Medical history predicts phenome-wide disease onset and enables the rapid response to emerging health threats
The COVID-19 pandemic exposed a global deficiency of systematic, data-driven guidance to identify high-risk individuals. Here, we illustrate the utility of routinely recorded medical history to predict the risk for 1741 diseases across clinical specialties and support the rapid response to emerging health threats such as COVID-19. We developed a neural network to learn from health records of 502,489 UK Biobank participants. Importantly, we observed discriminative improvements over basic demographic predictors for 1546 (88.8%) endpoints. After transferring the unmodified risk models to the All of US cohort, we replicated these improvements for 1115 (78.9%) of 1414 investigated endpoints, demonstrating generalizability across healthcare systems and historically underrepresented groups. Ultimately, we showed how this approach could have been used to identify individuals vulnerable to severe COVID-19. Our study demonstrates the potential of medical history to support guidance for emerging pandemics by systematically estimating risk for thousands of diseases at once at minimal cost.
Authors
- Maik Pietzner (ORCID: https://orcid.org/0000-0003-3437-9963)
- Andre Vauvelle (ORCID: https://orcid.org/0000-0003-2957-4166)
- John Deanfield (ORCID: https://orcid.org/0000-0001-8806-6052)
- Thore Buergel (ORCID: https://orcid.org/0000-0003-1159-007X)
- Benjamin Wild (ORCID: https://orcid.org/0000-0002-7492-8448)
- Roland Eils (ORCID: https://orcid.org/0000-0002-0034-4036)
- Stefan Hegselmann (ORCID: https://orcid.org/0000-0002-2145-3258)
- Claudia Langenberg (ORCID: https://orcid.org/0000-0002-5017-7344)
- Julius Upmeier zu Belzen (ORCID: https://orcid.org/0000-0002-0966-4458)
- Jakob Steinfeldt (ORCID: https://orcid.org/0000-0003-1387-2054)
- Spiros Denaxas (ORCID: https://orcid.org/0000-0001-9612-7791)
- Ulf Landmesser (ORCID: https://orcid.org/0000-0002-0214-3203)
- Harry Hemingway (ORCID: https://orcid.org/0000-0003-2279-0624)
Institutions
- Queen Mary University of London (GB)
- Heidelberg University (DE)
- University of Münster (DE)
- University of Cambridge (GB)
- University Hospital Heidelberg (DE)
- British Heart Foundation (GB)
- National Institute for Health and Care Research (GB)
- Humboldt-Universität zu Berlin (DE)
- Deutsches Herzzentrum der Charité (DE)
- German Centre for Cardiovascular Research (DE)
- Berlin Institute of Health at Charité - Universitätsmedizin Berlin (DE)
- Health Data Research UK (GB)
- MRC Epidemiology Unit (GB)
- University College London (GB)
- Massachusetts Institute of Technology (US)
- Freie Universität Berlin (DE)
- Charité - Universitätsmedizin Berlin (DE)
Publication Details
- Journal
- Nature Communications
- Published
- 2025-01-10
- DOI
- https://doi.org/10.1038/s41467-025-55879-x
- Citations
- 6
- Primary Topic
- Machine Learning in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 14.89
Funders
- National Institute for Health and Care Research
- British Heart Foundation
- University College London
- Deutsche Forschungsgemeinschaft
- Bundesministerium für Bildung und Forschung
- Einstein Stiftung Berlin
- University College London Hospitals NHS Foundation Trust