{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T15:09:59Z","timestamp":1778684999997,"version":"3.51.4"},"reference-count":28,"publisher":"BMJ","issue":"1","license":[{"start":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T00:00:00Z","timestamp":1778630400000},"content-version":"unspecified","delay-in-days":132,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000005","name":"Department of Defense","doi-asserted-by":"crossref","award":["W81XWH-22-1-0894"],"award-info":[{"award-number":["W81XWH-22-1-0894"]}],"id":[{"id":"10.13039\/100000005","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["BMJ Neurol Open"],"accepted":{"date-parts":[[2026,4,22]]},"published-print":{"date-parts":[[2026,1]]},"abstract":"<jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>To develop and validate a large language model (LLM) prompt capable of ascertaining a patient\u2019s depression from their multiple sclerosis (MS) neurologist\u2019s note and to explore its potential for earlier detection of depression in MS care.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials and methods<\/jats:title>\n                    <jats:p>This single-center retrospective study analysed prospectively collected electronic health record notes. In phase I, an institutionally secure ChatGPT-4 prompt was iteratively refined to infer the presence of depression using the neurologist\u2019s note and compared with manual annotation of the neurologist\u2019s impression (depression: present, absent, no mention) and patient-reported outcomes (PROs): Hospital Anxiety and Depression Scale or Patient Health Questionnaire-9. In phase II, longitudinal analysis compared timing of depression detection by the prompt and the neurologist across 5 years of notes for 250 patients.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>In phase I (n=278 adults with MS), the LLM prompt detected depression in 60.4% of notes (168\/278). When compared with neurologist impression in the clinical notes, the prompt achieved 97.3% sensitivity and 84.4% accuracy. Specificity was more modest (68.3%): when neurologists did not mention depression, the prompt inferred depression based on symptoms, history and medications. When PRO and neurologist impression disagreed, the prompt aligned with PROs 61.9% of the time. In phase II, the LLM inferred depression earlier than the neurologist in 18.8% of patients, at an average of 2.45 (SD 1.54) years earlier.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>The prompt was highly sensitive to neurologist documentation of depression in clinical notes; it inferred both present\/treated depression from other note components. Potential applications include quality improvement initiatives aiming to improve depression care on a cohort level.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1136\/bmjno-2025-001502","type":"journal-article","created":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T14:20:11Z","timestamp":1778682011000},"page":"e001502","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":0,"title":["Large language models to infer depression in patients with neurological conditions"],"prefix":"10.1136","volume":"8","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-1789-1876","authenticated-orcid":false,"given":"Maya","family":"Julian-Kwong","sequence":"first","affiliation":[{"name":"Neurology, University of California San Francisco Weill Institute for Neurosciences, San Francisco, California, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shane","family":"Poole","sequence":"additional","affiliation":[{"name":"Neurology, University of California San Francisco Weill Institute for Neurosciences, San Francisco, California, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kyra","family":"Henderson","sequence":"additional","affiliation":[{"name":"Neurology, University of California San Francisco Weill Institute for Neurosciences, San Francisco, California, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jaeleene","family":"Wijangco","sequence":"additional","affiliation":[{"name":"Neurology, University of California San Francisco Weill Institute for Neurosciences, San Francisco, California, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-7270-595X","authenticated-orcid":false,"given":"Nikki","family":"Sisodia","sequence":"additional","affiliation":[{"name":"Neurology, University of California San Francisco Weill Institute for Neurosciences, San Francisco, California, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5121-3319","authenticated-orcid":false,"given":"Jeffrey Marc","family":"Gelfand","sequence":"additional","affiliation":[{"name":"Neurology, University of California San Francisco Weill Institute for Neurosciences, San Francisco, California, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chu-Yueh","family":"Guo","sequence":"additional","affiliation":[{"name":"Neurology, University of California San Francisco Weill Institute for Neurosciences, San Francisco, California, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2034-8800","authenticated-orcid":false,"given":"Riley","family":"Bove","sequence":"additional","affiliation":[{"name":"Neurology, University of California San Francisco Weill Institute for Neurosciences, San Francisco, California, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"239","published-online":{"date-parts":[[2026,5,13]]},"reference":[{"key":"2026051307200567000_8.1.e001502.1","doi-asserted-by":"crossref","DOI":"10.2196\/20346","article-title":"The Effectiveness of Artificial Intelligence Conversational Agents in Health Care: Systematic Review","volume":"22","author":"Milne-Ives","year":"2020","journal-title":"J Med Internet Res"},{"key":"2026051307200567000_8.1.e001502.2","doi-asserted-by":"publisher","DOI":"10.1080\/0142159X.2023.2271159"},{"key":"2026051307200567000_8.1.e001502.3","article-title":"The role of artificial intelligence in hastening time to recruitment in clinical trials","volume":"5","author":"Ismail","year":"2023","journal-title":"BJR Open"},{"key":"2026051307200567000_8.1.e001502.4","doi-asserted-by":"crossref","first-page":"308","DOI":"10.1002\/ana.27251","article-title":"Detecting New Lesions Using a Large Language Model: Applications in Real-World Multiple Sclerosis Datasets","volume":"98","author":"Poole","year":"2025","journal-title":"Ann Neurol"},{"key":"2026051307200567000_8.1.e001502.5","doi-asserted-by":"crossref","DOI":"10.1186\/s11689-025-09612-w","article-title":"Automated extraction of functional biomarkers of verbal and ambulatory ability from multi-institutional clinical notes using large language models","volume":"17","author":"Kaster","year":"2025","journal-title":"J Neurodev Disord"},{"key":"2026051307200567000_8.1.e001502.6","doi-asserted-by":"crossref","first-page":"1972","DOI":"10.1212\/WNL.0000000000002174","article-title":"Differences in the burden of psychiatric comorbidity in MS vs the general population","volume":"85","author":"Marrie","year":"2015","journal-title":"Neurology (ECronicon)"},{"key":"2026051307200567000_8.1.e001502.7","doi-asserted-by":"publisher","DOI":"10.1080\/09540261.2017.1322555"},{"key":"2026051307200567000_8.1.e001502.8","doi-asserted-by":"publisher","DOI":"10.1177\/1352458514564487"},{"key":"2026051307200567000_8.1.e001502.9","doi-asserted-by":"crossref","DOI":"10.1007\/s10916-016-0677-0","article-title":"Literature Review of Patient Record Structures from the Physician\u2019s Perspective","volume":"41","author":"Forsvik","year":"2017","journal-title":"J Med Syst"},{"key":"2026051307200567000_8.1.e001502.10","doi-asserted-by":"crossref","first-page":"408","DOI":"10.1177\/1352458517747407","article-title":"Harnessing electronic medical records to advance research on multiple sclerosis","volume":"25","author":"Damotte","year":"2019","journal-title":"Mult Scler"},{"key":"2026051307200567000_8.1.e001502.11","unstructured":"A closed-loop digital health tool to improve depression care in multiple sclerosis (MS CATCH). ClinicalTrials.gov identifier: NCT05865405 . 2024. Available: https:\/\/clinicaltrials.gov\/ct2\/show\/NCT05865405 [Accessed 6 Sep 2025]."},{"key":"2026051307200567000_8.1.e001502.12","unstructured":"Multiple Sclerosis Falls Insight Track (MS FIT) . ClinicalTrials.gov identifier: NCT05837949 . 2024. Available: https:\/\/clinicaltrials.gov\/ct2\/show\/NCT05837949 [Accessed 6 Sep 2025]."},{"key":"2026051307200567000_8.1.e001502.13","doi-asserted-by":"publisher","DOI":"10.1046\/j.1525-1497.2001.016009606.x"},{"key":"2026051307200567000_8.1.e001502.14","doi-asserted-by":"publisher","DOI":"10.1016\/S0022-3999(01)00296-3"},{"key":"2026051307200567000_8.1.e001502.15","doi-asserted-by":"crossref","first-page":"707","DOI":"10.1053\/j.gastro.2023.12.019","article-title":"A Comparison of a Large Language Model vs Manual Chart Review for the Extraction of Data Elements From the Electronic Health Record","volume":"166","author":"Ge","year":"2024","journal-title":"Gastroenterology"},{"key":"2026051307200567000_8.1.e001502.16","unstructured":"Microsoft. n.d. Available: https:\/\/learn.microsoft.com\/en-us\/azure\/ai-services\/openai\/concepts\/models?tabs=global-standard%2Cstandard-chat-completions"},{"key":"2026051307200567000_8.1.e001502.17","doi-asserted-by":"publisher","DOI":"10.1016\/j.msard.2015.04.001"},{"key":"2026051307200567000_8.1.e001502.18","first-page":"1639","article-title":"Using Medical Text Extraction, Reasoning and Mapping System (MTERMS) to process medication information in outpatient clinical notes","volume":"2011","author":"Zhou","year":"2011","journal-title":"AMIA Annu Symp Proc"},{"key":"2026051307200567000_8.1.e001502.19","first-page":"452","article-title":"Performance of Artificial Intelligence in Predicting Future Depression Levels","volume":"305","author":"Aziz","year":"2023","journal-title":"Stud Health Technol Inform"},{"key":"2026051307200567000_8.1.e001502.20","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2023.104336"},{"key":"2026051307200567000_8.1.e001502.21","doi-asserted-by":"crossref","first-page":"1629725","DOI":"10.3389\/fcomp.2025.1629725","article-title":"Large language models for depression recognition in spoken language integrating psychological knowledge","volume":"7","author":"Li","year":"2025","journal-title":"Front Comput Sci"},{"key":"2026051307200567000_8.1.e001502.22","doi-asserted-by":"crossref","DOI":"10.1038\/s41746-025-01982-8","article-title":"Using a fine-tuned large language model for symptom-based depression evaluation","volume":"8","author":"Weber","year":"2025","journal-title":"NPJ Digit Med"},{"key":"2026051307200567000_8.1.e001502.23","doi-asserted-by":"publisher","DOI":"10.1080\/14737175.2020.1746645"},{"key":"2026051307200567000_8.1.e001502.24","unstructured":"American Academy of Neurology . Universal neurology quality measurement set. 2018. Available: https:\/\/www.aan.com\/siteassets\/home-page\/policy-and-guidelines\/quality\/quality-measures\/other-neurologic-conditions\/2018universalneurologymeasurementset.pdf [Accessed 7 Sep 2025]."},{"key":"2026051307200567000_8.1.e001502.25","doi-asserted-by":"crossref","first-page":"105648","DOI":"10.1016\/j.msard.2024.105648","article-title":"Prevalence, treatment and correlates of depression in multiple sclerosis","volume":"87","author":"Young","year":"2024","journal-title":"Mult Scler Relat Disord"},{"key":"2026051307200567000_8.1.e001502.26","doi-asserted-by":"publisher","DOI":"10.1016\/j.neurol.2014.07.017"},{"key":"2026051307200567000_8.1.e001502.27","doi-asserted-by":"crossref","DOI":"10.1001\/jamanetworkopen.2024.13140","article-title":"Virtual Scribes and Physician Time Spent on Electronic Health Records","volume":"7","author":"Rotenstein","year":"2024","journal-title":"JAMA Netw Open"},{"key":"2026051307200567000_8.1.e001502.28","doi-asserted-by":"crossref","first-page":"103705","DOI":"10.1016\/j.ajp.2023.103705","article-title":"Artificial intelligence in psychiatry research, diagnosis, and therapy","volume":"87","author":"Sun","year":"2023","journal-title":"Asian J Psychiatr"}],"container-title":["BMJ Neurology Open"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/bmjno-2025-001502","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,13]],"date-time":"2026-05-13T14:20:16Z","timestamp":1778682016000},"score":1,"resource":{"primary":{"URL":"https:\/\/neurologyopen.bmj.com\/lookup\/doi\/10.1136\/bmjno-2025-001502"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1]]},"references-count":28,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,5,13]]},"published-print":{"date-parts":[[2026,1]]}},"alternative-id":["10.1136\/bmjno-2025-001502"],"URL":"https:\/\/doi.org\/10.1136\/bmjno-2025-001502","relation":{},"ISSN":["2632-6140"],"issn-type":[{"value":"2632-6140","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1]]}}}