AI and Drug Information: Where OpenEvidence Stands Out—and Where Pharmacists Still Win

The integration of artificial intelligence (AI) into health care augments the work of health care professionals (HCPs) in their day-to-day practice.¹ Although not specifically designed for health care use, tools such as ChatGPT, Google Gemini, and Microsoft Copilot are being used in various ways to assist clinicians in their practice.² The use of these platforms highlights the growing supportive role of generative AI in delivering evidence-based practice. One generative AI tool specific to health care and, as such, a platform widely used among clinicians, is OpenEvidence.³

OpenEvidence is an official partner of several premier professional associations, including the JAMA Network, The New England Journal of Medicine, the National Comprehensive Cancer Network, and Cochrane Systematic Reviews.³ In December 2025, the American Diabetes Association (ADA) also partnered with OpenEvidence to advance AI integration into clinical practice.⁴ This collaboration uses clinicians' feedback to identify areas of clinical uncertainty in current ADA guidelines and guides the development of improved recommendations.⁴ According to OpenEvidence, over 100 million Americans will be treated annually by clinicians who use this tool. In addition, to date, over 200 million clinical consultations have been powered by the tool.⁵

AI in Pharmacy Practice

Like other HCPs, pharmacists use AI-driven tools to support the delivery of pharmacy-related services.¹ The use of these tools is incorporated into a variety of activities in a variety of practice settings. Examples include tools that support adverse drug reaction detection, clinical decision support systems, computerized prescriber order entry, dosing recommendations, detection of drug interactions, medication adherence, and error detection.¹

In addition to the system-wide use of AI, pharmacists can use such tools for obtaining drug information and for clinical decision-making.⁶ The accuracy of the information provided by generative AI tools such as ChatGPT and Google Gemini has been assessed in several studies that reported varying degrees of inaccuracy.^7-10 However, the accuracy of drug information generated by OpenEvidence, a tool geared toward HCPs, is more limited. A PubMed search using the terms “OpenEvidence” and “drug information” identified only 2 studies that evaluated the accuracy of information generated by this tool.^11,12

Studies Involving OpenEvidence

In a study by Ipema et al, 30 drug information (DI) questions previously answered by DI pharmacists were queried in 5 AI chatbots: OpenEvidence, Clair, GlassHealth, DougallGPT, and ChatGPT.¹¹ Notably, Clair, GlassHealth, and DougallGPT are also generative AI tools specific to health care. The quality of the responses was assessed independently by 3 pharmacists using the 25-point validated CLEAR scoring framework (completeness of content, lack of false information, evidence supporting the content, appropriateness, and relevance). The investigators reported that OpenEvidence had the highest mean total CLEAR score (17.82/25), followed by ChatGPT (15.72/25).¹¹

In another study by Surbaugh et al, 4 generative AI tools (ChatGPT, Microsoft Copilot Chat, Perplexity, and OpenEvidence) were assessed for accuracy with respect to their ability to provide dosing recommendations for both initiation and missed-dosing scenarios for long-acting injectable antipsychotics.¹² The investigators reported that accuracy rates among the tools varied significantly. OpenEvidence was the most accurate, yielding a 95% accuracy rate (providing an inaccurate response to only 1 query). ChatGPT was the least accurate, with a 40% accuracy rate. The investigators also noted that most of the inaccuracies occurred when AI tools provided recommendations for missed-dose scenarios. The investigators concluded that tools rooted in more scientific data, such as OpenEvidence, may be more useful for providing accurate information quickly than conversation-based tools, such as ChatGPT.¹²

Conclusion

AI has and will continue to revolutionize the practice of health care. Such tools have proven to increase efficiency in the delivery of certain elements of medical services.¹³ Likewise, the delivery of health services related to medication use has also benefited from the integration of AI technology.^14-16 Specifically, generative AI technologies have been used by pharmacists to obtain and provide drug information and while multiple tools are available for this purpose, pharmacists should be cognizant of the fact that certain tools may be better than others. Our search of the literature found that when compared to other generative AI tools, OpenEvidence provided outputs that were deemed to be of better quality, by multiple measures.^11,12 However, as reported by Ipema et al, when compared with pharmacists' responses, all AI chatbot responses, including those generated by OpenEvidence, had statistically lower scores.¹¹ This highlights the fact that human insights remain essential for ensuring that answers to drug information questions are of high quality.

The authors would like to acknowledge Tracy Cheng, BPS, PharmD candidate, for her assistance in developing this manuscript.

REFERENCES

1. Chalasani SH, Syed J, Ramesh M, Patil V, Kumar TMP. Artificial intelligence in the field of pharmacy practice: a literature review. Explor Re Clin Soc Pharm. 2023;12:100346. doi:10.1016/j.rcsop.2023.100346

2. Bhuyan SS, Sateesh V, Mukul N, et al. Generative artificial intelligence use in healthcare: opportunities for clinical excellence and administrative efficiency. J Med Syst. 2025;49(1):10. doi:10.1007/s10916-024-02136-1

3. OpenEvidence. Accessed July 13, 2026. https://www.openevidence.com/

4. The American Diabetes Association and OpenEvidence collaborte to accelerate evidence-based guideline delivery. News release. American Diabetes Association. December 3, 2025. Accessed July 13, 2026. https://diabetes.org/newsroom/press-releases/american-diabetes-association-and-openevidence-collaborate-accelerate

5. OpenEvidence is the leading medical information platform. OpenEvidence. Accessed July 13, 2026. https://www.openevidence.com/about

6. Alghitran A, Al Osaimi HM, Albuluwi A, et al. Integrating ChatGPT as a tool in pharmacy practice: a cross-sectional exploration among pharmacists in Saudi Arabia. Integr Pharm Res Pract. 2025;14:31-43. doi:10.2147/IPRP.S500689

7. Alzahrarni L, Alkhalaf A, Almehizia R, et al. Comparison of drug information questions’ responses between drug information pharmacists and artificial intelligence technologies: a real-world analysis in a tertiary care hospital. J Am Coll Clin Pharm. 2026;9(4):e70208. doi:10.1002/jac5.70208

8. Ready E, Patterson C, Gibbons S, et al. Assessing ChatGPT’s capability in understanding and reporting antiretroviral therapy drug-drug interaction effects: quantitative and qualitative results from the ACCURATE-DDI study. Brit J Clin Pharmacol. 2025;91(S1):16-17. doi:10.1002/bcp.70223

9. Frazer A, Yoon E, Durant KM. Artificial intelligence for drug information: evaluating its use in real-world pharmacy practice. Am J Health Syst Pharm. 2026;83(12):669-676. doi:10.1093/ajhp/zxaf350

10. Li L, Du P, Huang X, et al. Comparative analysis of generative artificial intelligence systems in solving clinical pharmacy problems: mixed methods study. JMIR Med Inform. 2025;13:e76128. doi:10.2196/76128

11. Ipema H, Munir F, Sarna K, Wasynczuk J, Saiyad Z. Evaluation of clinically-focused artificial intelligence chatbots for answering drug information questions. J Am Coll Clin Pharm. 2026;9(6):e70231. doi:10.1002/jac5.70231

12. Surbaugh LA, Moeller KE. Assessing the accurate of 4 generative AI tools to guide dosing of long-acting injectable antipsychotics. Am J Health-Syst Pharm. 2026;zxag089. doi:10.1093/ajhp/zxag089

13. Jeong J, Kim S, Pan L, et al. Reducing the workload of medical diagnosis through artificial intelligence: a narrative review. Medicine (Baltimore). 2025;104(6):e41470. doi:10.1097/MD.000000000041470

14. Damiani G, Altamura G, Zedda M, et al. Potentiality of algorithms and artificial intelligence adoption to improve medication management in primary care: a systematic review. BMJ Open. 2023;13(3):e065301. doi:10.1136/bmjopen-2022-065301

15. Shakarbaev N. AI-enhanced clinical decision support reduces medication errors and adverse drug events in a multicenter teaching hospital network: a prospective randomized controlled trial. Int J Med Inform. 2026;215:106457. doi:10.1016/j.ijmedinf.2026.106457

16. Levivien C, Cavagna P, Grah A, et al. Assessment of a hybrid decision support system using machine learning with artificial intelligence to safely rule out prescriptions from medication review in daily practice. Int J Clin Pharm. 2022;44(2):459-465. doi:10.1007/s11096-021-01366-4