Cardiovascular prevention models
These models predict different outcomes in different populations over different horizons. Choose the model validated for your patient's population — they are not interchangeable.
| Characteristic | ASCVD (PCE) | PREVENT | QRISK3 Pending verification | Framingham | SCORE2 Pending verification |
|---|---|---|---|---|---|
| Clinical question | What is the 10-year risk of a first atherosclerotic cardiovascular event to guide statin therapy? | What is the 10- and 30-year risk of total CVD, ASCVD and heart failure in adults without CVD? | What is the 10-year cardiovascular risk in a UK primary-care population? | What is the 10-year risk of any first cardiovascular event? | What is the 10-year risk of fatal and non-fatal CVD in European adults aged 40–69? |
| Outcome predicted | First hard ASCVD event (nonfatal MI, CHD death, fatal/nonfatal stroke) | Total CVD (ASCVD + heart failure), ASCVD, heart failure | CVD (CHD, stroke, TIA) | General CVD (coronary death, MI, coronary insufficiency, angina, stroke, TIA, PAD, heart failure) | Fatal and non-fatal CVD |
| Time horizon | 10 years | 10 and 30 years | 10 years | 10 years | 10 years |
| Intended population | Pooled US community cohorts (ARIC, CHS, CARDIA, Framingham Original and Offspring). | More than 3 million US adults aged 30–79 from 25 datasets (derivation and validation). | UK QResearch primary-care database. | Framingham Original and Offspring cohort participants aged 30–74, free of CVD at baseline. | — |
| Evidence level | Moderate evidence | Emerging evidence | Strong evidence | Moderate evidence | Strong evidence |
| Validation | — | Development and validation (Khan 2024): Derivation in 3,281,919 adults (1,839,828 female, 1,442,091 male) from 25 datasets; validated internally and externally. | — | — | — |
| Discrimination | — | — | — | — | — |
| Calibration | Overestimation reported in several contemporary US cohorts. | — | — | — | — |
| Geographic applicability | United States populations. | United States. | United Kingdom | United States (predominantly white cohort). | Europe (4 risk regions) |
| Guideline status | Basis of the 2018 ACC/AHA cholesterol and 2019 ACC/AHA primary-prevention guidelines. | Developed by the AHA as part of the cardiovascular-kidney-metabolic (CKM) health initiative. | — | — | — |
| Important limitations |
|
| — |
| — |
Inputs required
Which variables each model uses. Models with more inputs are not automatically better — validation in your population matters most.
| Input | ASCVD (PCE) | PREVENT | QRISK3 | Framingham | SCORE2 |
|---|---|---|---|---|---|
| Age | ●used | ●used | ●used | ●used | ●used |
| Sex | ●used | ●used | ●used | ●used | ●used |
| Race | ●used | –not used | –not used | –not used | –not used |
| Total cholesterol | ●used | ●used | –not used | ●used | –not used |
| HDL | ●used | ●used | –not used | ●used | –not used |
| Systolic BP | ●used | ●used | ●used | ●used | ●used |
| BP treatment | ●used | ●used | –not used | ●used | –not used |
| Diabetes | ●used | ●used | ●used | ●used | –not used |
| Smoking | ●used | ●used | ●used | ●used | ●used |
| Statin use | –not used | ●used | –not used | –not used | –not used |
| eGFR | –not used | ●used | –not used | –not used | –not used |
| BMI | –not used | ●used | ●used | –not used | –not used |
| Ethnicity | –not used | –not used | ●used | –not used | –not used |
| Deprivation | –not used | –not used | ●used | –not used | –not used |
| Total cholesterol/HDL ratio | –not used | –not used | ●used | –not used | –not used |
| SBP variability | –not used | –not used | ●used | –not used | –not used |
| CKD | –not used | –not used | ●used | –not used | –not used |
| AF | –not used | –not used | ●used | –not used | –not used |
| Migraine | –not used | –not used | ●used | –not used | –not used |
| SLE | –not used | –not used | ●used | –not used | –not used |
| Severe mental illness | –not used | –not used | ●used | –not used | –not used |
| Atypical antipsychotics | –not used | –not used | ●used | –not used | –not used |
| Steroids | –not used | –not used | ●used | –not used | –not used |
| Erectile dysfunction | –not used | –not used | ●used | –not used | –not used |
| Family history | –not used | –not used | ●used | –not used | –not used |
| Non-HDL cholesterol | –not used | –not used | –not used | –not used | ●used |
| Risk region | –not used | –not used | –not used | –not used | ●used |
Comparison shows model characteristics only. Different models predict different outcomes in different populations, so their numerical results are not directly comparable.