{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,22]],"date-time":"2026-10-22T19:03:23Z","timestamp":1792695803319,"version":"4.5.0"},"reference-count":33,"publisher":"Oxford University Press (OUP)","issue":"11","license":[{"start":{"date-parts":[[2026,8,7]],"date-time":"2026-08-07T00:00:00Z","timestamp":1786060800000},"content-version":"vor","delay-in-days":1,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000050","name":"National Heart, Lung, and Blood Institute","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000050","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"publisher","award":["R01HL161458"],"award-info":[{"award-number":["R01HL161458"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000056","name":"National Institute of Nursing Research","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100000056","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Reducing Health Differences Through Informatics Training Program","award":["T32NR007969"],"award-info":[{"award-number":["T32NR007969"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,11,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>To develop and systematically compare a human-led and LLM-assisted hybrid deductive\u2013inductive workflow for qualitative analyses.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Materials and methods<\/jats:title>\n                    <jats:p>We analyzed 122 transcripts (n\u2009=\u200961 research clinical consultations; n\u2009=\u200961 reflexive interviews) from a video ethnography study of patients with heart failure. Human-led thematic analysis used Dedoose software, and LLM-based analysis was conducted using ChatGPT Edu (GPT-5.2; OpenAI) with an eleven-prompt protocol. Both applied a hybrid deductive\u2013inductive approach. The research team compared outputs across 63 human\u2013LLM theme pairs using human consensus and LLM-based evaluation, integrated themes into a final framework, and manually verified quotation fidelity against the original transcripts.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>Human-led and LLM-generated analyses produced complementary cross-cutting themes (7 human; 9 LLM), all judged valid and integrated into 14 final themes across three domains. Thematic overlap was moderate to substantial (Hit Rate 1.00; Jaccard 0.44\u20130.51). Robustness testing across three runs revealed recurrence of five core concepts alongside variability in theme labels and counts. Quotation fidelity showed 68% verbatim, 20% paraphrased, 6% partial and 3% full hallucinations, and 3% truncated excerpts; verbatim quotations did not always clearly support their assigned themes.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>LLM-assisted analysis is feasible for large-scale qualitative health research within a HIPAA-compliant environment using an adaptable eleven-prompt protocol. Human oversight remained essential for contextual interpretation, quotation verification, and assessment of theme\u2013quotation support.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>LLMs are best positioned as analytic partners rather than autonomous coders. Transparent workflows with human-in-the-loop validation are essential for responsible AI integration in health and biomedical informatics.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/jamia\/ocag123","type":"journal-article","created":{"date-parts":[[2026,7,4]],"date-time":"2026-07-04T11:50:37Z","timestamp":1783165837000},"page":"2079-2091","source":"Crossref","is-referenced-by-count":2,"title":["Developing and evaluating human-led and large language model\u2013assisted hybrid deductive\u2013inductive workflows for qualitative analysis"],"prefix":"10.1093","volume":"33","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6493-1950","authenticated-orcid":false,"given":"So Hyeon","family":"Bang","sequence":"first","affiliation":[{"name":"Columbia University School of Nursing , New York, NY,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7130-818X","authenticated-orcid":false,"given":"Soojeong","family":"Han","sequence":"additional","affiliation":[{"name":"Columbia University School of Nursing , New York, NY,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6264-6320","authenticated-orcid":false,"given":"Meghan","family":"Reading Turchioe","sequence":"additional","affiliation":[{"name":"Columbia University School of Nursing , New York, NY,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Melani","family":"Ellison","sequence":"additional","affiliation":[{"name":"Columbia University School of Nursing , New York, NY,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stacey","family":"Dai","sequence":"additional","affiliation":[{"name":"Columbia University School of Nursing , New York, NY,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mary Beth","family":"Happ","sequence":"additional","affiliation":[{"name":"The Ohio State University College of Nursing , Columbus, OH,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8433-0821","authenticated-orcid":false,"given":"David","family":"Russell","sequence":"additional","affiliation":[{"name":"Department of Sociology, Appalachian State University , Boone, NC,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4238-9902","authenticated-orcid":false,"given":"Ruth","family":"Masterson Creber","sequence":"additional","affiliation":[{"name":"Columbia University School of Nursing , New York, NY,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,8,6]]},"reference":[{"key":"2026102204071851600_ocag123-B1","doi-asserted-by":"publisher","first-page":"2743","DOI":"10.1093\/jamia\/ocab195","article-title":"Guidance for publishing qualitative research in informatics","volume":"28","author":"Ancker","year":"2021","journal-title":"J Am Med Inform Assoc"},{"key":"2026102204071851600_ocag123-B2","doi-asserted-by":"publisher","first-page":"11","DOI":"10.18295\/2075-0528.2688","article-title":"Qualitative research and its uses in health care","volume":"8","author":"Al-Busaidi","year":"2008","journal-title":"Sultan Qaboos Univ Med J"},{"key":"2026102204071851600_ocag123-B3","doi-asserted-by":"crossref","first-page":"6","DOI":"10.3928\/00989134-20250307-01","article-title":"Emerging models of care using IT in long-term\/post-acute care: a comparative analysis of human and AI-driven qualitative insights","volume":"51","author":"Alexander","year":"2025","journal-title":"J. Gerontol. Nurs"},{"key":"2026102204071851600_ocag123-B4","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1177\/16094069261426100","article-title":"Large Language Models in qualitative analysis: comparing traditional and researcher-interpreted approaches","volume":"25","author":"Misra","year":"2026","journal-title":"Int J Qual Methods"},{"key":"2026102204071851600_ocag123-B5","doi-asserted-by":"publisher","first-page":"6","DOI":"10.1038\/s41746-023-00970-0","article-title":"Large language models to identify social determinants of health in electronic health records","volume":"7","author":"Guevara","year":"2024","journal-title":"NPJ Digit Med"},{"key":"2026102204071851600_ocag123-B6","doi-asserted-by":"publisher","first-page":"e85057","DOI":"10.2196\/85057","article-title":"Large language model-based analysis of statin therapy discussions and sentiment on social media: cross-sectional observational study","volume":"28","author":"Liu","year":"2026","journal-title":"J Med Internet Res."},{"key":"2026102204071851600_ocag123-B7","doi-asserted-by":"publisher","first-page":"843","DOI":"10.1093\/inthealth\/ihaf015","article-title":"Large language models for analyzing open text in global health surveys: why children are not accessing vaccine services in the Democratic Republic of the Congo","volume":"17","author":"Burstein","year":"2025","journal-title":"Int Health"},{"key":"2026102204071851600_ocag123-B8","doi-asserted-by":"publisher","first-page":"e59050","DOI":"10.2196\/59050","article-title":"ChatGPT for automated qualitative research: Content analysis","volume":"26","author":"Bijker","year":"2024","journal-title":"J Med Internet Res."},{"key":"2026102204071851600_ocag123-B9","doi-asserted-by":"publisher","first-page":"5491","DOI":"10.1007\/s11135-025-02199-3","article-title":"Thematic analysis of interview data with ChatGPT: designing and testing a reliable research protocol for qualitative research","volume":"59","author":"Goyanes","year":"2025","journal-title":"Qual Quant"},{"key":"2026102204071851600_ocag123-B10","doi-asserted-by":"publisher","first-page":"100144","DOI":"10.1016\/j.chbah.2025.100144","article-title":"Harnessing the power of AI in qualitative research: exploring, using and redesigning ChatGPT","volume":"4","author":"Zhang","year":"2025","journal-title":"Comput Hum Behav Artif Hum"},{"key":"2026102204071851600_ocag123-B11","doi-asserted-by":"publisher","DOI":"10.1177\/10778004251401851","article-title":"We reject the use of generative artificial intelligence for reflexive qualitative research","author":"Jowsey","year":"2025","journal-title":"Qual Inq"},{"key":"2026102204071851600_ocag123-B12","doi-asserted-by":"crossref","first-page":"610","DOI":"10.1145\/3442188.3445922","volume-title":"Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency","author":"Bender","year":"2021"},{"key":"2026102204071851600_ocag123-B13","doi-asserted-by":"publisher","first-page":"108356","DOI":"10.1016\/j.cmpb.2024.108356","article-title":"Inductive thematic analysis of healthcare qualitative interviews using open-source large language models: how does it compare to traditional methods?","volume":"255","author":"Mathis","year":"2024","journal-title":"Comput Methods Programs Biomed"},{"key":"2026102204071851600_ocag123-B14","doi-asserted-by":"publisher","author":"Raza","year":"2025","DOI":"10.48550\/arXiv.2502.01620"},{"key":"2026102204071851600_ocag123-B15","doi-asserted-by":"publisher","author":"Xu","DOI":"10.1145\/3828752"},{"issue":"3","key":"2026102204071851600_ocag123-B16","doi-asserted-by":"publisher","first-page":"e70038","DOI":"10.1111\/aphw.70038","article-title":"Generative AI for thematic analysis in a maternal health study: coding semistructured interviews using large language models","volume":"17","author":"Qiao","year":"2025","journal-title":"Appl Psychol Health Well Being"},{"key":"2026102204071851600_ocag123-B17","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1177\/16094069231211248","article-title":"Exploring the use of artificial intelligence for qualitative data analysis: the case of ChatGPT","volume":"22","author":"Morgan","year":"2023","journal-title":"Int J Qual Methods"},{"key":"2026102204071851600_ocag123-B18","doi-asserted-by":"publisher","first-page":"e54974","DOI":"10.2196\/54974","article-title":"Harnessing ChatGPT for thematic analysis: are we ready?","volume":"26","author":"Lee","year":"2024","journal-title":"J Med Internet Res"},{"key":"2026102204071851600_ocag123-B19","doi-asserted-by":"publisher","first-page":"1134","DOI":"10.1097\/ACM.0000000000006134","article-title":"Artificial intelligence to support qualitative data analysis: promises, approaches, pitfalls","volume":"100","author":"Cook","year":"2025","journal-title":"Acad Med"},{"key":"2026102204071851600_ocag123-B20","doi-asserted-by":"publisher","author":"Swain","year":"2018","DOI":"10.4135\/9781526435477"},{"key":"2026102204071851600_ocag123-B21","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1177\/0022146509361185","article-title":"Cultural health capital: a theoretical approach to understanding health care interactions and the dynamics of unequal treatment: a theoretical approach to understanding health care interactions and the dynamics of unequal treatment","volume":"51","author":"Shim","year":"2010","journal-title":"J Health Soc Behav"},{"key":"2026102204071851600_ocag123-B22","doi-asserted-by":"publisher","first-page":"224","DOI":"10.1007\/bf00989485","article-title":"Managing chronic illness at home: three lines of work","volume":"8","author":"Corbin","year":"1985","journal-title":"Qual Sociol"},{"key":"2026102204071851600_ocag123-B23","doi-asserted-by":"publisher","first-page":"1379","DOI":"10.1111\/1467-9566.13114","article-title":"Patient data work: filtering and sensing patient-reported outcomes","volume":"42","author":"Torenholt","year":"2020","journal-title":"Sociol Health Illness"},{"key":"2026102204071851600_ocag123-B24","author":"SocioCultural Research Consultants, LLC","year":"2025"},{"issue":"4","key":"2026102204071851600_ocag123-B25","doi-asserted-by":"publisher","first-page":"e0001189","DOI":"10.1371\/journal.pdig.0001189","article-title":"Large language models for thematic analysis in healthcare research: A blinded mixed-methods comparison with human analysts","volume":"5","author":"Hill","year":"2026","journal-title":"PLOS Digit Health"},{"key":"2026102204071851600_ocag123-B26","doi-asserted-by":"publisher","first-page":"e54482","DOI":"10.2196\/54482","article-title":"Comparing the efficacy and efficiency of human and generative AI: qualitative thematic analyses","volume":"3","author":"Prescott","year":"2024","journal-title":"JMIR AI."},{"issue":"5","key":"2026102204071851600_ocag123-B27","doi-asserted-by":"publisher","first-page":"ooaf126","DOI":"10.1093\/jamiaopen\/ooaf126","article-title":"Leveraging ChatGPT for thematic analysis of medical best practice advisory data","volume":"8","author":"Jeong","year":"2025","journal-title":"JAMIA Open"},{"key":"2026102204071851600_ocag123-B28","doi-asserted-by":"publisher","first-page":"1283","DOI":"10.1007\/s11606-025-10071-1","article-title":"Same prompt, different answer: why AI disagrees with itself","volume":"41","author":"Mee","year":"2026","journal-title":"J Gen Intern Med"},{"key":"2026102204071851600_ocag123-B29","doi-asserted-by":"publisher","author":"Chen","year":"2024","DOI":"10.1162\/99608f92.5317da47"},{"key":"2026102204071851600_ocag123-B30","doi-asserted-by":"publisher","first-page":"1622292","DOI":"10.3389\/frai.2025.1622292","article-title":"Survey and analysis of hallucinations in large language models: attribution to prompting strategies or model behavior","volume":"8","author":"Anh-Hoang","year":"2025","journal-title":"Front Artif Intell"},{"key":"2026102204071851600_ocag123-B31","doi-asserted-by":"publisher","first-page":"364","DOI":"10.1093\/jamia\/ocy173","article-title":"Natural language processing of symptoms documented in free-text narratives of electronic health records: a systematic review","volume":"26","author":"Koleck","year":"2019","journal-title":"J Am Med Inform Assoc"},{"key":"2026102204071851600_ocag123-B32","doi-asserted-by":"publisher","first-page":"e76743","DOI":"10.2196\/76743","article-title":"Real-world evidence synthesis of digital scribes using ambient listening and generative artificial intelligence for clinician documentation workflows: Rapid review","volume":"4","author":"Kanaparthy","year":"2025","journal-title":"Jmir AI."},{"key":"2026102204071851600_ocag123-B33","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1177\/20539517251347598","article-title":"A novel, human-in-the-loop computational grounded theory framework for big social data","volume":"12","author":"Alqazlan","year":"2025","journal-title":"Big Data Soc"}],"container-title":["Journal of the American Medical Informatics Association"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/jamia\/advance-article-pdf\/doi\/10.1093\/jamia\/ocag123\/70512941\/ocag123.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/jamia\/article-pdf\/33\/11\/2079\/70512941\/ocag123.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/jamia\/article-pdf\/33\/11\/2079\/70512941\/ocag123.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,22]],"date-time":"2026-10-22T08:07:25Z","timestamp":1792656445000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/jamia\/article\/33\/11\/2079\/8753705"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,6]]},"references-count":33,"journal-issue":{"issue":"11","published-online":{"date-parts":[[2026,8,6]]},"published-print":{"date-parts":[[2026,11,1]]}},"URL":"https:\/\/doi.org\/10.1093\/jamia\/ocag123","relation":{},"ISSN":["1067-5027","1527-974X"],"issn-type":[{"value":"1067-5027","type":"print"},{"value":"1527-974X","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,11]]},"published":{"date-parts":[[2026,8,6]]}}}