PT - JOURNAL ARTICLE AU - Rudland, Simon V AU - Shah, Nisar H AU - Nevill, Alan TI - Community-based cardiovascular risk assessment using the Cardisio<sup>TM</sup> AI test: a prospective cohort study AID - 10.3399/BJGPO.2024.0183 DP - 2025 Oct 01 TA - BJGP Open PG - BJGPO.2024.0183 VI - 9 IP - 3 4099 - http://bjgpopen.org/content/9/3/BJGPO.2024.0183.short 4100 - http://bjgpopen.org/content/9/3/BJGPO.2024.0183.full SO - BJGP Open2025 Oct 01; 9 AB - Background Cardiovascular disease (CVD) accounts for significant morbidity and mortality disproportionately affecting hard-to-reach individuals. New technology that enables community testing rather than attending hospital may address health inequalities and facilitate new care pathways.Aim To explore whether the Cardisio test, which interprets three-dimensional vectorcardiography activity using a cloud-based artificial intelligence (AI) algorithm, can identify asymptomatic CVD.Design &amp; setting Prospective cohort study in three settings: general practice, pharmacy, and a community health centre. Recruitment targeted asymptomatic adults aged ≥18 years, with a QRISK3 score ≥10% or CVD risk factors.Method A 10-minute test using five electrodes (four chest, one back). The Cardisio results are classified into red, amber, or green based on the Cardisio test’s perfusion (P), structure (S), and arrhythmia (A) parameters. Pre- and post-test questionnaires provided feedback on participants’ experiences. Results reviewed by a chief investigator ([CI] independent consultant cardiologist) and dealt with according to the study participants’ results and medical profile.Results In total, 628 tests were performed, 51% male (n = 320), 49% (n = 308) female, with a mean age of 54 years (18–75 years). In the opinion of the CI, there was a strong association between one or more Cardisio red test results and referral to cardiology clinic being indicated (P&lt;0.001). The test was understood as easy to perform, with an 87.5% recommendation rate among participants (n = 492 of the 560).Conclusion This simple, near-patient test afforded high-risk hard-to-reach individuals with access to an acceptable test that can facilitate appropriate referral. The automated test does not rely on interpretation of electrocardiogram (ECG) readouts and so is more effective at identifying underlying CVD than a traditional 12-lead ECG.