{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T19:09:28Z","timestamp":1789758568745,"version":"4.0.1"},"update-to":[{"DOI":"10.1371\/journal.pmed.1004965","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T00:00:00Z","timestamp":1789689600000}}],"reference-count":26,"publisher":"Public Library of Science (PLoS)","issue":"9","license":[{"start":{"date-parts":[[2026,9,15]],"date-time":"2026-09-15T00:00:00Z","timestamp":1789430400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000060","name":"National Institute of Allergy and Infectious Diseases","doi-asserted-by":"publisher","award":["R01AI177326"],"award-info":[{"award-number":["R01AI177326"]}],"id":[{"id":"10.13039\/100000060","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["www.plosmedicine.org"],"crossmark-restriction":false},"short-container-title":["PLoS Med"],"abstract":"<jats:sec id=\"sec001\">\n                    <jats:title>Background<\/jats:title>\n                    <jats:p>Fluoroquinolones (FQs) are a cornerstone of most all-oral, shorter regimens endorsed by the World Health Organization for the treatment of rifampicin-resistant or multidrug-resistant tuberculosis (RR\/MDR-TB). Knowledge of resistance to FQs can help guide regimen selection at the point of care. In settings where rapid testing for FQ resistance is unavailable, prediction models could support treatment decisions by identifying FQ resistance based on patient characteristics observable at the point of care. These prediction models have been typically developed and evaluated within a single country, and their generalizability across different geographic settings is unclear.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec002\">\n                    <jats:title>Methods and findings<\/jats:title>\n                    <jats:p>\n                      We used data from 5,175 patients with RR-TB and available FQ drug susceptibility testing (DST) results submitted to the TB Portals, an open-access data-sharing platform curated by the National Institute of Allergy and Infectious Diseases, from eight countries (Azerbaijan, Belarus, Georgia, Kazakhstan, Kyrgyzstan, Moldova, Romania and Ukraine) between 2012 and 2024. Among these patients, 1,772 (34.2%) had FQ-resistant TB. We developed prediction models for FQ resistance using logistic regression, neural networks, and XGBoost. Models were evaluated under three strategies: (1)\n                      <jats:italic>pooled models<\/jats:italic>\n                      trained on multi-country data; (2)\n                      <jats:italic>within-country models<\/jats:italic>\n                      trained and evaluated using internal validation; and (3)\n                      <jats:italic>cross-country models<\/jats:italic>\n                      trained on subsets of countries and externally validated on held-out countries. Model discrimination was assessed using the area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (AUPRC). Across the three algorithms, pooled models showed moderate optimism-corrected discrimination, with AUROC ranging from 0.70 to 0.72 and AUPRC ranging from 0.57 to 0.59. Within-country models demonstrated slightly better discrimination, with AUROC and AUPRC reaching 0.8 for some countries. Cross-country external validation showed that performance loss from using a model trained on external data could be negligible to &gt;0.1 AUPROC or AUROC, depending on the country and algorithm. A limited set of predictors, including case definition and treatment-history-related variables, were among the most consistently informative predictors, whereas demographic, comorbidity, social-risk, education, and employment variables showed more variable contributions across countries and algorithms. A limitation of our study is that the incidence of RR-TB and FQ resistance was relatively stable in our analysis dataset. Hence, the results may not generalize to scenarios with marked changes in MDR-TB dynamics.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec id=\"sec003\">\n                    <jats:title>Conclusions<\/jats:title>\n                    <jats:p>Predicting FQ resistance using demographic and clinical characteristics showed moderate ability to identify FQ resistance among patients with RR-TB, but their performance and predictor patterns varied across countries. Models developed for one or several countries cannot be assumed to generalize to other settings without rigorous external validation. These findings highlight the limitations of globally trained prediction models for RR\/MDR-TB and underscore the need for locally informed prediction models to support clinical decision-making.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1371\/journal.pmed.1004965","type":"journal-article","created":{"date-parts":[[2026,9,15]],"date-time":"2026-09-15T18:02:23Z","timestamp":1789495343000},"page":"e1004965","update-policy":"https:\/\/doi.org\/10.1371\/journal.pmed.corrections_policy","source":"Crossref","is-referenced-by-count":0,"title":["Predicting resistance to fluoroquinolones among patients with rifampicin-resistant tuberculosis: A cross-country validation study"],"prefix":"10.1371","volume":"23","author":[{"ORCID":"https:\/\/orcid.org\/0009-0005-0338-9909","authenticated-orcid":true,"given":"Tianfang","family":"Shao","sequence":"first","affiliation":[{"name":"Yale School of Public Health","place":["New Haven, Connecticut, United States of America"],"department":["Department of Epidemiology of Microbial Diseases"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mariana R.","family":"Neves","sequence":"additional","affiliation":[{"name":"Philip R. Lee Institute for Health Policy Studies, University of California San Francisco","place":["San Francisco, California, United States of America"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4890-5728","authenticated-orcid":true,"given":"Molly","family":"Franke","sequence":"additional","affiliation":[{"name":"Harvard Medical School","place":["Boston, Massachusetts, United States of America"],"department":["Department of Global Health and Social Medicine"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Carole","family":"Mitnick","sequence":"additional","affiliation":[{"name":"Harvard Medical School","place":["Boston, Massachusetts, United States of America"],"department":["Department of Global Health and Social Medicine"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0825-7199","authenticated-orcid":true,"given":"Jennifer","family":"Furin","sequence":"additional","affiliation":[{"name":"Harvard Medical School","place":["Boston, Massachusetts, United States of America"],"department":["Department of Global Health and Social Medicine"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8091-7198","authenticated-orcid":true,"given":"Ted","family":"Cohen","sequence":"additional","affiliation":[{"name":"Yale School of Public Health","place":["New Haven, Connecticut, United States of America"],"department":["Department of Epidemiology of Microbial Diseases"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9276-5750","authenticated-orcid":true,"given":"Reza","family":"Yaesoubi","sequence":"additional","affiliation":[{"name":"Philip R. Lee Institute for Health Policy Studies, University of California San Francisco","place":["San Francisco, California, United States of America"]},{"name":"University of California San Francisco","place":["San Francisco, California, United States of America"],"department":["Department of Epidemiology and Biostatistics"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,9,15]]},"reference":[{"key":"pmed.1004965.ref001","unstructured":"World Health Organization. Global Tuberculosis Report 2025. 2025."},{"key":"pmed.1004965.ref002","doi-asserted-by":"crossref","first-page":"107325","DOI":"10.1016\/j.ijid.2024.107325","article-title":"Insights from the 2024 WHO Global Tuberculosis Report - more comprehensive action, innovation, and investments required for achieving WHO end TB goals","volume":"150","author":"D Goletti","year":"2025","journal-title":"Int J Infect Dis"},{"issue":"10","key":"pmed.1004965.ref003","doi-asserted-by":"crossref","first-page":"617","DOI":"10.1038\/s41579-024-01025-1","article-title":"Drug-resistant tuberculosis: a persistent global health concern","volume":"22","author":"M Farhat","year":"2024","journal-title":"Nat Rev Microbiol"},{"issue":"1","key":"pmed.1004965.ref004","doi-asserted-by":"crossref","first-page":"1225","DOI":"10.1186\/s12879-025-11566-2","article-title":"Global burden of MDR-TB and XDR-TB: trends, inequities, and future implications for public health planning","volume":"25","author":"E-L Tan","year":"2025","journal-title":"BMC Infect Dis"},{"issue":"10024","key":"pmed.1004965.ref005","doi-asserted-by":"crossref","first-page":"1211","DOI":"10.1016\/S0140-6736(15)00151-8","article-title":"Tuberculosis","volume":"387","author":"K Dheda","year":"2016","journal-title":"Lancet"},{"key":"pmed.1004965.ref006","unstructured":"World Health Organization. WHO consolidated guidelines on tuberculosis. Module 4: treatment - drug-resistant tuberculosis treatment, 2022 update. World Health Organization; 2022."},{"key":"pmed.1004965.ref007","doi-asserted-by":"crossref","first-page":"794","DOI":"10.3389\/fmicb.2019.00794","article-title":"Molecular diagnosis of drug-resistant tuberculosis; a literature review","volume":"10","author":"TNA Nguyen","year":"2019","journal-title":"Front Microbiol"},{"key":"pmed.1004965.ref008","unstructured":"World Health Organization. Use of targeted next-generation sequencing to detect drug-resistant tuberculosis. 2023."},{"issue":"6","key":"pmed.1004965.ref009","article-title":"Predicting resistance to fluoroquinolones among patients with rifampicin-resistant tuberculosis using machine learning methods","volume":"1","author":"S You","year":"2022","journal-title":"PLOS Digit Health"},{"key":"pmed.1004965.ref010","article-title":"Model for predicting drug resistance based on the clinical profile of tuberculosis patients using machine learning techniques","volume":"10","author":"IWS Falcao","year":"2024","journal-title":"PeerJ Comput Sci"},{"issue":"1","key":"pmed.1004965.ref011","doi-asserted-by":"crossref","first-page":"27","DOI":"10.1186\/s12941-020-00369-9","article-title":"Nomogram to predict multidrug-resistant tuberculosis","volume":"19","author":"S Wang","year":"2020","journal-title":"Ann Clin Microbiol Antimicrob"},{"issue":"2","key":"pmed.1004965.ref012","first-page":"345","article-title":"From development to deployment: dataset shift, causality, and shift-stable models in health AI","volume":"21","author":"A Subbaswamy","year":"2020","journal-title":"Biostatistics"},{"key":"pmed.1004965.ref013","article-title":"External validation of clinical prediction models using big datasets from e-health records or IPD meta-analysis: opportunities and challenges","volume":"353","author":"RD Riley","year":"2016","journal-title":"BMJ"},{"issue":"11","key":"pmed.1004965.ref014","doi-asserted-by":"crossref","first-page":"3267","DOI":"10.1128\/JCM.01013-17","article-title":"The TB portals: an open-access, web-based platform for global drug-resistant-tuberculosis data sharing and analysis","volume":"55","author":"A Rosenthal","year":"2017","journal-title":"J Clin Microbiol"},{"issue":"9","key":"pmed.1004965.ref015","first-page":"1322","article-title":"Missing data in clinical research: a tutorial on multiple imputation","volume":"37","author":"PC Austin","year":"2021","journal-title":"Can J Cardiol"},{"key":"pmed.1004965.ref016","article-title":"TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods","volume":"385","year":"2024","journal-title":"BMJ"},{"key":"pmed.1004965.ref017","doi-asserted-by":"crossref","unstructured":"Chen T, Guestrin C. XGBoost: a scalable tree boosting system. In: Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, San Francisco, California, USA, 2016. p. 785\u201394.","DOI":"10.1145\/2939672.2939785"},{"key":"pmed.1004965.ref018","doi-asserted-by":"crossref","DOI":"10.1136\/bmj-2023-078378","article-title":"TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods","volume":"385","author":"GS Collins","year":"2024","journal-title":"BMJ"},{"issue":"1","key":"pmed.1004965.ref019","article-title":"Transparent Reporting of a multivariable prediction model for Individual Prognosis or Diagnosis (TRIPOD): explanation and elaboration","volume":"162","author":"KGM Moons","year":"2015","journal-title":"Ann Intern Med"},{"issue":"11","key":"pmed.1004965.ref020","doi-asserted-by":"crossref","DOI":"10.1111\/cts.70056","article-title":"Practical guide to SHAP analysis: explaining supervised machine learning model predictions in drug development","volume":"17","author":"AV Ponce-Bobadilla","year":"2024","journal-title":"Clin Transl Sci"},{"key":"pmed.1004965.ref021","first-page":"2825","article-title":"Scikit-learn: machine learning in Python","volume":"12","author":"F Pedregosa","year":"2011","journal-title":"J Mach Learn Res"},{"key":"pmed.1004965.ref022","author":"SM Lundberg"},{"issue":"1","key":"pmed.1004965.ref023","doi-asserted-by":"crossref","first-page":"1391","DOI":"10.1186\/s12879-024-10282-7","article-title":"Machine learning-based prediction of antibiotic resistance in Mycobacterium tuberculosis clinical isolates from Uganda","volume":"24","author":"SR Babirye","year":"2024","journal-title":"BMC Infect Dis"},{"issue":"9","key":"pmed.1004965.ref024","doi-asserted-by":"crossref","first-page":"2768","DOI":"10.1177\/0962280218785504","article-title":"A framework for meta-analysis of prediction model studies with binary and time-to-event outcomes","volume":"28","author":"TP Debray","year":"2019","journal-title":"Stat Methods Med Res"},{"issue":"1","key":"pmed.1004965.ref025","first-page":"230","article-title":"Topic Group \u2018Evaluating diagnostic tests and prediction models\u2019 of the STRATOS initiative. Calibration: the Achilles heel of predictive analytics","volume":"17","author":"B Van Calster","year":"2019","journal-title":"BMC Med"},{"key":"pmed.1004965.ref026","article-title":"We are underprepared for bedaquiline resistance: a call for clinical and programmatic readiness","author":"GB Cross","year":"2026","journal-title":"Clin Infect Dis"}],"updated-by":[{"DOI":"10.1371\/journal.pmed.1004965","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T00:00:00Z","timestamp":1789689600000}}],"container-title":["PLOS Medicine"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pmed.1004965","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,18]],"date-time":"2026-09-18T18:06:05Z","timestamp":1789754765000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pmed.1004965"}},"subtitle":[],"editor":[{"given":"Amitabh Bipin","family":"Suthar","sequence":"first","affiliation":[{"name":"PLOS Medicine Editorial Board","place":["UNITED STATES OF AMERICA"]}],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,9,15]]},"references-count":26,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2026,9,15]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pmed.1004965","relation":{},"ISSN":["1549-1676"],"issn-type":[{"value":"1549-1676","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,15]]}}}