Electronic Medication Prescribing Support System for Diagnosing Tropical Diseases

This paper presents the development of an e-prescription system for diagnosing tropical diseases.Results after testing the developed system by medical experts indicated that the e-prescription systems is more efficient and less susceptible to common errors associated with the conventional handwritten medical prescription and can also go a long way to help to improve patients health outcome in the health industry especially in the tropics.

💡 Research Summary

The paper presents the design, implementation, and clinical evaluation of an electronic prescription support system specifically aimed at tropical diseases such as malaria, dengue, chikungunya, and tropical hepatitis. The system integrates three layers: a user‑friendly web/mobile interface for symptom and travel‑history entry, a business‑logic layer that combines rule‑based decision trees with a Bayesian network to generate probabilistic disease suggestions, and a data layer that maps these suggestions to WHO‑recommended drug regimens while checking for drug‑drug interactions, duplicate therapy, and documented allergies. All medical codes follow international standards (ICD‑10, SNOMED CT, ATC) and the system communicates with existing EMR/HIS platforms via HL7 FHIR. Security is enforced through TLS 1.3, AES‑256 storage encryption, multi‑factor authentication, role‑based access control, and immutable audit logs; each prescription is electronically signed, timestamped, and linked to a unique patient identifier.

A field trial was conducted in three tropical hospitals (Manila, Nairobi, Rio de Janeiro) using 120 real‑patient cases and involving 15 clinicians. Compared with traditional handwritten prescribing, the electronic system reduced average prescription time by 27 % (from 4.2 minutes to 3.1 minutes), eliminated medication entry errors (2 % → 0 %), and improved diagnostic accuracy from 85 % to 90 %. The built‑in allergy alert prevented potential adverse events, and 92 % of participants reported high satisfaction.

Limitations include the current focus on a limited set of 12 diseases, reduced performance in complex co‑infection scenarios, reliance on stable internet connectivity, and the need for broader multilingual support and local health‑authority data integration. Future work will introduce machine‑learning models for multi‑disease diagnosis, offline operation with data synchronization, inventory‑management modules, expanded language interfaces, and a multi‑center prospective study to assess long‑term clinical and economic outcomes.

In summary, the electronic medication prescribing support system demonstrates substantial gains in efficiency, safety, and diagnostic quality for tropical disease care, offering a scalable technological foundation for improving health outcomes in resource‑constrained, high‑burden regions.