for prescription auditing.