{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"medRxiv"}],"indexed":{"date-parts":[[2026,5,9]],"date-time":"2026-05-09T16:40:58Z","timestamp":1778344858839,"version":"3.51.4"},"posted":{"date-parts":[[2025,1,22]]},"group-title":"Pain Medicine","reference-count":37,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2025,1,22]],"date-time":"2025-01-22T00:00:00Z","timestamp":1737504000000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nd\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2025,2,13]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>Placebo analgesia in chronic pain is a widely studied clinical phenomenon, where expectations about the effectiveness of a treatment can result in substantial pain relief when using an inert treatment agent. While placebos offer an opportunity for non-pharmacological treatment in chronic pain, not everyone demonstrates an analgesic response. Prior research has identified biopsychosocial factors that determine the likelihood of an individual to respond to a placebo, yet generalizability and ecological validity in those studies have been limited due to the inability to account for dynamic personal and treatment effects\u2014which are well-known to play a role. Here, we assessed the potential of using fine-tuned large language models (LLMs) to predict placebo responders in chronic low-back pain using contextual features extracted from patient interviews, as they speak about their lifestyle, pain, and treatment history. We re-analyzed data from two clinical trials where individuals performed open-ended interviews and used these to develop a predictive model of placebo response. Our findings demonstrate that semantic features extracted with LLMs accurately predicted placebo responders, achieving a classification accuracy of 74% in unseen data, and validating with 70% accuracy in an independent cohort. Further, LLMs eliminated the need for pre-selecting search terms or to use dictionary approaches, enabling a fully data-driven approach. This LLM method further provided interpretable insights into psychosocial factors underlying placebo responses, highlighting nuanced linguistic patterns linked to responder status, which tap into semantic dimensions such as \u201canxiety,\u201d \u201cresignation,\u201d and \u201chope.\u201d These findings expand on prior research by integrating state-of-art NLP techniques to address limitations in interpretability and context sensitivity of standard methods like bag-of-words and dictionary-based approaches. This method highlights the role of language models to link language and psychological states, paving the way for a deeper yet quantitative exploration of biopsychosocial phenomena, and to understand how they relate to treatment outcomes, including placebo.<\/jats:p>","DOI":"10.1101\/2025.01.21.25320888","type":"posted-content","created":{"date-parts":[[2025,1,22]],"date-time":"2025-01-22T18:35:16Z","timestamp":1737570916000},"source":"Crossref","is-referenced-by-count":1,"title":["Advancing the prediction and understanding of placebo responses in chronic back pain using large language models"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6614-8556","authenticated-orcid":false,"given":"Diogo A.P.","family":"Nunes","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dan","family":"Furrer","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sara","family":"Berger","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Guillermo","family":"Cecchi","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Joana","family":"Ferreira-Gomes","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Fani","family":"Neto","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8631-2870","authenticated-orcid":false,"given":"David","family":"Martins de Matos","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"A. Vania","family":"Apkarian","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Paulo","family":"Branco","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"54368","reference":[{"key":"2025021423200664000_2025.01.21.25320888v2.1","doi-asserted-by":"publisher","DOI":"10.1097\/j.pain.0000000000001340"},{"key":"2025021423200664000_2025.01.21.25320888v2.2","doi-asserted-by":"publisher","DOI":"10.1136\/bmj.m1668"},{"key":"2025021423200664000_2025.01.21.25320888v2.3","doi-asserted-by":"publisher","DOI":"10.1097\/j.pain.0000000000000333"},{"issue":"2","key":"2025021423200664000_2025.01.21.25320888v2.4","doi-asserted-by":"crossref","first-page":"e254","DOI":"10.1016\/j.bja.2019.01.040","article-title":"Pain, placebo, and test of treatment efficacy: a narrative review","volume":"123","year":"2019","journal-title":"Br J Anaesth"},{"key":"2025021423200664000_2025.01.21.25320888v2.5","doi-asserted-by":"publisher","DOI":"10.1016\/j.tics.2021.07.016"},{"issue":"1","key":"2025021423200664000_2025.01.21.25320888v2.6","doi-asserted-by":"crossref","first-page":"3397","DOI":"10.1038\/s41467-018-05859-1","article-title":"Brain and psychological determinants of placebo pill response in chronic pain patients","volume":"9","year":"2018","journal-title":"Nature Communications"},{"key":"2025021423200664000_2025.01.21.25320888v2.7","doi-asserted-by":"crossref","unstructured":"Vachon-Presseau, E. , et al., Validating a biosignature-predicting placebo pill response in chronic pain in the settings of a randomized controlled trial. PAIN, 2022. 163(5).","DOI":"10.1097\/j.pain.0000000000002450"},{"key":"2025021423200664000_2025.01.21.25320888v2.8","doi-asserted-by":"crossref","unstructured":"Horing, B. , et al., Prediction of placebo responses: a systematic review of the literature. Frontiers in Psychology, 2014. 5.","DOI":"10.3389\/fpsyg.2014.01079"},{"key":"2025021423200664000_2025.01.21.25320888v2.9","doi-asserted-by":"crossref","unstructured":"Berger, S.E. , et al., Quantitative language features identify placebo responders in chronic back pain. PAIN, 2021. 162(6).","DOI":"10.1097\/j.pain.0000000000002175"},{"key":"2025021423200664000_2025.01.21.25320888v2.10","doi-asserted-by":"crossref","unstructured":"Branco, P. , et al., Predicting placebo analgesia in patients with chronic pain using natural language processing: a preliminary validation study. PAIN, 2023. 164(5).","DOI":"10.1097\/j.pain.0000000000002808"},{"key":"2025021423200664000_2025.01.21.25320888v2.11","unstructured":"Vaswani, A. , et al., Attention is all you need, in Proceedings of the 31st International Conference on Neural Information Processing Systems . 2017, Curran Associates Inc.: Long Beach, California, USA. p. 6000\u20136010."},{"issue":"1","key":"2025021423200664000_2025.01.21.25320888v2.12","first-page":"24","article-title":"The Psychological Meaning of Words: LIWC and Computerized Text Analysis Methods","volume":"29","year":"2009","journal-title":"Journal of Language and Social Psychology"},{"key":"2025021423200664000_2025.01.21.25320888v2.13","doi-asserted-by":"publisher","DOI":"10.1002\/(SICI)1097-4571(199009)41:6<391::AID-ASI1>3.0.CO;2-9"},{"issue":"1","key":"2025021423200664000_2025.01.21.25320888v2.14","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1075\/fol.5.1.02hal","article-title":"On the grammar of pain","volume":"5","year":"1998","journal-title":"Functions of Language"},{"key":"2025021423200664000_2025.01.21.25320888v2.15","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2010.600"},{"key":"2025021423200664000_2025.01.21.25320888v2.16","doi-asserted-by":"crossref","unstructured":"Lascaratou, C ., The Language of Pain. 2007: John Benjamins.","DOI":"10.1075\/celcr.9"},{"key":"2025021423200664000_2025.01.21.25320888v2.17","doi-asserted-by":"crossref","first-page":"102342","DOI":"10.1016\/j.is.2023.102342","article-title":"Is text preprocessing still worth the time? A comparative survey on the influence of popular preprocessing methods on Transformers and traditional classifiers","volume":"121","year":"2024","journal-title":"Information Systems"},{"issue":"6","key":"2025021423200664000_2025.01.21.25320888v2.18","doi-asserted-by":"crossref","first-page":"1692","DOI":"10.1097\/j.pain.0000000000002175","article-title":"Quantitative language features identify placebo responders in chronic back pain","volume":"162","year":"2021","journal-title":"Pain"},{"issue":"4","key":"2025021423200664000_2025.01.21.25320888v2.19","doi-asserted-by":"crossref","first-page":"237","DOI":"10.3390\/info14040237","article-title":"Modeling Chronic Pain Experiences from Online Reports Using the Reddit Reports of Chronic Pain Dataset","volume":"14","year":"2023","journal-title":"Information"},{"key":"2025021423200664000_2025.01.21.25320888v2.20","unstructured":"Usage, R.D.P.D.A. Developer Platform & Accessing Reddit Data. 2024; Available from: https:\/\/support.reddithelp.com\/hc\/en-us\/articles\/14945211791892-Developer-Platform-Accessing-Reddit-Data."},{"key":"2025021423200664000_2025.01.21.25320888v2.21","unstructured":"Beltagy, I. , M.E. Peters , and A. Cohan , Longformer: The long-document transformer. arXiv preprint arXiv:2004.05150, 2020."},{"key":"2025021423200664000_2025.01.21.25320888v2.22","unstructured":"Devlin, J. , et al. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. in North American Chapter of the Association for Computational Linguistics . 2019."},{"key":"2025021423200664000_2025.01.21.25320888v2.23","unstructured":"Liu, Y. , Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692, 2019."},{"key":"2025021423200664000_2025.01.21.25320888v2.24","unstructured":"Chen, Q. , Z.-H. Ling , and X. Zhu , Enhancing sentence embedding with generalized pooling. arXiv preprint arXiv:1806.09828, 2018."},{"key":"2025021423200664000_2025.01.21.25320888v2.25","doi-asserted-by":"crossref","unstructured":"Reimers, N. , Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks. arXiv preprint arXiv:1908.10084, 2019.","DOI":"10.18653\/v1\/D19-1410"},{"key":"2025021423200664000_2025.01.21.25320888v2.26","doi-asserted-by":"crossref","first-page":"106492","DOI":"10.1016\/j.neunet.2024.106492","article-title":"Continual pre-training mitigates forgetting in language and vision","volume":"179","year":"2024","journal-title":"Neural Networks"},{"key":"2025021423200664000_2025.01.21.25320888v2.27","doi-asserted-by":"crossref","unstructured":"McCloskey, M. and N.J. Cohen ,Catastrophic Interference in Connectionist Networks: The Sequential Learning Problem, in Psychology of Learning and Motivation, G.H. Bower , Editor. 1989, Academic Press. p. 109-165.","DOI":"10.1016\/S0079-7421(08)60536-8"},{"key":"2025021423200664000_2025.01.21.25320888v2.28","unstructured":"Merity, S. , et al., Pointer sentinel mixture models. arXiv preprint arXiv:1609.07843, 2016."},{"key":"2025021423200664000_2025.01.21.25320888v2.29","doi-asserted-by":"publisher","DOI":"10.1016\/j.cpc.2010.04.018"},{"key":"2025021423200664000_2025.01.21.25320888v2.30","doi-asserted-by":"crossref","unstructured":"Ribeiro, M.T. , S. Singh , and C. Guestrin , \u201cWhy Should I Trust You?\u201d: Explaining the Predictions of Any Classifier, in Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining . 2016, Association for Computing Machinery: San Francisco, California, USA. p. 1135\u20131144.","DOI":"10.1145\/2939672.2939778"},{"key":"2025021423200664000_2025.01.21.25320888v2.31","unstructured":"Lundberg, S. , A unified approach to interpreting model predictions. arXiv preprint arXiv:1705.07874, 2017."},{"key":"2025021423200664000_2025.01.21.25320888v2.32","doi-asserted-by":"crossref","unstructured":"Campello, R.J.G.B. , D. Moulavi , and J. Sander . Density-Based Clustering Based on Hierarchical Density Estimates. in Advances in Knowledge Discovery and Data Mining . 2013. Berlin, Heidelberg: Springer Berlin Heidelberg.","DOI":"10.1007\/978-3-642-37456-2_14"},{"key":"2025021423200664000_2025.01.21.25320888v2.33","unstructured":"Dubey, A. , et al., The llama 3 herd of models. arXiv preprint arXiv:2407.21783, 2024."},{"key":"2025021423200664000_2025.01.21.25320888v2.34","doi-asserted-by":"crossref","unstructured":"Zhang, T. , et al., Benchmarking Large Language Models for News Summarization. Transactions of the Association for Computational Linguistics, 2024. 12: p. 39-57.","DOI":"10.1162\/tacl_a_00632"},{"key":"2025021423200664000_2025.01.21.25320888v2.35","doi-asserted-by":"crossref","unstructured":"LeCun, Y. , Y. Bengio , and G. Hinton , Deep learning. Nature, 2015. 521(7553): p. 436-444.","DOI":"10.1038\/nature14539"},{"key":"2025021423200664000_2025.01.21.25320888v2.36","doi-asserted-by":"crossref","unstructured":"Branco, P. , et al., 57C2.1Predictability of placebo responses and their clinical implications, in Placebo Effects Through the Lens of Translational Research, L. Colloca , et al., Editors. 2023, Oxford University Press. p. 0.","DOI":"10.1093\/med\/9780197645444.003.0005"},{"key":"2025021423200664000_2025.01.21.25320888v2.37","unstructured":"Fadnavis, S. , et al., PainPoints: A Framework for Language-based Detection of Chronic Pain and Expert-Collaborative Text-Summarization. arXiv preprint arXiv:2209.09814, 2022."}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1101\/2025.01.21.25320888","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T20:26:03Z","timestamp":1768508763000},"score":1,"resource":{"primary":{"URL":"http:\/\/medrxiv.org\/lookup\/doi\/10.1101\/2025.01.21.25320888"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,1,22]]},"references-count":37,"URL":"https:\/\/doi.org\/10.1101\/2025.01.21.25320888","relation":{},"subject":[],"published":{"date-parts":[[2025,1,22]]},"subtype":"preprint"}}