 the Appropriate Use of AI in Pharmacy Education


By Blake Travis, PharmD Candidate | Jessica Kennedy, PharmD Candidate Nicole Gattas, PharmD, BCPS, FAPhA, ChatGPT*


Travis S


tudents are oſt en early adopters of new forms of technology such as artifi cial intelligence (AI). T ey are able to easily adapt to changing situations, especially when looking at individuals in healthcare programs, given that healthcare on its own is a rapidly evolving fi eld. Students may also see the benefi t of AI as it can


be an eff ective tool to help streamline the learning process and help them use their time more effi ciently. T is is where generative AI (gAI) programs may come into play, but this may not always be for the better. Generative AI is a form of artifi cial intelligence where text, images,


audio, and videos can be input into a program and new modalities are created to form a new product. Current uses for gAI are extensive and include creating recipes, making schedules, grammatical and style corrections, coding, and more. In pharmacy education, AI can be helpful to create study aids, double-check math problems, correct grammar, and change tone of a written communication, And, with the good, comes the potential for more malicious uses...such as completing an assignment or inappropriately assessing patients. Students may unintentionally submit protected health information (PHI) or research fi ndings to ask questions, not aware if that data is being stored and further utilized.


Testing gAI’s Case Assessment Abilities To better understand the clinical capabilities of gAI in a healthcare


setting, we used ChatGPT to assess a patient case that was presented in a didactic course. During our initial analysis, we asked ChatGPT 3.5 to provide a complete assessment and plan when given specifi c subjective and objective information. We then took this assessment and plan and compared it to the answers provided by the professor. We later performed a second analysis using the updated ChatGPT 4.0 and compared its assessment to that of ChatGPT 3.5. Table 1 highlights key diff erences in ChatGPT’s capabilities in assessing patient disease states.


Table 1. Kennedy Gattas


Abbreviations: STI – sexually transmitted infection; AUD – alcohol use disorder; OUD – opioid use disorder; DVT – deep vein thrombosis; AMS – altered mental status; SUD – substance use disorder; CPGs – clinical practice guidelines.


ChatGPT 3.5 did a mediocre job of providing an assessment and plan. It failed to pick up on the patient’s need for DVT prophylaxis in an inpatient setting and need for STI treatment. While the program did provide some general pharmacologic recommendations for the patient’s disease states (e.g. using a benzodiazepine for AUD), it did not indicate a specifi c agent, route, dose, or frequency, nor cite any clinical practice guidelines. ChatGPT 4.0 provided a better assessment and plan, noticing that


the patient needed STI treatment unlike ChatGPT 3.5, but still did not notice the patient’s need for DVT prophylaxis. T e program also recommended specifi c pharmacologic agents for the patient’s various disease states (e.g. diazepam for AUD, ceſt riaxone and doxycycline for STI, etc.) with specifi c doses, routes, and frequencies. It also referenced clinical practice guidelines such those provided by Centers for Disease Control and American Psychiatric Association. Overall, ChatGPT 4.0 came closer to mimicking a student’s


assessment and plan of a patient, whereas ChatGPT 3.5 was severely lacking in its capabilities. ChatGPT’s capabilities can only be expected to improve as each update comes out and will be able to provide more accurate and detailed assessments as time goes on. T is analysis was completed using the free tier of ChatGPT and these fi ndings cannot necessarily be extrapolated to other gAI programs.


30 Missouri PHARMACIST | Volume 98, Issue III | Fall 2024


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48