The DOI matches its Crossref record; 1 of the 1 references it cites were found in a registry or an index.
Assessing ChatGPT's capability in understanding and reporting antiretroviral therapy drug–drug interactions (ACCURATE-DDI)
Abstract
OBJECTIVES: Artificial intelligence tools are transforming access to medication information. However, their ability to accurately identify antiretroviral (ARV) drug-drug interactions (DDIs) remains unclear. This study evaluated ChatGPT's analysis of ARV-related DDIs compared to established HIV-specific DDI resources. DESIGN: Cross-sectional observational study. METHODS: Using ChatGPT4o-mini in November 2024, we tested 94 ARV DDI pairs. ChatGPT was asked to classify each interaction as "no interaction," "potential interaction," or "serious interaction." Accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated, using the HIV/HCV Drug Therapy Guide and the Liverpool HIV Drug Interactions Checker as the correct classification. A subset of 25 pairs was qualitatively analyzed for responses to the prompt, "Can I take Drug A with Drug B?" using a composite score incorporating severity, mechanism, clinical effects, and management (0 = incorrect, 1 = mixed, 2 = correct). Responses were independently assessed by two HIV-specialized pharmacist/pharmacologist reviewers. RESULTS: ChatGPT correctly classified just 40.4% (38/94) of DDI pairs; errors were primarily false negatives (34/56 errors; 60.7%). Sensitivity was 46.0%, specificity 29.0%, PPV 56.9%, and NPV 20.9%. The mean composite score was 3.9/8. ChatGPT showed limited accuracy in identifying severity of serious DDIs (0.4/2), despite often correctly identifying mechanism (1.4/2) and clinical effects (1.6/2). Serious DDI management was often incorrect (0.9/2). ChatGPT performed poorly with ritonavir DDIs involving mechanisms other than CYP3A4 inhibition. CONCLUSION: ChatGPT demonstrated limited accuracy and specificity, often generating responses that combined correct and incorrect information. Currently, ChatGPT should not be considered a reliable tool for ARV DDI information.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Despite LLMs' potential, notable challenges include hallucination risk, response sensitivity to prompting method, cost, and data privacy.
“Despite LLMs' potential to support pharmacist-led direct patient care tasks, notable challenges include hallucination risk, response sensitivity to prompting method, cost, and data privacy.16,53,92”
From Large Language Models in Pharmacist-Led Direct Patient Care: A Rapid Review of Published Evaluations (2026) · cites this paper as [92]
Citations
Cites
The publisher hasn’t deposited this paper’s reference list.
Research works printed but not linked: 23
- Juhi, 2023, The capability of ChatGPT in predicting and explaining common drug–drug interactions, Cureus, 15, e36272
- Roosan, 2024, Effectiveness of ChatGPT in clinical pharmacy and the role of artificial intelligence in medication therapy management, J Am Pharm Assoc, 64, 422, 10.1016/j.japh.2023.11.023 · DOI
- Hsu, 2023, Examining real-world medication consultations and drug–herb interactions: ChatGPT performance evaluation, JMIR Med Educ, 9, e48433, 10.2196/48433 · DOI · Paper page
- Shariff, 2021, Assessing consistency of drug–drug interaction-related information across various drug information resources, Cureus, 13, e13766
- Spina, 2012, Clinically significant drug interactions with newer antidepressants, CNS Drugs, 26, 39, 10.2165/11594710-000000000-00000 · DOI
- Foley, 2006, Bupropion: pharmacology and therapeutic applications, Expert Rev Neurother, 6, 1249, 10.1586/14737175.6.9.1249 · DOI
- Tseng, 2017, Cobicistat versus ritonavir: similar pharmacokinetic enhancers but some important differences, Ann Pharmacother, 51, 1008, 10.1177/1060028017717018 · DOI
- Chahine, 2021, Fostemsavir: the first oral attachment inhibitor for treatment of HIV-1 infection, Am J Health Syst Pharm, 78, 376, 10.1093/ajhp/zxaa416 · DOI
- Moore, 2019, Methadone and buprenorphine pharmacokinetics and pharmacodynamics when coadministered with fostemsavir to opioid-dependent, human immunodeficiency virus seronegative participants, Br J Clin Pharmacol, 85, 1771, 10.1111/bcp.13964 · DOI
- Stringer, 2009, Methadone-associated Q-T interval prolongation and torsades de pointes, Am J Health Syst Pharm, 66, 825, 10.2146/ajhp070392 · DOI
- Boyle, 2019, Clinical pharmacodynamics, pharmacokinetics, and drug interaction profile of doravirine, Clin Pharmacokinet, 58, 1553, 10.1007/s40262-019-00806-9 · DOI
- Burger, 2020, Cobicistat: a case of mislabelled drug-drug interaction risk?, Br J Clin Pharmacol, 86, 834, 10.1111/bcp.14262 · DOI
- Marzolini, 2025, Drug-drug interactions potential with the HIV-1 capsid inhibitor lenacapavir, Expert Opin Drug Metab Toxicol, 21, 161, 10.1080/17425255.2024.2415295 · DOI
- van Nuland, 2024, Assessing accuracy of ChatGPT in response to questions from day to day pharmaceutical care in hospitals, Explor Res Clin Soc Pharm, 15, 100464
- Burger, 2008, The effect of atazanavir and atazanavir/ritonavir on UDP-glucuronosyltransferase using lamotrigine as a phenotypic probe, Clin Pharmacol Ther, 84, 698, 10.1038/clpt.2008.106 · DOI
- van der Lee, 2006, Lopinavir/ritonavir reduces lamotrigine plasma concentrations in healthy subjects, Clin Pharmacol Ther, 80, 159, 10.1016/j.clpt.2006.04.014 · DOI
- Grossman, 2024, Appropriateness of ChatGPT as a resource for medication-related questions, Br J Clin Pharmacol, 90, 2691, 10.1111/bcp.16212 · DOI
- Hughes, 2015, Managing drug interactions in HIV-infected adults with comorbid illness, CMAJ, 187, 36, 10.1503/cmaj.131626 · DOI
- Carpenter, 2019, Clinically relevant drug–drug interactions in primary care, Am Fam Physician, 99, 558
- Deeks, 2014, Cobicistat: a review of its use as a pharmacokinetic enhancer of atazanavir and darunavir in patients with HIV-1 infection, Drugs, 74, 195, 10.1007/s40265-013-0160-x · DOI
- Sherman, 2015, Cobicistat: review of a pharmacokinetic enhancer for HIV infection, Clin Ther, 37, 1876, 10.1016/j.clinthera.2015.07.022 · DOI
- Renjifo, 2015, Pharmacokinetic enhancement in HIV antiretroviral therapy: a comparison of ritonavir and cobicistat, AIDS Rev, 17, 37
- Bischof, 2025, Chat GPT vs. clinical decision support systems in the analysis of drug–drug interactions, Clin Pharmacol Ther, 117, 1142, 10.1002/cpt.3585 · DOI
Cited by
Sorted by citations
Large Language Models in Pharmacist-Led Direct Patient Care: A Rapid Review of Published Evaluations
Yakob, R. A., Bamgboje-Ayodele, A., Collins, J. C., & Aslani, P. (2026) · Research in Social and Administrative Pharmacy · doi:10.1016/j.sapharm.2026.09.007 · Cited by 1
Sunagawa, S. W., Sherman, E. M., Badowski, M., Cluck, D. B., Durham, S. H., Havens, J. P., Huesgen, E., Ma, J. K., Rohani, R., Sahloff, E. G., et al. (2026) · British Journal of Clinical Pharmacology · doi:10.1002/bcp.70884
Metadata
- Authors
- Ready, E. A., Patterson, C., Gibbons, S., Yoong, D., Marzolini, C., Walker, L. E., & Tseng, A.
- Affiliations
- University of British Columbia (CA)
- University of Liverpool (GB)
- St. Michael's Hospital (CA)
- University of Toronto (CA)
- Corresponding author
- Ready, Erin A.
- Venue
- AIDS
- Published
- 2026-04-11
- Volume
- 40
- Issue
- 9
- Pages
- 1291-1296
- ISSN
- 0269-9370 (ISSN-L)
- 1473-5571
- OpenAlex
- W7154141451
- Cited by
- 2
- References
- 0
- Citations by year
- 2026: 2
- FWCI
- 3.91
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- AIDS (publishedVersion)
- PubMed (publishedVersion)
- IRIS (submittedVersion)
- Topics
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Medicine › Infectious Diseases › HIV/AIDS drug development and treatment
- Health Sciences › Health Professions › Pharmacy › Pharmacy and Medical Practices
- Keywords
- ChatGPT, antiretroviral drugs, drug-drug interactions, artificial intelligence, HIV, ritonavir, CYP3A4 inhibition, pharmacist review
- MeSH terms
- Generative Artificial Intelligence, Cross-Sectional Studies, Drug Interactions, Humans, Sensitivity and Specificity, HIV Infections, Anti-Retroviral Agents
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
