{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"medRxiv"}],"indexed":{"date-parts":[[2026,10,26]],"date-time":"2026-10-26T22:11:55Z","timestamp":1793052715291,"version":"4.5.0"},"posted":{"date-parts":[[2026,2,18]]},"group-title":"Neurology","reference-count":38,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2026,2,18]],"date-time":"2026-02-18T00:00:00Z","timestamp":1771372800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.medrxiv.org\/about\/FAQ#license"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,2,18]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>Causal discovery algorithms are often leveraged for inferring causal relationships and recovering a causal model from data. However, causal discovery from data alone is limited by the structural constraints of the used dataset, the lack of causal logic, and the lack of external knowledge. Thus, data-driven causal discovery can only suggest possible causal relationships at best. To overcome these limitations, Large Language Models (LLMs) and knowledge systems, such as Retrieval-Augmented Generation (RAG), have been proposed as alternatives to data-driven causal discovery and as a method to augment causal discovery algorithms. Using an expert-defined causal graph of chronic lower back pain, we further propose knowledge graph based RAG systems, such as GraphRAG, as an improvement over RAG systems for augmenting causal discovery (F1 0.745), benchmarking its performance against augmenting causal discovery with an LLM (F1 0.636), augmenting causal discovery with RAG (F1 0.714), and causal discovery alone (F1 0.396). We also explore the impact of different prompting methods for causality, such as querying for the plausibility of causal relationships, the presence of statistical associations, and the existence of temporal causal relationships, as inspired by the methodology of the domain experts constructing our ground truth. Lastly, we discuss how applications of LLMs, RAG, and graph-based RAG systems can impact and accelerate the causal modeling of chronic lower back pain by bridging the gap between domain knowledge and data driven approaches to causal modeling.<\/jats:p>\n                <jats:sec>\n                  <jats:title>Graphical Abstract<\/jats:title>\n                  <jats:fig id=\"ufig1\" position=\"float\" orientation=\"portrait\" fig-type=\"figure\">\n                    <jats:graphic xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" xlink:href=\"26346255v1_ufig1\" position=\"float\" orientation=\"portrait\"\/>\n                  <\/jats:fig>\n                <\/jats:sec>","DOI":"10.64898\/2026.02.13.26346255","type":"posted-content","created":{"date-parts":[[2026,2,19]],"date-time":"2026-02-19T01:30:16Z","timestamp":1771464616000},"source":"Crossref","is-referenced-by-count":1,"title":["Knowledge augmented causal discovery through large language models and knowledge graphs: application in chronic low back pain"],"prefix":"10.64898","author":[{"given":"Damon","family":"Lin","sequence":"first","affiliation":[{"name":"Department of Computer Science and Software Engineering, California Polytechnic State University, San Luis Obispo, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Marzieh Mussavi","family":"Rizi","sequence":"additional","affiliation":[{"name":"School of Public Health Sciences, University of Waterloo, Waterloo, Ontario, Canada"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Conor","family":"O\u2019Neill","sequence":"additional","affiliation":[{"name":"Department of Orthopaedic Surgery, University of California San Francisco, San Francisco, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jeffrey C","family":"Lotz","sequence":"additional","affiliation":[{"name":"Department of Orthopaedic Surgery, University of California San Francisco, San Francisco, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Paul","family":"Anderson","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Software Engineering, California Polytechnic State University, San Luis Obispo, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9787-8738","authenticated-orcid":false,"given":"Abel","family":"Torres-Espin","sequence":"additional","affiliation":[{"name":"School of Public Health Sciences, University of Waterloo, Waterloo, Ontario, Canada"},{"name":"Department of Neurological Surgery, University of California San Francisco, San Francisco, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026022021550906000_2026.02.13.26346255v1.1","doi-asserted-by":"crossref","unstructured":"J. Russell Huie , Rohit Vashisht , Anoop Galivanche , Constance Hadjadj , Saam Morshed , Atul J. Butte , Adam R. Ferguson , and Conor O\u2019Neill . Toward a causal model of chronic back pain: Challenges and oppor-tunities. Frontiers in Computational Neuroscience, Volume 16-2022, 2023.","DOI":"10.3389\/fncom.2022.1017412"},{"key":"2026022021550906000_2026.02.13.26346255v1.2","first-page":"642","article-title":"Synthesizing subject-matter expertise for vari-able selection in causal effect estimation: A case study. Epidemiology (Cambridge","volume":"35","year":"2024","journal-title":"Mass"},{"key":"2026022021550906000_2026.02.13.26346255v1.3","doi-asserted-by":"crossref","first-page":"108258","DOI":"10.1016\/j.engappai.2024.108258","article-title":"A survey of causal discovery based on functional causal model","volume":"133","year":"2024","journal-title":"Engineering Applications of Artificial Intelligence"},{"key":"2026022021550906000_2026.02.13.26346255v1.4","doi-asserted-by":"crossref","unstructured":"Clark Glymour , Kun Zhang , and Peter Spirtes . Review of causal discovery methods based on graphical models. Frontiers in Genetics, 10, 2019.","DOI":"10.3389\/fgene.2019.00524"},{"key":"2026022021550906000_2026.02.13.26346255v1.5","doi-asserted-by":"crossref","first-page":"101","DOI":"10.1016\/j.ijar.2022.09.004","article-title":"A survey on causal discovery: Theory and practice","volume":"151","year":"2022","journal-title":"International Journal of Approximate Reasoning"},{"key":"2026022021550906000_2026.02.13.26346255v1.6","doi-asserted-by":"crossref","unstructured":"Peter Spirtes , Clark N Glymour , and Richard Scheines . Causation, prediction, and search. MIT press, 2000.","DOI":"10.7551\/mitpress\/1754.001.0001"},{"issue":"1","key":"2026022021550906000_2026.02.13.26346255v1.7","doi-asserted-by":"crossref","first-page":"19","DOI":"10.1038\/s44260-024-00017-9","article-title":"Triangulation for causal loop diagrams: constructing biopsychoso-cial models using group model building, literature review, and causal discovery","volume":"1","year":"2024","journal-title":"npj Complexity"},{"key":"2026022021550906000_2026.02.13.26346255v1.8","unstructured":"Aniket Vashishtha , Abbavaram Gowtham Reddy , Abhinav Kumar , Saketh Bachu , Vineeth N Balasubramanian , and Amit Sharma . Causal inference using llm-guided discovery. arXiv preprint arXiv:2310.15117, 2023."},{"key":"2026022021550906000_2026.02.13.26346255v1.9","unstructured":"Kristy Choi , Chris Cundy , Sanjari Srivastava , and Stefano Ermon . Lmpriors: Pre-trained language models as task-specific priors. arXiv preprint arXiv:2210.12530, 2022."},{"key":"2026022021550906000_2026.02.13.26346255v1.10","unstructured":"Stephanie Long , Tibor Schuster , and Alexandre Pich\u00e9 . Can large lan-guage models build causal graphs? arXiv preprint arXiv:2303.05279, 2023."},{"key":"2026022021550906000_2026.02.13.26346255v1.11","unstructured":"Emre Kiciman , Robert Ness , Amit Sharma , and Chenhao Tan . Causal reasoning and large language models: Opening a new frontier for causal-ity. Transactions on Machine Learning Research, 2023."},{"issue":"6","key":"2026022021550906000_2026.02.13.26346255v1.12","doi-asserted-by":"crossref","first-page":"1154","DOI":"10.1016\/j.spinee.2024.12.029","article-title":"Chronic low back pain causal risk factors identified by mendelian randomization: a cross-sectional cohort analysis","volume":"25","year":"2025","journal-title":"The Spine Journal"},{"key":"2026022021550906000_2026.02.13.26346255v1.13","unstructured":"Thomas Jiralerspong , Xiaoyin Chen , Yash More , Vedant Shah , and Yoshua Bengio . Efficient causal graph discovery using large language models. arXiv preprint arXiv:2402.01207, 2024."},{"key":"2026022021550906000_2026.02.13.26346255v1.14","unstructured":"Kai-Hendrik Cohrs , Gherardo Varando , Emiliano Diaz , Vasileios Sitokonstantinou , and Gustau Camps-Valls . Large language models for constrained-based causal discovery. arXiv preprint arXiv:2406.07378, 2024."},{"key":"2026022021550906000_2026.02.13.26346255v1.15","doi-asserted-by":"crossref","unstructured":"Taiyu Ban , Lyuzhou Chen , Derui Lyu , Xiangyu Wang , Qinrui Zhu , Qiang Tu , and Huanhuan Chen . Integrating large language model for improved causal discovery. IEEE Transactions on Artificial Intelligence, 2025.","DOI":"10.1109\/TAI.2025.3560927"},{"key":"2026022021550906000_2026.02.13.26346255v1.16","unstructured":"Victor-Alexandru Darvariu , Stephen Hailes , and Mirco Musolesi . Large language models are effective priors for causal graph discovery. arXiv preprint arXiv:2405.13551, 2024."},{"key":"2026022021550906000_2026.02.13.26346255v1.17","unstructured":"Elahe Khatibi , Mahyar Abbasian , Zhongqi Yang , Iman Azimi , and Amir M Rahmani . Alcm: Autonomous llm-augmented causal discov-ery framework. arXiv preprint arXiv:2405.01744, 2024."},{"key":"2026022021550906000_2026.02.13.26346255v1.18","unstructured":"Yuzhe Zhang , Yipeng Zhang , Yidong Gan , Lina Yao , and Chen Wang . Causal graph discovery with retrieval-augmented generation based large language models. arXiv preprint arXiv:2402.15301, 2024."},{"key":"2026022021550906000_2026.02.13.26346255v1.19","unstructured":"Darren Edge , Ha Trinh , Newman Cheng , Joshua Bradley , Alex Chao , Apurva Mody , Steven Truitt , Dasha Metropolitansky , Robert Osazuwa Ness , and Jonathan Larson . From local to global: A graph rag approach to query-focused summarization, 2025."},{"key":"2026022021550906000_2026.02.13.26346255v1.20","unstructured":"Trisha F Hue , Jeffrey C Lotz , Patricia Zheng , Dennis M Black , Jeannie Bailey , Susan K Ewing , Aaron J Fields , Wolf Mehling , Aaron Schef-fler , Irina Strigo , et al. Design of the comeback and backhome studies, longitudinal cohorts for comprehensive deep phenotyping of adults with chronic low-back pain (clbp): a part of the bacpac research program. medRxiv, 2024."},{"issue":"1","key":"2026022021550906000_2026.02.13.26346255v1.21","first-page":"322","article-title":"Ev-idence synthesis for constructing directed acyclic graphs (esc-dags): a novel and systematic method for building directed acyclic graphs","volume":"49","year":"2019","journal-title":"In-ternational Journal of Epidemiology"},{"key":"2026022021550906000_2026.02.13.26346255v1.22","doi-asserted-by":"publisher","DOI":"10.1007\/s11606-009-0981-1"},{"key":"2026022021550906000_2026.02.13.26346255v1.23","unstructured":"Brandon T Willard and R\u00e9mi Louf . Efficient guided generation for large language models. arXiv preprint arXiv:2307.09702, 2023."},{"key":"2026022021550906000_2026.02.13.26346255v1.24","unstructured":"Tomaz Bratanic . Implementing \u2018from local to global\u2019 graphrag with neo4j and langchain: Constructing the graph. https:\/\/medium.com\/neo4j\/implementing-from-local-to-global-graphrag-with-neo4j-and-langchain-constructing-the-graph-73924cc5bab4, July 2024. Medium."},{"issue":"4","key":"2026022021550906000_2026.02.13.26346255v1.25","doi-asserted-by":"crossref","first-page":"474","DOI":"10.26421\/JDI3.4-5","article-title":"Improved methods to aid unsupervised evidence-based fact checking for online health news","volume":"3","year":"2022","journal-title":"Journal of Data Intelligence"},{"key":"2026022021550906000_2026.02.13.26346255v1.26","doi-asserted-by":"crossref","unstructured":"V. A. Traag , L. Waltman , and N. J. van Eck . From louvain to lei-den: guaranteeing well-connected communities. Scientific Reports, 9(1), March 2019.","DOI":"10.1038\/s41598-019-41695-z"},{"key":"2026022021550906000_2026.02.13.26346255v1.27","doi-asserted-by":"crossref","unstructured":"Matthijs Douze , Alexandr Guzhva , Chengqi Deng , Jeff Johnson , Gergely Szilvasy , Pierre-Emmanuel Mazar\u00e9 , Maria Lomeli , Lucas Hosseini , and Herv\u00e9 J\u00e9gou. The faiss library. 2024.","DOI":"10.1109\/TBDATA.2025.3618474"},{"key":"2026022021550906000_2026.02.13.26346255v1.28","unstructured":"Albert Q. Jiang , Alexandre Sablayrolles , Arthur Mensch , Chris Bam-ford , Devendra Singh Chaplot , Diego de las Casas , Florian Bressand , Gi-anna Lengyel , Guillaume Lample , Lucile Saulnier , L\u00e9lio Renard Lavaud , Marie-Anne Lachaux , Pierre Stock , Teven Le Scao , Thibaut Lavril , Thomas Wang , Timoth\u00e9e Lacroix , and William El Sayed . Mistral 7b, 2023."},{"key":"2026022021550906000_2026.02.13.26346255v1.29","unstructured":"Jason Wei , Xuezhi Wang , Dale Schuurmans , Maarten Bosma , Brian Ichter , Fei Xia , Ed Chi , Quoc Le , and Denny Zhou . Chain-of-thought prompting elicits reasoning in large language models, 2023."},{"issue":"3","key":"2026022021550906000_2026.02.13.26346255v1.30","first-page":"1","article-title":"Learning bayesian networks with the bnlearn R package","volume":"35","year":"2010","journal-title":"Journal of Statistical Software"},{"key":"2026022021550906000_2026.02.13.26346255v1.31","unstructured":"Nir Friedman , Moises Goldszmidt , and Abraham Wyner . Data analysis with bayesian networks: A bootstrap approach, 2013."},{"key":"2026022021550906000_2026.02.13.26346255v1.32","doi-asserted-by":"crossref","unstructured":"Marco Scutari and Radhakrishnan Nagarajan . On identifying significant edges in graphical models of molecular networks, 2013.","DOI":"10.1016\/j.artmed.2012.12.006"},{"key":"2026022021550906000_2026.02.13.26346255v1.33","doi-asserted-by":"crossref","unstructured":"Jeffrey C. Lotz , Glen E. P. Ropella , Paul Anderson , Qian Yang , Michael A. Hedderich , Jeannie F Bailey , and C. Anthony Hunt . An exploration of knowledge-organizing technologies to advance transdisci-plinary back pain research. JOR Spine, 6, 2023.","DOI":"10.1002\/jsp2.1300"},{"key":"2026022021550906000_2026.02.13.26346255v1.34","first-page":"686","article-title":"Issls prize in clinical science 2020. examining causal effects of body mass index on back pain: a mendelian randomization study","volume":"29","year":"2019","journal-title":"European Spine Journal"},{"key":"2026022021550906000_2026.02.13.26346255v1.35","doi-asserted-by":"crossref","unstructured":"Renato Polimanti , Roseann E. Peterson , Jue-Sheng Ong , Stuart Mac-gregor , Alexis C. Edwards , Toni-Kim Clarke , Josef Frank , Zachary F. Gerring , Nathan A. Gillespie , Penelope A. Lind , Hermine H. M. Maes , Nicholas G. Martin , Hamdi Mbarek , Sarah E. Medland , Fabian Streit , Arpana Agrawal , Howard J. Edenberg , Kenneth S. Kendler , Cathryn M. Lewis , Patrick F. Sullivan , Naomi R. Wray , Joel Gelernter , and Eske M. Derks . Evidence of causal effect of major depression on alcohol depen-dence: Findings from the psychiatric genomics consortium. bioRxiv, 2018.","DOI":"10.1101\/412098"},{"key":"2026022021550906000_2026.02.13.26346255v1.36","doi-asserted-by":"crossref","unstructured":"Chen Zhu , Qihui Chen , Wei Si , Yingxiang Li , Gang Chen , and Qiran Zhao . Alcohol use and depression: A mendelian randomization study from china. Frontiers in Genetics, 11, 2020.","DOI":"10.3389\/fgene.2020.585351"},{"key":"2026022021550906000_2026.02.13.26346255v1.37","first-page":"3","article-title":"The association of cigarette smoking with depression and anx-iety: A systematic review","volume":"19","year":"2016","journal-title":"Nicotine & Tobacco Research"},{"key":"2026022021550906000_2026.02.13.26346255v1.38","doi-asserted-by":"publisher","DOI":"10.1017\/S003329172100180X"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2026.02.13.26346255","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,21]],"date-time":"2026-02-21T05:55:23Z","timestamp":1771653323000},"score":1,"resource":{"primary":{"URL":"http:\/\/medrxiv.org\/lookup\/doi\/10.64898\/2026.02.13.26346255"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,18]]},"references-count":38,"URL":"https:\/\/doi.org\/10.64898\/2026.02.13.26346255","relation":{},"subject":[],"published":{"date-parts":[[2026,2,18]]},"subtype":"preprint"}}