The value of additional risk factors in addition to current prediction rules to better predict 10-year cardiovascular risk

Date of Approval: 
2020-08-24 00:00:00
Lay Summary: 
In order to tailor preventive treatment for cardiovascular disease (conditions such as a heart attack, heart failure or stroke) to individuals, it is important to accurately assess their risk of cardiovascular disease. For this, models exist which can predict this risk, based on several risk factors. However, sometimes there is more information available than used in the model. For example, one of someone’s parents had cardiovascular problems at a younger age. This may increase someone’s risk of cardiovascular disease, but it is unclear how to exactly implement this information in the risk prediction. The goal of this study is to evaluate a methodology to use such additional data and to evaluate the effect of a list of extra predictors. Results of the study will be directly implemented in online risk calculators for easy clinical use.
Technical Summary: 
The use of prediction models for estimating cardiovascular risk is recommended by European and American guidelines and can help to tailor preventive treatment to the individual patient. Several prediction models are available in the primary prevention to 10-year cardiovascular risk, e.g. the SCORE-model and the ASCVD pooled cohort equation. These models are widely-used and practical because they use easy to measure and generally available risk factors to calculate 10-year cardiovascular risk. In clinical practice however, often other risk factors are known apart from those in the prediction model, for example family history or a coronary calcium score. It is unclear how this additional data affects risk prediction or how best to handle this in clinical practice. The goal of the current study is to validate a methodology of flexible prediction, the naïve method. This method allows extra predictors to be used without the need of a new regression model for every combination of predictor availability. Using this methodology, the effect of the most common additional predictors on top of a basic model will be quantified. Results of the study will be directly implemented in online risk calculators such as for easy clinical use.
Health Outcomes to be Measured: 
fatal and non-fatal stroke; fatal and non-fatal myocardial infarction; vascular death; all-cause mortality
Application Number: 

Olaf Klungel - Chief Investigator - Utrecht University
Patrick Souverein - Corresponding Applicant - Utrecht University
Frank Visseren - Collaborator - Utrecht University
Jannick Dorresteijn - Collaborator - UMC Utrecht
Romin Pajouheshnia - Collaborator - Utrecht University
Steven Hageman - Collaborator - Utrecht University

HES Admitted Patient Care;ONS Death Registration Data