In an effort to address the challenges of prescription errors and patient compliance in medication, Oxford University has developed DrugGPT, an AI tool designed to improve the prescribing process.
This innovation will serve as a support system for healthcare professionals, offering them instant recommendations, potential adverse effects, and drug interactions upon entering patient data.
Prof David Clifton, leading the initiative, emphasizes the tool's ability to justify its recommendations with evidence-based guidance.
Unlike general-purpose AI chatbots, which pose risks of disseminating inaccurate information, DrugGPT has demonstrated proficiency on par with human experts in medical licensing exams, as per preliminary research.
It aims to alleviate the burden on doctors by providing them with up-to-date medical guidance, acting as a "co-pilot" rather than a replacement for human judgment.
The necessity for such technological advancements is underscored by studies highlighting the annual occurrence of approximately 237 million medication errors in England, with significant implications for both cost and patient safety.
DrugGPT not only aims to reduce these errors but also to enhance patient understanding and adherence to medication regimens, thereby optimizing treatment outcomes.
Supported by the National Institute for Health and Care Research, the project has garnered interest from healthcare professionals, including Dr. Lucy Mackillop, who acknowledges its potential to alleviate the pressures faced by doctors.
The Royal College of GPs also recognizes the importance of introducing advanced safety measures, stressing the need for careful implementation and adequate resources in general practice to ensure the effective use of such tools in delivering safe patient care.