{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T14:04:56Z","timestamp":1790863496788,"version":"4.1.0"},"reference-count":40,"publisher":"BMJ","issue":"4","license":[{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"unspecified","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc\/4.0\/"}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["bmjph"],"accepted":{"date-parts":[[2026,9,9]]},"published-print":{"date-parts":[[2026,10]]},"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Physical activity (PA) is a critical health behaviour often disrupted by public health crises. Prior research has primarily relied on self-reported surveys, limiting timely and scalable insights. This study examines how large-scale digital exercise data can enhance real-time PA prediction and support interventions to strengthen PA resilience during global public health crises.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>We analysed PA data from 5.17\u2009million individuals across 65 countries, collected via a smartphone PA tracking app between January 2019 and June 2021. The dataset includes longitudinal exercise histories, sociodemographic characteristics, geographic locations and social network connections. We developed statistical and machine learning models\u2014including linear regressions, neural networks (NNs) and graph neural networks (GNNs)\u2014to predict weekly PA adherence (defined as at least 150\u2009min of exercise per week in line with WHO recommendations) before and during the COVID-19 pandemic. Models predicted each week\u2019s adherence using the individual\u2019s exercise history from the preceding 1\u20133\u2009weeks, together with sociodemographic, weather and pandemic-related variables. We also applied surrogate models and econometric analyses to assess the associations of individual and urban-level attributes with PA resilience globally.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>In a US county subsample where all model classes could be compared (Middlesex County, MA), the GNN achieved the highest predictive performance for weekly PA adherence, with 89.9% accuracy pre-pandemic and 82.1% during the pandemic. At the global level, the NN model demonstrated similar predictive performance (accuracy 87.6% and 82.8%). Personal exercise history was the strongest predictor, improving model accuracy by 20\u2009percentage points, while increased model complexity or the addition of social network features yielded modest gains. Marked global disparities were observed: cities in lower- and upper-middle-income countries experienced greater reductions in PA during the pandemic compared with cities in high-income countries. Females, socially connected individuals and residents in cities with better access to outdoor sports infrastructure were more likely to maintain regular PA during the pandemic.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Conclusion<\/jats:title>\n                    <jats:p>Digital PA data can enable accurate, real-time prediction of PA adherence and inform individual and environmental factors influencing global PA resilience during crises. This approach holds promise for informing targeted interventions and urban policy strategies to enhance health resilience and reduce global health disparities.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1136\/bmjph-2025-003670","type":"journal-article","created":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T13:41:37Z","timestamp":1790862097000},"page":"e003670","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":0,"title":["Global physical activity resilience during the COVID-19 pandemic: an observational study of 5.17 million exercise app users across 65 countries"],"prefix":"10.1136","volume":"4","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8400-3863","authenticated-orcid":false,"given":"Yichun","family":"Fan","sequence":"first","affiliation":[{"name":"Nicholas School of the Environment, Duke University, Durham, North Carolina, USA"},{"name":"Department of Urban Studies and Planning, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yuchen","family":"Chai","sequence":"additional","affiliation":[{"name":"Department of Urban Studies and Planning, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"F\u00e1bio","family":"Duarte","sequence":"additional","affiliation":[{"name":"Department of Urban Studies and Planning, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Stefanie","family":"Jegelka","sequence":"additional","affiliation":[{"name":"Department of Electrical Engineering & Computer Science, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Siqi","family":"Zheng","sequence":"additional","affiliation":[{"name":"Department of Urban Studies and Planning, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"239","published-online":{"date-parts":[[2026,10,1]]},"reference":[{"key":"2026100106151374000_4.4.e003670.1","doi-asserted-by":"publisher","DOI":"10.1016\/S0140-6736(12)60898-8"},{"key":"2026100106151374000_4.4.e003670.2","doi-asserted-by":"crossref","DOI":"10.1038\/s41598-022-26406-5","article-title":"Urbanization and physical activity in the global Prospective Urban and Rural Epidemiology study","volume":"13","author":"Boakye","year":"2023","journal-title":"Sci Rep"},{"key":"2026100106151374000_4.4.e003670.3","doi-asserted-by":"publisher","DOI":"10.1161\/CIRCULATIONAHA.116.024651"},{"key":"2026100106151374000_4.4.e003670.4","doi-asserted-by":"publisher","DOI":"10.1016\/S0140-6736(17)31634-3"},{"key":"2026100106151374000_4.4.e003670.5","doi-asserted-by":"publisher","DOI":"10.1038\/sj.ijo.0802821"},{"key":"2026100106151374000_4.4.e003670.6","doi-asserted-by":"publisher","DOI":"10.1001\/jama.2010.1843"},{"key":"2026100106151374000_4.4.e003670.7","doi-asserted-by":"crossref","DOI":"10.3390\/jcm8020253","article-title":"Cardiovascular Health Benefits of Exercise Training in Persons Living with Type 1 Diabetes: A Systematic Review and Meta-Analysis","volume":"8","author":"Wu","year":"2019","journal-title":"J Clin Med"},{"key":"2026100106151374000_4.4.e003670.8","doi-asserted-by":"publisher","DOI":"10.1111\/j.1464-5491.2007.01991.x"},{"key":"2026100106151374000_4.4.e003670.9","doi-asserted-by":"publisher","DOI":"10.7326\/0003-4819-144-2-200601170-00004"},{"key":"2026100106151374000_4.4.e003670.10","doi-asserted-by":"crossref","DOI":"10.1038\/s41598-020-72987-4","article-title":"Cross-sectional and prospective relationship between physical activity and depression symptoms","volume":"10","author":"Marques","year":"2020","journal-title":"Sci Rep"},{"key":"2026100106151374000_4.4.e003670.11","doi-asserted-by":"publisher","DOI":"10.1152\/japplphysiol.00977.2009"},{"key":"2026100106151374000_4.4.e003670.12","doi-asserted-by":"crossref","first-page":"054026","DOI":"10.1088\/1748-9326\/abecfd","article-title":"Health perception and commuting choice: a survey experiment measuring behavioral trade-offs between physical activity benefits and pollution exposure risks","volume":"16","author":"Fan","year":"2021","journal-title":"Environ Res Lett"},{"key":"2026100106151374000_4.4.e003670.13","doi-asserted-by":"crossref","first-page":"165","DOI":"10.1016\/j.reseneeco.2017.05.004","article-title":"Alerts work! Air quality warnings and cycling","volume":"49","author":"Saberian","year":"2017","journal-title":"Resource and Energy Economics"},{"key":"2026100106151374000_4.4.e003670.14","doi-asserted-by":"crossref","first-page":"347","DOI":"10.1038\/s41584-020-0427-z","article-title":"Combating physical inactivity during the COVID-19 pandemic","volume":"16","author":"Pinto","year":"2020","journal-title":"Nat Rev Rheumatol"},{"key":"2026100106151374000_4.4.e003670.15","doi-asserted-by":"crossref","DOI":"10.1136\/bmjgh-2021-007190","article-title":"Impact of COVID-19 pandemic lockdown on movement behaviours of children and adolescents: a systematic review","volume":"7","author":"Kharel","year":"2022","journal-title":"BMJ Glob Health"},{"key":"2026100106151374000_4.4.e003670.16","doi-asserted-by":"crossref","first-page":"2147","DOI":"10.1111\/apa.15420","article-title":"Physical activity during COVID-19 quarantine","volume":"109","author":"Shahidi","year":"2020","journal-title":"Acta Paediatr"},{"key":"2026100106151374000_4.4.e003670.17","doi-asserted-by":"crossref","DOI":"10.1073\/pnas.2016632118","article-title":"Lifestyle and mental health disruptions during COVID-19","volume":"118","author":"Giuntella","year":"2021","journal-title":"Proc Natl Acad Sci U S A"},{"key":"2026100106151374000_4.4.e003670.18","doi-asserted-by":"crossref","DOI":"10.1038\/ncomms14753","article-title":"Exercise contagion in a global social network","volume":"8","author":"Aral","year":"2017","journal-title":"Nat Commun"},{"key":"2026100106151374000_4.4.e003670.19","doi-asserted-by":"crossref","first-page":"444","DOI":"10.1089\/hs.2020.0062","article-title":"Factors Associated with Psychological Distress and Physical Activity During the COVID-19 Pandemic","volume":"18","author":"Sfendla","year":"2020","journal-title":"Health Secur"},{"key":"2026100106151374000_4.4.e003670.20","doi-asserted-by":"crossref","DOI":"10.1136\/bmjsem-2020-000960","article-title":"Changes in physical activity and sedentary behaviours from before to during the COVID-19 pandemic lockdown: a systematic review","volume":"7","author":"Stockwell","year":"2021","journal-title":"BMJ Open Sport Exerc Med"},{"key":"2026100106151374000_4.4.e003670.21","doi-asserted-by":"crossref","DOI":"10.1136\/bmjgh-2018-001044","article-title":"Objective measurement of physical activity: improving the evidence base to address non-communicable diseases in Africa","volume":"3","author":"Barr","year":"2018","journal-title":"BMJ Glob Health"},{"key":"2026100106151374000_4.4.e003670.22","doi-asserted-by":"publisher","DOI":"10.1007\/s11205-013-0359-y"},{"key":"2026100106151374000_4.4.e003670.23","doi-asserted-by":"publisher","DOI":"10.1038\/nature23018"},{"key":"2026100106151374000_4.4.e003670.24","doi-asserted-by":"crossref","first-page":"1391","DOI":"10.1249\/MSS.0000000000002602","article-title":"Factors Protecting against a Decline in Physical Activity during the COVID-19 Pandemic","volume":"53","author":"Fearnbach","year":"2021","journal-title":"Med Sci Sports Exerc"},{"key":"2026100106151374000_4.4.e003670.25","doi-asserted-by":"crossref","first-page":"835","DOI":"10.1093\/tbm\/ibaa086","article-title":"Determinants of physical activity maintenance during the Covid-19 pandemic: a focus on fitness apps","volume":"10","author":"Yang","year":"2020","journal-title":"Transl Behav Med"},{"key":"2026100106151374000_4.4.e003670.26","doi-asserted-by":"crossref","DOI":"10.3390\/jcm10143069","article-title":"Changes in Chinese Adults\u2019 Physical Activity Behavior and Determinants before and during the COVID-19 Pandemic","volume":"10","author":"Wang","year":"2021","journal-title":"J Clin Med"},{"key":"2026100106151374000_4.4.e003670.27","doi-asserted-by":"crossref","DOI":"10.1186\/s12889-022-14613-8","article-title":"Factors associated with meeting physical activity guidelines during the COVID-19 pandemic","volume":"22","author":"Heredia","year":"2022","journal-title":"BMC Public Health"},{"key":"2026100106151374000_4.4.e003670.28","doi-asserted-by":"crossref","first-page":"588","DOI":"10.1111\/bjhp.12497","article-title":"Determinants of physical activity among adults in the United Kingdom during the COVID-19 pandemic: The DUK-COVID study","volume":"26","author":"Spence","year":"2021","journal-title":"Br J Health Psychol"},{"key":"2026100106151374000_4.4.e003670.29","doi-asserted-by":"crossref","DOI":"10.7189\/jogh.12.05003","article-title":"Impact of COVID-19 on physical activity: A rapid review","volume":"12","author":"Park","year":"2022","journal-title":"J Glob Health"},{"key":"2026100106151374000_4.4.e003670.30","doi-asserted-by":"crossref","DOI":"10.1038\/s41746-019-0121-1","article-title":"Best practices for analyzing large-scale health data from wearables and smartphone apps","volume":"2","author":"Hicks","year":"2019","journal-title":"NPJ Digit Med"},{"key":"2026100106151374000_4.4.e003670.31","doi-asserted-by":"crossref","unstructured":"Gong L , Cheng Q . Exploiting edge features for graph neural networks. 2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR); Long Beach, CA, USA. doi:10.1109\/CVPR.2019.00943","DOI":"10.1109\/CVPR.2019.00943"},{"key":"2026100106151374000_4.4.e003670.32","doi-asserted-by":"crossref","first-page":"4","DOI":"10.1109\/TNNLS.2020.2978386","article-title":"A Comprehensive Survey on Graph Neural Networks","volume":"32","author":"Wu","year":"2021","journal-title":"IEEE Trans Neural Netw Learn Syst"},{"key":"2026100106151374000_4.4.e003670.33","unstructured":"Li Y , Yu R , Shahabi C , et al . Diffusion convolutional recurrent neural network: data-driven traffic forecasting. arXiv [cs.LG] [Preprint] 2017."},{"key":"2026100106151374000_4.4.e003670.34","doi-asserted-by":"publisher","DOI":"10.1016\/j.aiopen.2021.01.001"},{"key":"2026100106151374000_4.4.e003670.35","doi-asserted-by":"publisher","DOI":"10.1109\/TNN.2008.2005605"},{"key":"2026100106151374000_4.4.e003670.36","unstructured":"Lundberg S , Lee S-I . A unified approach to interpreting model predictions. arXiv [cs.AI] [Preprint] 2017."},{"key":"2026100106151374000_4.4.e003670.37","unstructured":"Porter J . 2003. Estimation in the regression discontinuity model. unpublished manuscript, department of economics, university of wisconsin at madison. 2003.5\u201319."},{"key":"2026100106151374000_4.4.e003670.38","doi-asserted-by":"publisher","DOI":"10.1257\/aer.104.9.2763"},{"key":"2026100106151374000_4.4.e003670.39","doi-asserted-by":"crossref","first-page":"53","DOI":"10.1016\/j.jdeveco.2015.11.006","article-title":"Divorce, abortion, and the child sex ratio: The impact of divorce reform in China","volume":"120","author":"Sun","year":"2016","journal-title":"J Dev Econ"},{"key":"2026100106151374000_4.4.e003670.40","doi-asserted-by":"crossref","unstructured":"Nohara Y , Matsumoto K , Soejima H , et al . Explanation of machine learning models using improved shapley additive explanation. BCB \u201919; Niagara Falls NY USA, September 4, 2019:546. doi:10.1145\/3307339.3343255 Available: https:\/\/dl.acm.org\/doi\/proceedings\/10.1145\/3307339","DOI":"10.1145\/3307339.3343255"}],"container-title":["BMJ Public Health"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/bmjph-2025-003670","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T13:41:42Z","timestamp":1790862102000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmjpublichealth.bmj.com\/lookup\/doi\/10.1136\/bmjph-2025-003670"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10]]},"references-count":40,"journal-issue":{"issue":"4","published-online":{"date-parts":[[2026,10,1]]},"published-print":{"date-parts":[[2026,10]]}},"alternative-id":["10.1136\/bmjph-2025-003670"],"URL":"https:\/\/doi.org\/10.1136\/bmjph-2025-003670","relation":{},"ISSN":["2753-4294"],"issn-type":[{"value":"2753-4294","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10]]}}}