The abstract is read at the publisher; the summary is JClub's.
ClinicalAge, an EHR-derived ageing biomarker in UK Biobank, aligns closely with PhenoAge (R²=0·819) and predicts higher all-cause, cancer, and cardiovascular mortality.
- Why it matters: Routine longitudinal EHR data could enable earlier, non-invasive identification of accelerated ageing, but existing biomarkers do not fully capture ageing trajectories across the life course.
- What they did: A transformer deep-learning model was trained on UK Biobank and validated in three CPRD Aurum cohorts, using PhenoAge as reference, then tested associations with functional measures, biological ageing biomarkers, 31 ageing-related diseases, and mortality.
- The result: ClinicalAge significantly associated with 17 functional measures and ten biological ageing biomarkers, predicted incidence of all 20 non-cancer diseases, and showed adjusted hazard ratios of 1·23 (UK Biobank) and 1·36 (CPRD) for all-cause mortality.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.