The model is really good at telling which people are more likely to die in the next 10 years compared to others, based on their health numbers.
Claim Context
The CVD PREDICT micro-simulation model demonstrates high discrimination in predicting individual-level 10-year mortality risk, with an area under the ROC curve of 0.83 for all-cause mortality and 0.84 for CVD mortality, indicating that the model can effectively rank individuals by their risk of death.
“Areas under the ROC curves for model-predicted 10-year all-cause and CVD mortality risks were 0.83 (0.81–0.85) and 0.84 (0.81–0.88), respectively.”
Evidence from Studies
No evidence studies found yet.
What Would Prove This
Per GRADE and EBM methodology, here is what ideal scientific evidence would look like to definitively prove or disprove this claim, ordered from strongest to weakest.
Whether ROC performance is consistent across multiple CVD models and populations.
A systematic review of all published CVD risk prediction and simulation models that extracts AUC values for 10-year all-cause and CVD mortality from external validation studies, comparing performance across model types and populations.
The true individual-level risk of death over 10 years.
A prospective cohort of 10,000 U.S. adults aged 35–80 with baseline risk factor measurement and 10 years of follow-up for mortality, used to calculate observed individual risk and compare to model-predicted risk rankings.
Whether risk factor distributions match model assumptions at a point in time.
A cross-sectional survey of 5,000 U.S. adults aged 35–80 measuring age, sex, blood pressure, cholesterol, smoking, and diabetes status to compare with model input distributions.