{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:43:59Z","timestamp":1785545039834,"version":"3.56.0"},"reference-count":53,"publisher":"JMIR Publications Inc.","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["JMIR Form Res"],"abstract":"Abstract<\/jats:title>\n \n Background<\/jats:title>\n Patients with rheumatoid arthritis (RA) face higher risks of major adverse cardiovascular events than the general population. While machine learning offers powerful predictive capabilities, its clinical adoption is hindered by the \u201cblack-box\u201d nature of complex algorithms.<\/jats:p>\n <\/jats:sec>\n \n Objective<\/jats:title>\n This study aimed to develop interpretable survival models to predict the risk of major adverse cardiovascular events in patients with RA, providing transparent and actionable insights for personalized clinical prognosis management.<\/jats:p>\n <\/jats:sec>\n \n Methods<\/jats:title>\n Using data from the Taipei Medical University Clinical Research Database (2011-2022) for 2461 patients with RA, machine learning survival models, including random survival forest (RSF), DeepSurv, and Cox-Time, were compared with the traditional Cox proportional hazards model. Performance was evaluated using the C-index and integrated Brier score. Permutation importance and Shapley additive explanations (SHAP) analyses were integrated to provide explainability for individual-level risk predictions.<\/jats:p>\n <\/jats:sec>\n \n Results<\/jats:title>\n RSF demonstrated superior performance, achieving a C-index of 0.8771 and an integrated Brier score of 0.0775. Permutation importance identified key features, including creatinine, conventional synthetic disease-modifying antirheumatic drugs, C-reactive protein, alanine aminotransferase, and age at RA diagnosis. SHAP analysis further quantified feature-specific effects, revealing both protective and risk-increasing associations between medications and laboratory indicators.<\/jats:p>\n <\/jats:sec>\n \n Conclusions<\/jats:title>\n RSF outperformed traditional methods, and integrating SHAP enabled transparent, personalized risk interpretation, translating complex models into actionable insights for clinicians. This approach empowers clinicians to identify high-risk individuals and advances precision medicine in rheumatology. Future work should include temporal validation using data from later years and external validation using datasets from other health care systems to further assess model generalizability.<\/jats:p>\n <\/jats:sec>","DOI":"10.2196\/91790","type":"journal-article","created":{"date-parts":[[2026,6,5]],"date-time":"2026-06-05T17:43:12Z","timestamp":1780681392000},"page":"e91790-e91790","source":"Crossref","is-referenced-by-count":0,"title":["Interpretable Machine Learning Framework for Predicting Major Adverse Cardiovascular Events in Rheumatoid Arthritis Using Electronic Health Records: Multicenter Cohort Study"],"prefix":"10.2196","volume":"10","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-1022-3131","authenticated-orcid":false,"given":"Wei-Chen","family":"Chiang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7605-5556","authenticated-orcid":false,"given":"Guan-Ling","family":"Lin","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2622-7524","authenticated-orcid":false,"given":"Yu-Sheng","family":"Chang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4003-5393","authenticated-orcid":false,"given":"Yu-Chen","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2551-6199","authenticated-orcid":false,"given":"Chih-Wei","family":"Huang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7436-9041","authenticated-orcid":false,"given":"Phung Anh","family":"Nguyen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2997-2404","authenticated-orcid":false,"given":"Jason C","family":"Hsu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0649-6281","authenticated-orcid":false,"given":"Der-Ming","family":"Liou","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9198-0697","authenticated-orcid":false,"given":"Hsuan-Chia","family":"Yang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1010","published-online":{"date-parts":[[2026,6,5]]},"reference":[{"issue":"10416","key":"R1","doi-asserted-by":"publisher","first-page":"2019","DOI":"10.1016\/S0140-6736(23)01525-8","article-title":"Rheumatoid arthritis","volume":"402","author":"Di Matteo","journal-title":"Lancet"},{"issue":"1","key":"R2","doi-asserted-by":"publisher","first-page":"102925","DOI":"10.1016\/j.autrev.2021.102925","article-title":"Systematic review and meta-analysis of cardiovascular risk in rheumatological disease: symptomatic and non-symptomatic events in rheumatoid arthritis and systemic lupus erythematosus","volume":"21","author":"Restivo","journal-title":"Autoimmun Rev"},{"issue":"Suppl 3","key":"R3","doi-asserted-by":"publisher","first-page":"314","DOI":"10.1007\/s00393-024-01538-3","article-title":"All-cause and cause-specific mortality in rheumatoid arthritis: a meta-analysis","volume":"83","author":"Lee","journal-title":"Z Rheumatol"},{"issue":"2","key":"R4","doi-asserted-by":"publisher","DOI":"10.1136\/rmdopen-2018-000771","article-title":"Performance of the Expanded Cardiovascular Risk Prediction Score for Rheumatoid Arthritis in a geographically distant national register-based cohort: an external validation","volume":"4","author":"Ljung","journal-title":"RMD Open"},{"issue":"7","key":"R5","doi-asserted-by":"publisher","first-page":"587","DOI":"10.1002\/acr2.11432","article-title":"Time-varying association of rheumatoid arthritis disease activity to subsequent cardiovascular risk","volume":"4","author":"Yoshida","journal-title":"ACR Open Rheumatol"},{"issue":"5","key":"R6","doi-asserted-by":"publisher","first-page":"631","DOI":"10.1002\/acr.25481","article-title":"Risk of incident heart failure and heart failure subtypes in patients with rheumatoid arthritis","volume":"77","author":"Kawano","journal-title":"Arthritis Care Res (Hoboken)"},{"issue":"12","key":"R7","doi-asserted-by":"publisher","DOI":"10.1111\/1756-185X.15448","article-title":"Risk factors associated with major adverse cardiovascular events and malignancies in patients with rheumatoid arthritis in a real-world setting in Japan","volume":"27","author":"Yamaoka","journal-title":"Int J Rheum Dis"},{"key":"R8","doi-asserted-by":"publisher","DOI":"10.1016\/j.semarthrit.2024.152416","article-title":"Incidence and risk factors of major cardiovascular events in rheumatoid arthritis and psoriatic arthritis: a population-based cohort study","volume":"65","author":"Meng","journal-title":"Semin Arthritis Rheum"},{"issue":"22","key":"R9","doi-asserted-by":"publisher","DOI":"10.3390\/jcm11226812","article-title":"Associations between cardiovascular outcomes and rheumatoid arthritis: a nationwide population-based cohort study","volume":"11","author":"Kang","journal-title":"J Clin Med"},{"issue":"1","key":"R10","doi-asserted-by":"publisher","DOI":"10.1186\/s12944-025-02594-7","article-title":"Association of weight-adjusted waist index with all-cause and cardiovascular disease mortality among rheumatoid arthritis population: a cohort study from the NHANES 1999-2018","volume":"24","author":"Liu","journal-title":"Lipids Health Dis"},{"issue":"6","key":"R11","doi-asserted-by":"publisher","first-page":"1069","DOI":"10.1177\/09622802241242325","article-title":"Methods for non-proportional hazards in clinical trials: a systematic review","volume":"33","author":"Bardo","journal-title":"Stat Methods Med Res"},{"issue":"8","key":"R12","doi-asserted-by":"publisher","first-page":"2368","DOI":"10.1177\/0962280218779408","article-title":"Flexible and structured survival model for a simultaneous estimation of non-linear and non-proportional effects and complex interactions between continuous variables: performance of this multidimensional penalized spline approach in net survival trend analysis","volume":"28","author":"Remontet","journal-title":"Stat Methods Med Res"},{"issue":"1","key":"R13","doi-asserted-by":"publisher","first-page":"25","DOI":"10.1146\/annurev-statistics-032921-022127","article-title":"High-dimensional survival analysis: methods and applications","volume":"10","author":"Salerno","journal-title":"Annu Rev Stat Appl"},{"issue":"1","key":"R14","doi-asserted-by":"publisher","DOI":"10.4172\/2161-1165.1000216","article-title":"The effect of ignoring statistical interactions in regression analyses conducted in epidemiologic studies: an example with survival analysis using Cox proportional hazards regression model","volume":"6","author":"Vatcheva","journal-title":"Epidemiology (Sunnyvale)"},{"issue":"22","key":"R15","doi-asserted-by":"publisher","DOI":"10.1002\/cam4.71375","article-title":"A systematic review on machine learning techniques for survival analysis in cancer","volume":"14","author":"O\u2019Donnell","journal-title":"Cancer Med"},{"issue":"1","key":"R16","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-025-26714-6","article-title":"Global performance of machine learning models to predict all-cause mortality: systematic review and meta-analysis","volume":"15","author":"Delpino","journal-title":"Sci Rep"},{"issue":"1","key":"R17","doi-asserted-by":"publisher","DOI":"10.1186\/s12874-023-02078-1","article-title":"Application of machine learning in predicting survival outcomes involving real-world data: a scoping review","volume":"23","author":"Huang","journal-title":"BMC Med Res Methodol"},{"issue":"1","key":"R18","doi-asserted-by":"publisher","first-page":"31","DOI":"10.17849\/insm-47-01-31-39.1","article-title":"Random forest","volume":"47","author":"Rigatti","journal-title":"J Insur Med"},{"issue":"1","key":"R19","doi-asserted-by":"publisher","DOI":"10.1186\/s12874-018-0482-1","article-title":"DeepSurv: personalized treatment recommender system using a Cox proportional hazards deep neural network","volume":"18","author":"Katzman","journal-title":"BMC Med Res Methodol"},{"key":"R20","volume":"20","author":"Kvamme","journal-title":"J Mach Learn Res"},{"key":"R21","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-023-10681-3","article-title":"Deep learning for survival analysis: a review","volume":"57","author":"Wiegrebe","journal-title":"Artif Intell Rev"},{"issue":"1","key":"R22","doi-asserted-by":"publisher","DOI":"10.1186\/s12874-025-02694-z","article-title":"Comparison of machine learning methods versus traditional Cox regression for survival prediction in cancer using real-world data: a systematic literature review and meta-analysis","volume":"25","author":"Huang","journal-title":"BMC Med Res Methodol"},{"key":"R23","unstructured":"Lundberg S Lee SI . A unified approach to interpreting model predictions. arXiv. Preprint posted online on May 22, 2017. [doi: 10.48550\/arXiv.1705.07874]"},{"issue":"22","key":"R24","doi-asserted-by":"publisher","DOI":"10.3390\/cancers14225562","article-title":"Development and validation of novel deep-learning models using multiple data types for lung cancer survival","volume":"14","author":"Hsu","journal-title":"Cancers (Basel)"},{"issue":"13","key":"R25","doi-asserted-by":"publisher","first-page":"1360","DOI":"10.1001\/jama.2018.13103","article-title":"Diagnosis and management of rheumatoid arthritis: a review","volume":"320","author":"Aletaha","journal-title":"JAMA"},{"issue":"10086","key":"R26","doi-asserted-by":"publisher","first-page":"2338","DOI":"10.1016\/S0140-6736(17)31491-5","article-title":"Novel treatment strategies in rheumatoid arthritis","volume":"389","author":"Burmester","journal-title":"Lancet"},{"key":"R27","doi-asserted-by":"publisher","DOI":"10.1016\/j.jclinepi.2022.08.016","article-title":"Handling missing data in clinical research","volume":"151","author":"Heymans","journal-title":"J Clin Epidemiol"},{"issue":"1","key":"R28","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-025-09975-z","article-title":"A robust machine learning approach to predicting remission and stratifying risk in rheumatoid arthritis patients treated with bDMARDs","volume":"15","author":"Salehi","journal-title":"Sci Rep"},{"key":"R29","volume":"162","author":"Jarrett","journal-title":"Proc Mach Learn Res"},{"issue":"3","key":"R30","doi-asserted-by":"publisher","first-page":"841","DOI":"10.1214\/08-AOAS169","article-title":"Random survival forests","volume":"2","author":"Ishwaran","journal-title":"Ann Appl Stat"},{"key":"R31","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijmedinf.2021.104635","article-title":"Comparison of time-to-event machine learning models in predicting oral cavity cancer prognosis","volume":"157","author":"Adeoye","journal-title":"Int J Med Inform"},{"issue":"1","key":"R32","doi-asserted-by":"publisher","DOI":"10.1186\/s12911-023-02293-2","article-title":"A comparison of random survival forest and Cox regression for prediction of mortality in patients with hemorrhagic stroke","volume":"23","author":"Wang","journal-title":"BMC Med Inform Decis Mak"},{"key":"R33","doi-asserted-by":"publisher","DOI":"10.2196\/44417","article-title":"Predicting colorectal cancer survival using time-to-event machine learning: retrospective cohort study","volume":"25","author":"Yang","journal-title":"J Med Internet Res"},{"issue":"4","key":"R34","doi-asserted-by":"publisher","first-page":"531","DOI":"10.1002\/sam.11583","article-title":"Optimal ratio for data splitting","volume":"15","author":"Joseph","journal-title":"Stat Anal Data Min"},{"issue":"17-18","key":"R35","doi-asserted-by":"publisher","first-page":"2529","DOI":"10.1002\/(sici)1097-0258(19990915\/30)18:17\/183.0.co;2-5","article-title":"Assessment and comparison of prognostic classification schemes for survival data","volume":"18","author":"Graf","journal-title":"Stat Med"},{"issue":"9","key":"R36","doi-asserted-by":"publisher","first-page":"1567","DOI":"10.1093\/jamia\/ocac098","article-title":"Development and validation of a deep learning model to predict the survival of patients in ICU","volume":"29","author":"Tang","journal-title":"J Am Med Inform Assoc"},{"issue":"10","key":"R37","doi-asserted-by":"publisher","first-page":"1340","DOI":"10.1093\/bioinformatics\/btq134","article-title":"Permutation importance: a corrected feature importance measure","volume":"26","author":"Altmann","journal-title":"Bioinformatics"},{"key":"R38","unstructured":"Molnar C . Interpretable Machine Learning. Lulu.com; 2020. ISBN: 9780244768522"},{"issue":"2","key":"R39","doi-asserted-by":"publisher","first-page":"201","DOI":"10.1136\/ard-2024-226156","article-title":"Disease activity of rheumatoid arthritis and kidney function decline: a large prospective registry study","volume":"84","author":"Fukui","journal-title":"Ann Rheum Dis"},{"issue":"3","key":"R40","doi-asserted-by":"publisher","first-page":"161","DOI":"10.1080\/25785826.2025.2460267","article-title":"Chronic kidney disease and inflammatory cytokines in rheumatoid arthritis: a potential pathogenic link","volume":"48","author":"Hanaoka","journal-title":"Immunol Med"},{"issue":"3","key":"R41","doi-asserted-by":"publisher","first-page":"619","DOI":"10.1016\/j.cgh.2021.11.038","article-title":"An observational data meta-analysis on the differences in prevalence and risk factors between MAFLD vs NAFLD","volume":"21","author":"Lim","journal-title":"Clin Gastroenterol Hepatol"},{"issue":"4","key":"R42","doi-asserted-by":"publisher","first-page":"1135","DOI":"10.1007\/s12072-024-10706-1","article-title":"NAFLD and MAFLD independently increase the risk of major adverse cardiovascular events (MACE): a 20-year longitudinal follow-up study from regional Australia","volume":"18","author":"Vaz","journal-title":"Hepatol Int"},{"key":"R43","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejim.2023.08.002","article-title":"Clinical features and immune mechanisms directly linked to the altered liver function in patients with rheumatoid arthritis","volume":"118","author":"Arias-de la Rosa","journal-title":"Eur J Intern Med"},{"issue":"4","key":"R44","doi-asserted-by":"publisher","first-page":"1485","DOI":"10.1007\/s10067-025-07364-5","article-title":"Metabolic dysfunction-associated steatotic liver disease and cardiovascular risk factors in rheumatoid arthritis","volume":"44","author":"Saidi","journal-title":"Clin Rheumatol"},{"key":"R45","doi-asserted-by":"publisher","DOI":"10.1016\/j.ejim.2022.07.001","article-title":"C-reactive protein and 10-year cardiovascular risk in rheumatoid arthritis","volume":"104","author":"Erre","journal-title":"Eur J Intern Med"},{"issue":"1","key":"R46","doi-asserted-by":"publisher","first-page":"e58","DOI":"10.1016\/S2665-9913(20)30221-6","article-title":"Cardiovascular risk in inflammatory arthritis: rheumatoid arthritis and gout","volume":"3","author":"Hansildaar","journal-title":"Lancet Rheumatol"},{"issue":"3","key":"R47","doi-asserted-by":"publisher","DOI":"10.1136\/rmdopen-2025-005905","article-title":"Data-driven identification of subgroups in early rheumatoid arthritis: mortality and cardiovascular disease in a cohort from Western Norway","volume":"11","author":"Alsing","journal-title":"RMD Open"},{"issue":"4","key":"R48","doi-asserted-by":"publisher","DOI":"10.1136\/rmdopen-2023-003489","article-title":"Risk of major adverse cardiovascular events in patients with rheumatoid arthritis treated with conventional synthetic, biologic and targeted synthetic disease-modifying antirheumatic drugs: observational data from the German RABBIT register","volume":"9","author":"Meissner","journal-title":"RMD Open"},{"issue":"5","key":"R49","doi-asserted-by":"publisher","first-page":"648","DOI":"10.3899\/jrheum.200265","article-title":"The risk of cardiovascular events associated with disease-modifying antirheumatic drugs in rheumatoid arthritis","volume":"48","author":"Ozen","journal-title":"J Rheumatol"},{"issue":"9","key":"R50","doi-asserted-by":"publisher","first-page":"3171","DOI":"10.26355\/eurrev_202205_28734","article-title":"The effect of statin on major adverse cardiovascular events and mortality in patients with rheumatoid arthritis - a systematic review and meta-analysis","volume":"26","author":"Rahmadi","journal-title":"Eur Rev Med Pharmacol Sci"},{"issue":"9","key":"R51","doi-asserted-by":"publisher","first-page":"1437","DOI":"10.1002\/art.40892","article-title":"A multicenter, randomized, placebo-controlled trial of atorvastatin for the primary prevention of cardiovascular events in patients with rheumatoid arthritis","volume":"71","author":"Kitas","journal-title":"Arthritis Rheumatol"},{"issue":"12","key":"R52","doi-asserted-by":"publisher","first-page":"1522","DOI":"10.1136\/annrheumdis-2021-220577","article-title":"Short-term dose and duration-dependent glucocorticoid risk for cardiovascular events in glucocorticoid-naive patients with rheumatoid arthritis","volume":"80","author":"Ocon","journal-title":"Ann Rheum Dis"},{"issue":"11","key":"R53","doi-asserted-by":"publisher","first-page":"1585","DOI":"10.1002\/art.42928","article-title":"Association of cardiovascular outcomes with low-dose glucocorticoid prescription in patients with rheumatoid arthritis","volume":"76","author":"Coburn","journal-title":"Arthritis Rheumatol"}],"container-title":["JMIR Formative Research"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/formative.jmir.org\/2026\/1\/e91790","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,31]],"date-time":"2026-07-31T22:24:05Z","timestamp":1785536645000},"score":1,"resource":{"primary":{"URL":"https:\/\/formative.jmir.org\/2026\/1\/e91790"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,5]]},"references-count":53,"URL":"https:\/\/doi.org\/10.2196\/91790","relation":{},"ISSN":["2561-326X"],"issn-type":[{"value":"2561-326X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,5]]},"article-number":"v10i10416e91790"}}