| Title: | Development of a Multimodal Risk Stratification Framework to Guide Individualized Blood Pressure Targets |
| Lead Institution: | Beijing Anzhen Hospital, Capital Medical University |
| Principal investigator: |
Application 1116314
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
About
Hypertension is the leading modifiable determinant of cardiovascular and renal disease. Although intensive systolic blood pressure (SBP) lowering has been shown in large randomized trials to reduce morbidity and mortality, the applicability of these findings to heterogeneous populations is uncertain. Evidence is limited by selective eligibility criteria, modest follow-up, and underrepresentation of high-risk groups such as patients with peripheral artery disease, stroke, chronic kidney disease, or frailty. This project will use the UK Biobank's multimodal phenotyping and long-term outcomes to develop and validate an individualized framework for SBP target selection that quantifies net clinical benefit. Research questions Which baseline clinical, biochemical, imaging, lifestyle, and genetic features identify individuals most likely to benefit from intensive SBP lowering? How do vascular comorbidities and polygenic risk modify associations between SBP and cardiovascular or renal outcomes? Do multimodal models outperform conventional risk equations in stratifying benefit versus harm? Objectives Derive and validate a risk score integrating multimodal data to estimate individualized benefit-harm profiles of intensive BP control. Characterize heterogeneity of treatment effect across subgroups defined by age, sex, vascular disease, stroke, CKD, diabetes, and frailty. Compare machine learning methods (elastic-net, gradient boosting, random forest) with penalized Cox and competing-risk regression, evaluating calibration, discrimination, and clinical utility. Scientific rationale Current BP targets are based on population averages and may obscure heterogeneity. Multimodal integration enables granular risk stratification, aligning treatment with precision medicine principles. Findings will inform guideline development, support shared decision-making, and improve outcomes in high-risk groups.1 Publication
| Pub ID | Title | Author(s) | Year | Journal |
|---|---|---|---|---|
| 21279 | Overlapping airway diseases and risk of major adverse cardiovascular events: a UK biobank prospective cohort study | Pingsheng Hu (+6) | 2026 | Frontiers in Cardiovascular Medicine |
Enabling scientific discoveries that improve human health