{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T14:00:42Z","timestamp":1781100042474,"version":"3.54.1"},"reference-count":36,"publisher":"SAGE Publications","license":[{"start":{"date-parts":[[2026,2,1]],"date-time":"2026-02-01T00:00:00Z","timestamp":1769904000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"},{"start":{"date-parts":[[2026,2,1]],"date-time":"2026-02-01T00:00:00Z","timestamp":1769904000000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/journals.sagepub.com\/page\/policies\/text-and-data-mining-license"}],"funder":[{"DOI":"10.13039\/501100023674","name":"Deanship of Research and Graduate Studies at King Khalid University","doi-asserted-by":"publisher","award":["RGP2\/120\/47"],"award-info":[{"award-number":["RGP2\/120\/47"]}],"id":[{"id":"10.13039\/501100023674","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["journals.sagepub.com"],"crossmark-restriction":true},"short-container-title":["DIGITAL HEALTH"],"published-print":{"date-parts":[[2026,2]]},"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Large language models (LLMs) are increasingly used in medical education, but their performance and reliability on mechanistically demanding topics like acid\u2013base interpretation are unclear.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We conducted a cross-sectional head-to-head evaluation of three LLMs (ChatGPT, DeepSeek, Perplexity) using 75 textbook acid\u2013base cases related to management and drug treatment. One obsolete case was excluded, leaving 74 vignettes (29 metabolic, 10 respiratory, 35 mixed) that generated 510 multiple-choice questions (MCQs). Each MCQ was mapped to case category, one of seven cognitive domains, and one of four distractor-based difficulty bands. All questions were posed via public web interfaces in two independent phases (Phase I and II). Primary outcome was accuracy (proportion correct); consistency outcomes were identical answers across phases and reproducibly correct answers (correct in both phases).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Overall accuracy was 59.8% (305\/510) for ChatGPT, 58.8% (300\/510) for DeepSeek, and 53.7% (274\/510) for Perplexity (Cochran\u2019s Q p = 0.16); ChatGPT and DeepSeek each outperformed Perplexity (p = 0.0086 and p = 0.0374). In metabolic cases (164 items), Perplexity scored 50.6% versus 60.4% and 62.8% for ChatGPT and DeepSeek. Accuracy was highest for compensation\/expected response (116 items; 70.7% for ChatGPT and DeepSeek, 62.1% for Perplexity) and lowest for metabolic acidosis subtypes (59 items; 52.5%, 42.4%, 32.2%). Accuracy declined with difficulty, with Perplexity significantly lower than the other models in moderate\u2013hard and hard bands. Identical answers across phases occurred in 81.7% of items for ChatGPT, 81.1% for DeepSeek, and 76.1% for Perplexity, but reproducibly correct answers were 52.1%, 48.8%, and 43.2%.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>ChatGPT and DeepSeek showed moderate performance and generally higher accuracy than Perplexity. Stable repeated errors despite only moderate reproducibly correct responses support cautious interpretation of LLM outputs and argue against their use as stand-alone tools for acid\u2013base learning or clinical decision making of management and drug treatment .<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1177\/20552076261452489","type":"journal-article","created":{"date-parts":[[2026,6,8]],"date-time":"2026-06-08T15:09:11Z","timestamp":1780931351000},"update-policy":"https:\/\/doi.org\/10.1177\/sage-journals-update-policy","source":"Crossref","is-referenced-by-count":0,"title":["Head-to-head evaluation of ChatGPT, DeepSeek, and Perplexity on acid\u2013base disorder case clinical management and drug treatment: Accuracy, domain performance, and response consistency assessment"],"prefix":"10.1177","volume":"12","author":[{"given":"Moteb","family":"Khobrani","sequence":"first","affiliation":[{"name":"King Khalid University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Asaad Ahmed","family":"Asaad Khalil","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0866-7040","authenticated-orcid":false,"given":"Salman Ashfaq","family":"Ahmad","sequence":"additional","affiliation":[{"name":"Iqra University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Azfar Athar","family":"Ishaqui","sequence":"additional","affiliation":[{"name":"King Khalid University"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"179","published-online":{"date-parts":[[2026,6,8]]},"reference":[{"key":"e_1_3_6_2_2","volume-title":"Large Language Models in Healthcare","author":"Al-Garadi M","unstructured":"Al-Garadi M, Mungle T, Ahmed A, et al. Large Language Models in Healthcare. arXiv preprint arXiv:250304748 2025."},{"key":"e_1_3_6_3_2","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pdig.0000198"},{"key":"e_1_3_6_4_2","first-page":"263","volume-title":"Acid-Base Disorders: Clinical Evaluation and Management","author":"Reddi AS","year":"2019","unstructured":"Reddi AS. Acid\u2013Base Disorders in Critically Ill Patients. Acid-Base Disorders: Clinical Evaluation and Management. Springer, 2019, pp. 263\u2013279."},{"key":"e_1_3_6_5_2","doi-asserted-by":"publisher","DOI":"10.1053\/j.ackd.2016.02.005"},{"key":"e_1_3_6_6_2","first-page":"148","article-title":"Acid-Base Interpretation: A Practical Approach","volume":"111","author":"Morikawa MJ","year":"2025","unstructured":"Morikawa MJ, Ganesh PR. Acid-Base Interpretation: A Practical Approach. American Family Physician 2025; 111: 148\u2013155.","journal-title":"American Family Physician"},{"key":"e_1_3_6_7_2","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-025-02954-4"},{"key":"e_1_3_6_8_2","doi-asserted-by":"publisher","DOI":"10.1016\/B978-0-323-47794-9.00012-3"},{"key":"e_1_3_6_9_2","doi-asserted-by":"publisher","DOI":"10.1186\/s12909-024-06309-x"},{"key":"e_1_3_6_10_2","doi-asserted-by":"publisher","DOI":"10.3389\/frai.2025.1616145"},{"key":"e_1_3_6_11_2","first-page":"47","article-title":"Product: Perplexity","volume":"46","author":"Roy A","year":"2025","unstructured":"Roy A. Product: Perplexity. The Journal of the Canadian Health Libraries Association= Journal de l'Association des Biblioth\u00e8ques de la Sant\u00e9 du Canada 2025; 46: 47\u201352. https:\/\/doi.org\/10.29173\/jchla29877","journal-title":"The Journal of the Canadian Health Libraries Association= Journal de l'Association des Biblioth\u00e8ques de la Sant\u00e9 du Canada"},{"key":"e_1_3_6_12_2","doi-asserted-by":"publisher","DOI":"10.1002\/oto2.70075"},{"key":"e_1_3_6_13_2","first-page":"e1766","article-title":"Is ChatGPT-4 accurate and complete when answering questions on tuberculosis?","volume":"11","author":"De Vito A","year":"2025","unstructured":"De Vito A, Colpani A, Buonsenso D, et al. Is ChatGPT-4 accurate and complete when answering questions on tuberculosis? Results of the ChatGTB study. 2025; 11: e1766.","journal-title":"Results of the ChatGTB study"},{"key":"e_1_3_6_14_2","doi-asserted-by":"publisher","DOI":"10.3390\/diagnostics16030388"},{"key":"e_1_3_6_15_2","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2024.25953"},{"key":"e_1_3_6_16_2","doi-asserted-by":"publisher","DOI":"10.3389\/fmed.2025.1679153"},{"key":"e_1_3_6_17_2","volume-title":"Critical Concept Mastery Series: Acid-Base Disturbance Cases","author":"Healy Z","year":"2021","unstructured":"Healy Z. Critical Concept Mastery Series: Acid-Base Disturbance Cases. McGraw Hill Professional, 2021."},{"key":"e_1_3_6_18_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2024.104620"},{"key":"e_1_3_6_19_2","first-page":"9","volume-title":"Informatics","author":"Beheshti M","year":"2025","unstructured":"Beheshti M, Toubal IE, Alaboud K, et al. Evaluating the reliability of chatgpt for health-related questions: A systematic review. In: Informatics. MDPI, 2025, p. 9."},{"key":"e_1_3_6_20_2","volume-title":"Large language models for forecasting and anomaly detection: A systematic literature review","author":"Su J","unstructured":"Su J, Jiang C, Jin X, et al. Large language models for forecasting and anomaly detection: A systematic literature review. arXiv preprint arXiv:240210350 2024."},{"key":"e_1_3_6_21_2","doi-asserted-by":"publisher","DOI":"10.1007\/s40670-023-01956-z"},{"issue":"12","key":"e_1_3_6_22_2","first-page":"e76309","article-title":"Advancements in AI medical education: assessing ChatGPT\u2019s performance on USMLE-style questions across topics and difficulty levels","volume":"16","author":"Penny P","year":"2024","unstructured":"Penny P, Bane R, Riddle V. Advancements in AI medical education: assessing ChatGPT\u2019s performance on USMLE-style questions across topics and difficulty levels. Cureus 2024; 16(12): e76309. https:\/\/doi.org\/10.7759\/cureus.76309","journal-title":"Cureus"},{"key":"e_1_3_6_23_2","doi-asserted-by":"publisher","DOI":"10.1007\/s40670-024-02176-9"},{"key":"e_1_3_6_24_2","doi-asserted-by":"publisher","DOI":"10.1089\/thy.2023.0491"},{"key":"e_1_3_6_25_2","doi-asserted-by":"publisher","DOI":"10.3810\/pgm.2000.03.957"},{"key":"e_1_3_6_26_2","doi-asserted-by":"publisher","DOI":"10.1093\/pch\/18.1.29"},{"key":"e_1_3_6_27_2","doi-asserted-by":"publisher","DOI":"10.3389\/fmed.2024.1380148"},{"key":"e_1_3_6_28_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.ajp.2023.103770"},{"key":"e_1_3_6_29_2","doi-asserted-by":"publisher","DOI":"10.1007\/s40670-025-02293-z"},{"key":"e_1_3_6_30_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-024-63997-7"},{"key":"e_1_3_6_31_2","doi-asserted-by":"publisher","DOI":"10.3389\/fpubh.2025.1566982"},{"key":"e_1_3_6_32_2","doi-asserted-by":"publisher","DOI":"10.1002\/jac5.70038"},{"key":"e_1_3_6_33_2","doi-asserted-by":"publisher","DOI":"10.1186\/s12909-025-06916-2"},{"key":"e_1_3_6_34_2","doi-asserted-by":"publisher","DOI":"10.1101\/2025.02.28.25323115"},{"key":"e_1_3_6_35_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.amjmed.2023.06.012"},{"key":"e_1_3_6_36_2","doi-asserted-by":"publisher","DOI":"10.1186\/s12909-024-05239-y"},{"key":"e_1_3_6_37_2","volume-title":"Multiple choice questions and large languages models: A case study with fictional medical data","author":"Griot M","unstructured":"Griot M, Vanderdonckt J, Yuksel D, et al. Multiple choice questions and large languages models: A case study with fictional medical data. arXiv preprint arXiv:240602394 2024."}],"container-title":["DIGITAL HEALTH"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/20552076261452489","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/full-xml\/10.1177\/20552076261452489","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/journals.sagepub.com\/doi\/pdf\/10.1177\/20552076261452489","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T13:17:35Z","timestamp":1781097455000},"score":1,"resource":{"primary":{"URL":"https:\/\/journals.sagepub.com\/doi\/10.1177\/20552076261452489"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2]]},"references-count":36,"alternative-id":["10.1177\/20552076261452489"],"URL":"https:\/\/doi.org\/10.1177\/20552076261452489","relation":{},"ISSN":["2055-2076","2055-2076"],"issn-type":[{"value":"2055-2076","type":"print"},{"value":"2055-2076","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2]]},"article-number":"20552076261452489"}}