{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T15:01:58Z","timestamp":1785164518283,"version":"3.55.0"},"reference-count":36,"publisher":"BMJ","issue":"7","license":[{"start":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T00:00:00Z","timestamp":1785110400000},"content-version":"unspecified","delay-in-days":26,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100000925","name":"National Health and Medical Research Council","doi-asserted-by":"publisher","award":["2026765"],"award-info":[{"award-number":["2026765"]}],"id":[{"id":"10.13039\/501100000925","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Australian Medical Research Futures Fund","award":["2036232"],"award-info":[{"award-number":["2036232"]}]}],"content-domain":{"domain":["bmj.com"],"crossmark-restriction":true},"short-container-title":["BMJ Open"],"accepted":{"date-parts":[[2026,7,13]]},"published-print":{"date-parts":[[2026,7]]},"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Cardiovascular disease (CVD) remains a leading cause of preventable morbidity and mortality. Most existing CVD prediction tools use a baseline survival modelling framework, in which each predictor is represented by a single value recorded at the start of follow-up. Although risk calculators can be re-run using updated values, these models do not explicitly use the longitudinal, irregularly recorded information in primary-care electronic medical records (EMRs). Dynamic modelling approaches have the potential to generate updated CVD risk estimates at each general practitioner (GP) encounter, supporting both risk assessment and monitoring of changes over time. This protocol outlines the development and validation of transformer-based sequence models for encounter-level prediction of 5-year CVD risk, compared with dynamic landmark survival models and benchmarked against static CVD prediction tools.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods and analysis<\/jats:title>\n                    <jats:p>The study will use the New South Wales (NSW) Lumos linked health data asset, a large, privacy-preserving linkage of GP EMRs with hospital admissions, mortality and other administrative datasets across NSW, Australia. Eligible GP encounters will be included from 2018 onwards. The primary outcome is first fatal or non-fatal CVD event within 5 years of each eligible encounter, identified using ICD-10-AM codes in linked hospital and mortality data. Predictor variables include routinely collected demographics, chronic conditions, clinical measurements and medications, as well as healthcare utilisation patterns available at or before each encounter. Sex-specific dynamic landmark survival models (such as Cox proportional hazards models) with time-updated covariates (full and least absolute shrinkage and selection operator (LASSO)-regularised) will be developed alongside sex-specific transformer-based sequence models, fine-tuned for 5-year risk prediction at each encounter. Internal validation will use repeated resampling for landmark survival models with patient-level and temporal separation for sequence models. Geographic transportability will be evaluated using internal\u2013external validation across primary health networks (PHNs). Performance will be evaluated using discrimination, calibration, decision-curve analysis and subgroup analyses and will be compared with existing static predictive models. Missing-data will be imputed where necessary: a range of imputation approaches will be compared to optimise computational efficiency but also support predictive generalisability.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Ethics and dissemination<\/jats:title>\n                    <jats:p>This study is conducted under NSW Health governance arrangements with approval from the NSW Population and Health Services Research Ethics Committee (2019\/ETH00660). Analyses will be performed within a secure data environment using de-identified linked data. Results will be reported in accordance with TRIPOD-AI guidance and disseminated through peer-reviewed publications, scientific conferences and policy and consumer forums.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1136\/bmjopen-2026-122980","type":"journal-article","created":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T14:35:32Z","timestamp":1785162932000},"page":"e122980","update-policy":"https:\/\/doi.org\/10.1136\/crossmarkpolicy","source":"Crossref","is-referenced-by-count":0,"title":["Dynamic prediction of cardiovascular risk from linked primary care records in New South Wales, Australia: protocol for a retrospective prognostic modelling study"],"prefix":"10.1136","volume":"16","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8749-7280","authenticated-orcid":false,"given":"Nicholas I-Hsien","family":"Kuo","sequence":"first","affiliation":[{"name":"Centre for Big Data Research in Health, University of New South Wales, Sydney, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2220-3664","authenticated-orcid":false,"given":"Heidi","family":"Welberry","sequence":"additional","affiliation":[{"name":"Centre for Big Data Research in Health, University of New South Wales, Sydney, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9555-0261","authenticated-orcid":false,"given":"Sanja","family":"Lujic","sequence":"additional","affiliation":[{"name":"Centre for Big Data Research in Health, University of New South Wales, Sydney, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1057-0085","authenticated-orcid":false,"given":"Anna","family":"Campain","sequence":"additional","affiliation":[{"name":"The George Institute for Global Health, University of New South Wales, Sydney, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daniel","family":"Wilson","sequence":"additional","affiliation":[{"name":"NSW Ministry of Health, Sydney, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Katie","family":"Harris","sequence":"additional","affiliation":[{"name":"The George Institute for Global Health, University of New South Wales, Sydney, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Michael","family":"Kidd","sequence":"additional","affiliation":[{"name":"International Centre for Future Health Systems, University of New South Wales, Sydney, Australia"},{"name":"Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford, UK"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Patricia","family":"Correll","sequence":"additional","affiliation":[{"name":"NSW Ministry of Health, Sydney, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6898-3870","authenticated-orcid":false,"given":"David","family":"Peiris","sequence":"additional","affiliation":[{"name":"The George Institute for Global Health, University of New South Wales, Sydney, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0390-661X","authenticated-orcid":false,"given":"Louisa","family":"Jorm","sequence":"additional","affiliation":[{"name":"Centre for Big Data Research in Health, University of New South Wales, Sydney, Australia"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"239","published-online":{"date-parts":[[2026,7,27]]},"reference":[{"key":"2026072707352403000_16.7.e122980.1","unstructured":"National Heart Foundation of Australia . Key statistics: cardiovascular disease. 2025. Available: https:\/\/www.heartfoundation.org.au\/your-heart\/evidence-and-statistics\/key-stats-cardiovascular-disease"},{"key":"2026072707352403000_16.7.e122980.2","unstructured":"Australian Institute of Health and Welfare . Heart, stroke and vascular diseases: Australian facts \u2013 disease types. 2025. Available: https:\/\/www.aihw.gov.au\/reports\/heart-stroke-vascular-diseases\/hsvd-facts\/contents\/disease-types"},{"key":"2026072707352403000_16.7.e122980.3","doi-asserted-by":"publisher","DOI":"10.5694\/mja2.52280"},{"key":"2026072707352403000_16.7.e122980.4","unstructured":"National Heart Foundation of Australia . Guideline for managing cardiovascular disease. 2025. Available: https:\/\/www.heartfoundation.org.au\/for-professionals\/guideline-for-managing-cvd"},{"key":"2026072707352403000_16.7.e122980.5","doi-asserted-by":"publisher","DOI":"10.1016\/j.jacc.2013.11.005"},{"key":"2026072707352403000_16.7.e122980.6","doi-asserted-by":"publisher","DOI":"10.1016\/S0140-6736(18)30664-0"},{"key":"2026072707352403000_16.7.e122980.7","doi-asserted-by":"publisher","DOI":"10.1038\/s41591-024-02905-y"},{"key":"2026072707352403000_16.7.e122980.8","doi-asserted-by":"publisher","DOI":"10.1161\/CIRCULATIONAHA.123.067626"},{"key":"2026072707352403000_16.7.e122980.9","doi-asserted-by":"publisher","DOI":"10.5694\/mja2.52052"},{"key":"2026072707352403000_16.7.e122980.10","doi-asserted-by":"crossref","first-page":"263","DOI":"10.37765\/ajmc.2024.89509","article-title":"Primary care visit cadence and hospital admissions in high-risk patients","volume":"30","author":"Matsil","year":"2024","journal-title":"Am J Manag Care"},{"key":"2026072707352403000_16.7.e122980.11","doi-asserted-by":"crossref","DOI":"10.1186\/s41512-025-00199-3","article-title":"A comparison of modeling approaches for static and dynamic prediction of central-line bloodstream infections using electronic health records (part 1): regression models","volume":"9","author":"Gao","year":"2025","journal-title":"Diagn Progn Res"},{"key":"2026072707352403000_16.7.e122980.12","doi-asserted-by":"crossref","DOI":"10.1093\/gigascience\/giaf107","article-title":"Foundation model of electronic medical records for adaptive risk estimation","volume":"14","author":"Renc","year":"2025","journal-title":"Gigascience"},{"key":"2026072707352403000_16.7.e122980.13","doi-asserted-by":"publisher","DOI":"10.1093\/jamia\/ocw042"},{"key":"2026072707352403000_16.7.e122980.14","doi-asserted-by":"publisher","DOI":"10.1016\/j.jbi.2021.103980"},{"key":"2026072707352403000_16.7.e122980.15","doi-asserted-by":"publisher","DOI":"10.1093\/aje\/kwy018"},{"key":"2026072707352403000_16.7.e122980.16","doi-asserted-by":"publisher","DOI":"10.1093\/eurheartj\/ehw302"},{"key":"2026072707352403000_16.7.e122980.17","doi-asserted-by":"crossref","first-page":"30","DOI":"10.1093\/ehjdh\/ztad058","article-title":"Improving cardiovascular risk prediction through machine learning modelling of irregularly repeated electronic health records","volume":"5","author":"Li","year":"2024","journal-title":"Eur Heart J Digit Health"},{"key":"2026072707352403000_16.7.e122980.18","first-page":"310","article-title":"Machine-learning versus traditional approaches for atherosclerotic cardiovascular risk prognostication in primary prevention cohorts: a systematic review and meta-analysis","volume":"9","author":"Liu","year":"2023","journal-title":"Eur Heart J Qual Care Clin Outcomes"},{"key":"2026072707352403000_16.7.e122980.19","doi-asserted-by":"crossref","DOI":"10.1136\/ihj-2021-000074","article-title":"Lumos: a statewide linkage programme in Australia integrating general practice data to guide system redesign","volume":"3","author":"Correll","year":"2021","journal-title":"Integr Healthc J"},{"key":"2026072707352403000_16.7.e122980.20","doi-asserted-by":"crossref","DOI":"10.1136\/bmjopen-2025-102055","article-title":"An assessment of data quality and sociodemographic variation in health service utilisation of general practice, emergency department and admitted services in a New South Wales linked health data asset: a retrospective cohort study of Lumos","volume":"15","author":"Bouckley","year":"2025","journal-title":"BMJ Open"},{"key":"2026072707352403000_16.7.e122980.21","doi-asserted-by":"publisher","DOI":"10.1136\/heartjnl-2025-325776"},{"key":"2026072707352403000_16.7.e122980.22","doi-asserted-by":"crossref","first-page":"2101","DOI":"10.1002\/sim.9713","article-title":"A comparison of two approaches to dynamic prediction: Joint modeling and landmark modeling","volume":"42","author":"Li","year":"2023","journal-title":"Stat Med"},{"key":"2026072707352403000_16.7.e122980.23","unstructured":"Vaswani A , Shazeer N , Parmar N , et al . Attention is all you need. Adv Neural Inf Process Syst 2017."},{"key":"2026072707352403000_16.7.e122980.24","unstructured":"Devlin J , Chang MW , Lee K , et al . BERT: pre-training of deep bidirectional transformers for language understanding. Proceedings of NAACL-HLT; 2019:4171\u201386."},{"key":"2026072707352403000_16.7.e122980.25","doi-asserted-by":"crossref","first-page":"7155","DOI":"10.1038\/s41598-020-62922-y","article-title":"BEHRT: Transformer for Electronic Health Records","volume":"10","author":"Li","year":"2020","journal-title":"Sci Rep"},{"key":"2026072707352403000_16.7.e122980.26","doi-asserted-by":"crossref","first-page":"100873","DOI":"10.1016\/j.landig.2025.03.005","article-title":"Refined selection of individuals for preventive cardiovascular disease treatment with a transformer-based risk model","volume":"7","author":"Rao","year":"2025","journal-title":"Lancet Digit Health"},{"key":"2026072707352403000_16.7.e122980.27","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":"Collins","year":"2024","journal-title":"BMJ"},{"key":"2026072707352403000_16.7.e122980.28","first-page":"1142","article-title":"A profile of the Centre for Health Record Linkage","volume":"4","author":"Irvine","year":"2019","journal-title":"Int J Popul Data Sci"},{"key":"2026072707352403000_16.7.e122980.29","unstructured":"Australian Bureau of Statistics . Socio-Economic Indexes for Areas (SEIFA). Available: https:\/\/www.abs.gov.au\/websitedbs\/censushome.nsf\/home\/seifa [Accessed 18 Nov 2025]."},{"key":"2026072707352403000_16.7.e122980.30","unstructured":"World Health Organization . ATC classification. Available: https:\/\/www.who.int\/tools\/atc-ddd-toolkit\/atc-classification [Accessed 18 Nov 2025]."},{"key":"2026072707352403000_16.7.e122980.31","unstructured":"Medicare Benefits Schedule (MBS) Online . Australian Government Department of Health and Aged Care. Available: https:\/\/www.mbsonline.gov.au\/ [Accessed 17 Mar 2026]."},{"key":"2026072707352403000_16.7.e122980.32","unstructured":"National Centre for Classification in Health . The international statistical classification of diseases and related health problems, 10th revision, Australian modification (icd-10-am). Sydney NCCH; 2019."},{"key":"2026072707352403000_16.7.e122980.33","doi-asserted-by":"publisher","DOI":"10.1002\/sim.7992"},{"key":"2026072707352403000_16.7.e122980.34","doi-asserted-by":"crossref","DOI":"10.1038\/s41598-018-24271-9","article-title":"Recurrent Neural Networks for Multivariate Time Series with Missing Values","volume":"8","author":"Che","year":"2018","journal-title":"Sci Rep"},{"key":"2026072707352403000_16.7.e122980.35","unstructured":"Rossum G , Drake FL . Python tutorial. Amsterdam Centrum voor Wiskunde en Informatica."},{"key":"2026072707352403000_16.7.e122980.36","unstructured":"Paszke A , Gross S , Massa F , et al . PyTorch: an imperative style, high-performance deep learning library. Adv Neural Inf Process Syst 2019."}],"container-title":["BMJ Open"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.1136\/bmjopen-2026-122980","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T14:35:36Z","timestamp":1785162936000},"score":1,"resource":{"primary":{"URL":"https:\/\/bmjopen.bmj.com\/lookup\/doi\/10.1136\/bmjopen-2026-122980"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7]]},"references-count":36,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2026,7,27]]},"published-print":{"date-parts":[[2026,7]]}},"alternative-id":["10.1136\/bmjopen-2026-122980"],"URL":"https:\/\/doi.org\/10.1136\/bmjopen-2026-122980","relation":{},"ISSN":["2044-6055","2044-6055"],"issn-type":[{"value":"2044-6055","type":"print"},{"value":"2044-6055","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7]]}}}