Exploring Biases Related to the Use of Large Language Models in a Multilingual Depression Corpus

Exploring Biases Related to the Use of Large Language Models in a Multilingual Depression Corpus | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Exploring Biases Related to the Use of Large Language Models in a Multilingual Depression Corpus Paula Andrea Pérez-Toro, Judith Dineley, Raquel Iniesta, Yuezhou Zhang, and 17 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5731243/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 16 Oct, 2025 Read the published version in Scientific Reports → Version 1 posted 11 You are reading this latest preprint version Abstract Recent advancements in Large Language Models (LLMs) present promising opportunities for applying these technologies to aid the detection and monitoring of Major Depression Disorder (MDD). However, demographic biases in LLMs may present challenges in the extraction of key information. This study evaluates commonly used LLMs in the speech health literature, across a cohort comprised of English, Spanish, and Dutch speakers with recurrent MDD to observe the effects of different demographic imbalances. Results indicate demographics indeed influence model performance. Gender showed variable impacts across models, with age presenting more pronounced differences. Model performance also varied across language. This study emphasizes the necessity of incorporating demographic-aware models in health-related analyses. It raises awareness of the biases that may affect their application in mental health and suggests further research on methods to mitigate these biases and enhance model generalization. Physical sciences/Mathematics and computing/Applied mathematics Physical sciences/Mathematics and computing/Computational science Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 16 Oct, 2025 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 12 May, 2025 Reviews received at journal 25 Feb, 2025 Reviews received at journal 18 Feb, 2025 Reviewers agreed at journal 07 Feb, 2025 Reviewers agreed at journal 07 Feb, 2025 Reviewers agreed at journal 07 Feb, 2025 Reviewers invited by journal 07 Feb, 2025 Editor assigned by journal 07 Feb, 2025 Editor invited by journal 09 Jan, 2025 Submission checks completed at journal 08 Jan, 2025 First submitted to journal 29 Dec, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5731243","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":402043645,"identity":"57f79a06-8a14-478e-9412-f88bd38623e4","order_by":0,"name":"Paula Andrea Pérez-Toro","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABFklEQVRIiWNgGAWjYHACxgNg6gBcgPnAB5BwAx496FrYEmeQqoXHEK8Wc/YzBgd+MNjl8d0+fPAD4w6bfH7+Mx8bfrYxyPbj0GLZk2NwsIchuVjyXFqyBOOZNMuZDWc3Nva2MRjPxGGNwYEcgwM8DMyJG87wGEgwth02MDjYu/0BbxtD4oYDOLScf2Nw8A9DPVAL/+cfjG3/DewP8zxs/AvUsh+Xlhs5Bod5GA6DbGED2nLAwICNh7EZbAsuv8x4VnBYxuB44swzbGYWiWeSDSTOsBk2y5yTMJ6BwxZz/uSND99UVCf2nWF+fOPjDjsD/v7DDxvflNnI9uPyPhLJwJCIUCWBw1kIxRCAN8ZHwSgYBaNgxAIAkqBkV2D/DN8AAAAASUVORK5CYII=","orcid":"","institution":"Institute of Psychiatry, Psychology \u0026 Neuroscience, King's College London","correspondingAuthor":true,"prefix":"","firstName":"Paula","middleName":"Andrea","lastName":"Pérez-Toro","suffix":""},{"id":402043646,"identity":"8ad0beee-2fbd-4eee-b6b0-a12bdbbeddaa","order_by":1,"name":"Judith Dineley","email":"","orcid":"","institution":"Institute of Psychiatry, Psychology \u0026 Neuroscience, King's College London","correspondingAuthor":false,"prefix":"","firstName":"Judith","middleName":"","lastName":"Dineley","suffix":""},{"id":402043648,"identity":"746750d4-ffdc-4175-a94c-79b615c8f905","order_by":2,"name":"Raquel Iniesta","email":"","orcid":"","institution":"Institute of Psychiatry, Psychology \u0026 Neuroscience, King's College London","correspondingAuthor":false,"prefix":"","firstName":"Raquel","middleName":"","lastName":"Iniesta","suffix":""},{"id":402043649,"identity":"08275a96-a096-4402-983d-8b336ab68c58","order_by":3,"name":"Yuezhou Zhang","email":"","orcid":"","institution":"Institute of Psychiatry, Psychology \u0026 Neuroscience, King's College London","correspondingAuthor":false,"prefix":"","firstName":"Yuezhou","middleName":"","lastName":"Zhang","suffix":""},{"id":402043651,"identity":"4d33033f-1f56-4cea-9ce8-360e5756c3a3","order_by":4,"name":"Faith Matcham","email":"","orcid":"","institution":"Institute of Psychiatry, Psychology \u0026 Neuroscience, King's College London","correspondingAuthor":false,"prefix":"","firstName":"Faith","middleName":"","lastName":"Matcham","suffix":""},{"id":402043652,"identity":"fdb9a958-675c-405a-a3b7-51d373345439","order_by":5,"name":"Sara Siddi","email":"","orcid":"","institution":"Parc Sanitari Sant Joan de Deu, Fundació Sant Joan de Deu, CIBERSAM, Universitat de Barcelona","correspondingAuthor":false,"prefix":"","firstName":"Sara","middleName":"","lastName":"Siddi","suffix":""},{"id":402043654,"identity":"eace3887-7bf5-4e54-b9f2-26558f76bf1b","order_by":6,"name":"Femke Lamers","email":"","orcid":"","institution":"Department of Psychiatry, Amsterdam Public Health Research Institute and Amsterdam Neuroscience, Amsterdam University Medical Centre, Vrije Universiteit and GGZ InGeest","correspondingAuthor":false,"prefix":"","firstName":"Femke","middleName":"","lastName":"Lamers","suffix":""},{"id":402043656,"identity":"990bebac-2de5-4cd4-bebd-444d52374ee5","order_by":7,"name":"Josep Maria Haro","email":"","orcid":"","institution":"Parc Sanitari Sant Joan de Deu, Fundació Sant Joan de Deu, CIBERSAM, Universitat de Barcelona","correspondingAuthor":false,"prefix":"","firstName":"Josep","middleName":"Maria","lastName":"Haro","suffix":""},{"id":402043657,"identity":"ba6f90c4-6a4f-46fa-9745-28324ab7ee4a","order_by":8,"name":"Brenda W.J.H. Penninx","email":"","orcid":"","institution":"Department of Psychiatry, Amsterdam Public Health Research Institute and Amsterdam Neuroscience, Amsterdam University Medical Centre, Vrije Universiteit and GGZ InGeest","correspondingAuthor":false,"prefix":"","firstName":"Brenda","middleName":"W.J.H.","lastName":"Penninx","suffix":""},{"id":402043659,"identity":"ffa0bea0-31ec-4dd7-843f-4e9644d805f4","order_by":9,"name":"Amos A. Folarin","email":"","orcid":"","institution":"Institute of Psychiatry, Psychology \u0026 Neuroscience, King's College London","correspondingAuthor":false,"prefix":"","firstName":"Amos","middleName":"A.","lastName":"Folarin","suffix":""},{"id":402043660,"identity":"a1ed092e-02ea-4b03-a1bd-d6d93afbf5a7","order_by":10,"name":"Tomás Arias-Vergara","email":"","orcid":"","institution":"Pattern Recognition Lab. Friedrich-Alexander-Universität Erlangen-Nürnberg","correspondingAuthor":false,"prefix":"","firstName":"Tomás","middleName":"","lastName":"Arias-Vergara","suffix":""},{"id":402043661,"identity":"537ae3a5-945b-4cea-8f9b-a8513c2c4464","order_by":11,"name":"Juan Rafael Orozco-Arroyave","email":"","orcid":"","institution":"GITA Lab, Faculty of Engineering. Universidad de Antioquia","correspondingAuthor":false,"prefix":"","firstName":"Juan","middleName":"Rafael","lastName":"Orozco-Arroyave","suffix":""},{"id":402043662,"identity":"9fd791e2-16f7-4db3-b2dd-1fc0e4fe5224","order_by":12,"name":"Elmar Nöth","email":"","orcid":"","institution":"Pattern Recognition Lab. Friedrich-Alexander-Universität Erlangen-Nürnberg","correspondingAuthor":false,"prefix":"","firstName":"Elmar","middleName":"","lastName":"Nöth","suffix":""},{"id":402043663,"identity":"ee5ac9d1-e0b2-496e-bed4-1c32c7fd9107","order_by":13,"name":"Andreas Maier","email":"","orcid":"","institution":"Pattern Recognition Lab. Friedrich-Alexander-Universität Erlangen-Nürnberg","correspondingAuthor":false,"prefix":"","firstName":"Andreas","middleName":"","lastName":"Maier","suffix":""},{"id":402043664,"identity":"a7729005-14bb-4b4e-b0dd-80fe2894cadd","order_by":14,"name":"Til Wykes","email":"","orcid":"","institution":"Institute of Psychiatry, Psychology \u0026 Neuroscience, King's College London","correspondingAuthor":false,"prefix":"","firstName":"Til","middleName":"","lastName":"Wykes","suffix":""},{"id":402043665,"identity":"e3c3cf7b-0a0f-4d6f-807a-84649caa2f85","order_by":15,"name":"Srinivasan Vairavan","email":"","orcid":"","institution":"Institute of Psychiatry, Psychology \u0026 Neuroscience, King's College London","correspondingAuthor":false,"prefix":"","firstName":"Srinivasan","middleName":"","lastName":"Vairavan","suffix":""},{"id":402043666,"identity":"82ae923d-e094-4aa8-82ac-ad582420ce1b","order_by":16,"name":"Richard Dobson","email":"","orcid":"","institution":"Institute of Psychiatry, Psychology \u0026 Neuroscience, King's College London","correspondingAuthor":false,"prefix":"","firstName":"Richard","middleName":"","lastName":"Dobson","suffix":""},{"id":402043667,"identity":"ba4b1295-c293-4ac6-9f31-49d484452e4f","order_by":17,"name":"Vaibhav A. Narayan","email":"","orcid":"","institution":"Davos Alzheimer's Collaborative","correspondingAuthor":false,"prefix":"","firstName":"Vaibhav","middleName":"A.","lastName":"Narayan","suffix":""},{"id":402043668,"identity":"b2e31d8c-4bbe-48a8-8611-8ebb6952c951","order_by":18,"name":"Matthew Hotopf","email":"","orcid":"","institution":"Institute of Psychiatry, Psychology \u0026 Neuroscience, King's College London","correspondingAuthor":false,"prefix":"","firstName":"Matthew","middleName":"","lastName":"Hotopf","suffix":""},{"id":402043669,"identity":"d5fea222-32dd-43ef-870f-84816e847840","order_by":19,"name":"Nicholas Cummins","email":"","orcid":"","institution":"Institute of Psychiatry, Psychology \u0026 Neuroscience, King's College London","correspondingAuthor":false,"prefix":"","firstName":"Nicholas","middleName":"","lastName":"Cummins","suffix":""},{"id":402043670,"identity":"5941bf80-b8ef-414d-844c-09d4fd39021c","order_by":20,"name":"The RADAR-CNS Consortium","email":"","orcid":"","institution":"The RADAR-CNS Consortium: www.radar-cns.org","correspondingAuthor":false,"prefix":"","firstName":"The","middleName":"RADAR-CNS","lastName":"Consortium","suffix":""}],"badges":[],"createdAt":"2024-12-29 18:23:55","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5731243/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5731243/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-025-19980-x","type":"published","date":"2025-10-16T15:57:51+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":93956155,"identity":"d083cc0f-6a7c-4d67-90f9-e2f858997ea1","added_by":"auto","created_at":"2025-10-20 16:11:16","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":441283,"visible":true,"origin":"","legend":"","description":"","filename":"BiasesLLMsDepressionScientificReportsPerezToro.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5731243/v1_covered_4520371c-3055-451c-a7c2-a71df74c4f14.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Exploring Biases Related to the Use of Large Language Models in a Multilingual Depression Corpus","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-5731243/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5731243/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"Recent advancements in Large Language Models (LLMs) present promising opportunities for applying these technologies to aid the detection and monitoring of Major Depression Disorder (MDD). However, demographic biases in LLMs may present challenges in the extraction of key information. This study evaluates commonly used LLMs in the speech health literature, across a cohort comprised of English, Spanish, and Dutch speakers with recurrent MDD to observe the effects of different demographic imbalances. Results indicate demographics indeed influence model performance. Gender showed variable impacts across models, with age presenting more pronounced differences. Model performance also varied across language. This study emphasizes the necessity of incorporating demographic-aware models in health-related analyses. It raises awareness of the biases that may affect their application in mental health and suggests further research on methods to mitigate these biases and enhance model generalization.","manuscriptTitle":"Exploring Biases Related to the Use of Large Language Models in a Multilingual Depression Corpus","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-01-15 16:54:39","doi":"10.21203/rs.3.rs-5731243/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-05-12T07:09:12+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-02-25T14:21:18+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-02-18T13:53:48+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"325800598274465915195724619021003121101","date":"2025-02-07T18:11:40+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"93814257264859713751416339328557876745","date":"2025-02-07T17:04:29+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"120770033345080309663377597667171487744","date":"2025-02-07T15:43:17+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-02-07T15:18:51+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-02-07T15:15:27+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-01-09T15:51:46+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-01-09T04:37:32+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2024-12-29T18:15:33+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"9dd9f577-5a13-40b5-9131-9dea1c79cbdc","owner":[],"postedDate":"January 15th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":42850152,"name":"Physical sciences/Mathematics and computing/Applied mathematics"},{"id":42850153,"name":"Physical sciences/Mathematics and computing/Computational science"}],"tags":[],"updatedAt":"2025-10-20T16:06:38+00:00","versionOfRecord":{"articleIdentity":"rs-5731243","link":"https://doi.org/10.1038/s41598-025-19980-x","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2025-10-16 15:57:51","publishedOnDateReadable":"October 16th, 2025"},"versionCreatedAt":"2025-01-15 16:54:39","video":"","vorDoi":"10.1038/s41598-025-19980-x","vorDoiUrl":"https://doi.org/10.1038/s41598-025-19980-x","workflowStages":[]},"version":"v1","identity":"rs-5731243","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5731243","identity":"rs-5731243","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}