Synapse
⌘+K
Synapse
PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
April 13, 2026American Journal of Preventive CardiologyOpen Access

Independent External Validation and Head-to-Head Comparison of Guideline-Recommended CVD Risk Prediction Models

View Full Paper
Ask AI
Bookmark
Share

Why the study?

How do the PREVENT, SCORE2, and QRISK3 CVD risk prediction models compare in performance and transportability when externally validated in the UK Biobank population?

Population

502,157 participants from the UK Biobank (UKBB) with complete data for all models

Comparison

Application of three guideline-endorsed CVD risk… vs Head-to-head comparison between the three models

Design

Cohort

Authors

LKLum KastratiEAEleftheria Maria AlexandriMRMary E. Rupp

Discussion

Loading...

Member takes

Overview

The PREVENT, SCORE2, and QRISK3 CVD risk prediction models demonstrate similar discriminative performance but systematically overestimate risk in older individuals when applied to the UK Biobank cohort.

Key Points

  • This research aims to compare the performance of three guideline-recommended CVD risk prediction models and assess their validity in different populations.
  • Conducted external validation using the UK Biobank dataset.
  • Assessed model performance with Brier scores and AUC, stratified by sex.
  • Evaluated the risk of bias using the PROBAST tool.
  • Performed direct comparisons in primary and secondary analyses based on eligibility criteria.
  • AUC for female participants ranged from 0.7092 to 0.7468, while for males it ranged from 0.6813 to 0.6946.
  • Overall predictive performance was generally better for females.
  • Calibration was suboptimal, especially for older individuals, showing systematic risk overestimation.
  • Despite differences, the models showed consistent performance across various outcomes.

Structured PICO

How do the PREVENT, SCORE2, and QRISK3 CVD risk prediction models compare in performance and transportability when externally validated in the UK Biobank population?

P
Population
502,157 participants from the UK Biobank (UKBB) with complete data for all models
I
Intervention
Application of three guideline-endorsed CVD risk prediction models: PREVENT (AHA), SCORE2 (ESC), and QRISK3 (NICE)
C
Comparator
Head-to-head comparison between the three models
O
Outcome
Model performance assessed using Brier scores, Area Under the Receiver Operating Curve (AUC), and calibration across original and alternative outcome definitions, stratified by sex

The PREVENT, SCORE2, and QRISK3 CVD risk prediction models demonstrate similar discriminative performance but systematically overestimate risk in older individuals when applied to the UK Biobank cohort.

Limitations

  • All models were at high risk of bias

Cite This Study

Kastrati et al. (2026) studied this question.

synapsesocial.com/papers/69dc887f3afacbeac03ea5b0https://doi.org/10.1016/j.ajpc.2026.101625
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 3 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Global Burden of Cardiovascular Diseases and Risk Factors, 1990–20192020 · 11,659 citations
  2. 2UK Biobank: An Open Access Resource for Identifying the Causes of a Wide Range of Complex Diseases of Middle and Old Age2015 · 13,905 citations
  3. 3SCORE2 risk prediction algorithms: new models to estimate 10-year risk of cardiovascular disease in Europe2021 · 1,782 citations