{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,22]],"date-time":"2026-10-22T09:55:31Z","timestamp":1792662931544,"version":"4.5.0"},"reference-count":43,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J Med Internet Res"],"abstract":"<jats:sec>\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Large language models including GPT-4 (OpenAI) have opened new avenues in health care and qualitative research. Traditional qualitative methods are time-consuming and require expertise to capture nuance. Although large language models have demonstrated enhanced contextual understanding and inferencing compared with traditional natural language processing, their performance in qualitative analysis versus that of humans remains unexplored.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Objective<\/jats:title>\n                    <jats:p>We evaluated the effectiveness of GPT-4 versus human researchers in qualitative analysis of interviews with patients with adult-acquired buried penis (AABP).<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>Qualitative data were obtained from semistructured interviews with 20 patients with AABP. Human analysis involved a structured 3-stage process\u2014initial observations, line-by-line coding, and consensus discussions to refine themes. In contrast, artificial intelligence (AI) analysis with GPT-4 underwent two phases: (1) a na\u00efve phase, where GPT-4 outputs were independently evaluated by a blinded reviewer to identify themes and subthemes and (2) a comparison phase, where AI-generated themes were compared with human-identified themes to assess agreement. We used a general qualitative description approach.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>The study population (N=20) comprised predominantly White (17\/20, 85%), married (12\/20, 60%), heterosexual (19\/20, 95%) men, with a mean age of 58.8 years and BMI of 41.1 kg\/m2. Human qualitative analysis identified \u201curinary issues\u201d in 95% (19\/20) and GPT-4 in 75% (15\/20) of interviews, with the subtheme \u201cspray or stream\u201d noted in 60% (12\/20) and 35% (7\/20), respectively. \u201cSexual issues\u201d were prominent (19\/20, 95% humans vs 16\/20, 80% GPT-4), although humans identified a wider range of subthemes, including \u201cpain with sex or masturbation\u201d (7\/20, 35%) and \u201cdifficulty with sex or masturbation\u201d (4\/20, 20%). Both analyses similarly highlighted \u201cmental health issues\u201d (11\/20, 55%, both), although humans coded \u201cdepression\u201d more frequently (10\/20, 50% humans vs 4\/20, 20% GPT-4). Humans frequently cited \u201cissues using public restrooms\u201d (12\/20, 60%) as impacting social life, whereas GPT-4 emphasized \u201cstruggles with romantic relationships\u201d (9\/20, 45%). \u201cHygiene issues\u201d were consistently recognized (14\/20, 70% humans vs 13\/20, 65% GPT-4). Humans uniquely identified \u201ccontributing factors\u201d as a theme in all interviews. There was moderate agreement between human and GPT-4 coding (\u03ba=0.401). Reliability assessments of GPT-4\u2019s analyses showed consistent coding for themes including \u201cbody image struggles,\u201d \u201cchronic pain\u201d (10\/10, 100%), and \u201cdepression\u201d (9\/10, 90%). Other themes like \u201cmotivation for surgery\u201d and \u201cweight challenges\u201d were reliably coded (8\/10, 80%), while less frequent themes were variably identified across multiple iterations.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Large language models including GPT-4 can effectively identify key themes in analyzing qualitative health care data, showing moderate agreement with human analysis. While human analysis provided a richer diversity of subthemes, the consistency of AI suggests its use as a complementary tool in qualitative research. With AI rapidly advancing, future studies should iterate analyses and circumvent token limitations by segmenting data, furthering the breadth and depth of large language model\u2013driven qualitative analyses.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.2196\/56500","type":"journal-article","created":{"date-parts":[[2024,7,9]],"date-time":"2024-07-09T09:16:17Z","timestamp":1720516577000},"page":"e56500","source":"Crossref","is-referenced-by-count":57,"title":["Comparing GPT-4 and Human Researchers in Health Care Data Analysis: Qualitative Description Study"],"prefix":"10.2196","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0550-0250","authenticated-orcid":false,"given":"Kevin Danis","family":"Li","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9918-5783","authenticated-orcid":false,"given":"Adrian M","family":"Fernandez","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8125-9492","authenticated-orcid":false,"given":"Rachel","family":"Schwartz","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2583-1241","authenticated-orcid":false,"given":"Natalie","family":"Rios","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-6467-0758","authenticated-orcid":false,"given":"Marvin Nathaniel","family":"Carlisle","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5843-8641","authenticated-orcid":false,"given":"Gregory M","family":"Amend","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6475-670X","authenticated-orcid":false,"given":"Hiren V","family":"Patel","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0532-048X","authenticated-orcid":false,"given":"Benjamin N","family":"Breyer","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2024,8,21]]},"reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1016\/j.metrad.2023.100017"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1038\/s43856-023-00370-1"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1186\/s13040-023-00339-9"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1016\/j.iotcps.2023.04.003"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.2196\/48933"},{"issue":"9","key":"ref6","first-page":"369","volume":"3","author":"Almeida, F","year":"2017","journal-title":"Eur J Educ Stud"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1016\/j.jdin.2023.10.001"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1177\/10497323231217392"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1136\/fmch-2021-001287"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.3233\/jid-220013"},{"key":"ref11","author":"Achiam, J","journal-title":"ArXiv"},{"key":"ref12","author":"Zhang, H","journal-title":"ArXiv"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1177\/16094069231201504"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1177\/16094069231211248"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1177\/10497323241244669"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1186\/s12909-024-05316-2"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0300024"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1097\/JU.0000000000002667"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.2196\/50638"},{"key":"ref20","unstructured":"Otter.ai - AI Meeting Note Taker & Real-time AI Transcription2024-01-17https:\/\/otter.ai\/"},{"key":"ref21","unstructured":"One platform to connectZoom2024-01-17https:\/\/zoom.us\/"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1002\/1098-240x(200008)23:4<334::aid-nur9>3.0.co;2-g"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1002\/nur.20362"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1186\/1471-2288-9-52"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1186\/s12874-016-0200-9"},{"key":"ref26","unstructured":"Research with Dr Kriukow Youtube page202305192024-07-25Thematic analysis with ChatGPT | PART 1- Coding qualitative Data with ChatGPThttps:\/\/www.youtube.com\/watch?v=8dTs7D42ge0"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1016\/j.glmedi.2024.100051"},{"key":"ref28","unstructured":"Bengfort JNow Available: Versa, UCSF Generative AI Platform20242024-07-25Office of the Chancellorhttps:\/\/chancellor.ucsf.edu\/news\/now-available-versa-ucsf-generative-ai-platform"},{"key":"ref29","unstructured":"KohnRMastering token limits and memory in ChatGPT and other large language models20232024-07-25Mediumhttps:\/\/medium.com\/@russkohn\/mastering-ai-token-limits-and-memory-ce920630349a"},{"key":"ref30","author":"Baktash, JA","journal-title":"ArXiv"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyg.2023.1199058"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.3389\/fpsyg.2021.605191"},{"key":"ref33","unstructured":"TennySBrannanJBrannanGQualitative StudyStatPearls [Internet]2024-07-06https:\/\/www.ncbi.nlm.nih.gov\/books\/NBK470395\/"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.2196\/49771"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.2196\/51229"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.2196\/46482"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pdig.0000198"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2023.36997"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1001\/jamanetworkopen.2023.43689"},{"key":"ref40","unstructured":"RedmondWVeronaWMicrosoft and epic expand strategic collaboration with integration of azure openAI service202304172024-07-25Microsoft News Centerhttps:\/\/news.microsoft.com\/2023\/04\/17\/microsoft-and-epic-expand-strategic-collaboration-with-integration-of-azure-openai-service\/"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1038\/s41746-023-00873-0"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1038\/d41586-023-00816-5"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1016\/S2589-7500(23)00225-X"}],"container-title":["Journal of Medical Internet Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.jmir.org\/2024\/1\/e56500","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T17:45:31Z","timestamp":1782841531000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.jmir.org\/2024\/1\/e56500"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,8,21]]},"references-count":43,"URL":"https:\/\/doi.org\/10.2196\/56500","relation":{},"ISSN":["1438-8871"],"issn-type":[{"value":"1438-8871","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,8,21]]}}}