Do positive psychosocial factors contribute to the prediction of coronary artery disease? A UK Biobank–based machine learning approach
Authors' conclusion
Positive psychosocial factors were individually linked to CAD risk, in line with previous studies, and as reflected by the new European Society of Cardiology guidelines on cardiovascular disease prevention. However, including available PPFs in CAD-prediction models did not improve prediction compared with the FRS alone.
Study context
- Population
- UK Biobank participants without coronary artery disease (CAD) at baseline160,226 to 441,419 analysed
- Exposure
- Framingham Risk Score (FRS) plus positive psychosocial factors (PPFs) vs FRS alone for CAD prediction
- Outcome
- Prediction of AMI and chronic ischaemic heart disease (AUC change)
Methodological detailsmachine learning (XGBoost), logistic regression, validation sample
- Machine learning (XGBoost)
- Logistic regression
- Validation sample
Main finding
As reported in the abstract
Acute myocardial infarction (AMI) AUC change, logistic regression (no improvement)
0.02%
AMI AUC change, XGBoost (no improvement)
0.90%
Chronic ischaemic heart disease (CIHD) AUC change, logistic regression (no improvement)
−1.10%
CIHD AUC change, XGBoost (no improvement)
−0.88%
Observational data: associations, not effects.
Further findings
Association of individual PPFs with AMI and CIHD risk
- Happiness, satisfaction with health and life, and participation in social activities linked to lower AMI and CIHD risk
- all P-for-trend ≤ 0.04
- Social support not linked
Original publication
Summarised from the published abstract; every number is checked against it