Heart health
Understanding your 10-year cardiovascular risk score
9 min read
Cardiovascular risk calculators take a handful of routine inputs and return a single percentage: your estimated probability of a heart attack or stroke over the next ten years. Three widely used models — ASCVD, QRISK3 and SCORE2 — will often give you three different answers from identical inputs.
That disagreement is not a bug. Each was built on a different population, with a different definition of the outcome it predicts.
What each model is actually built on
The differences are worth understanding before you compare your numbers.
- ASCVD Pooled Cohort Equations: US cohorts, predicts first hard cardiovascular event (non-fatal MI, coronary death, stroke) in adults 40–79
- QRISK3: UK primary care records from millions of patients, adds deprivation, ethnicity, chronic kidney disease, migraine, corticosteroids, atypical antipsychotics and severe mental illness
- SCORE2 and SCORE2-Diabetes: European cohorts recalibrated by regional risk level, uses non-HDL cholesterol and reports fatal plus non-fatal events
- Framingham: the original, still useful, but derived from a largely white population from an era with very different treatment patterns
How to read the percentage
A 12% ten-year risk means that out of 100 people sharing your risk-factor profile, roughly 12 would be expected to have an event in a decade. It says nothing about which twelve, and nothing about you specifically.
Thresholds vary by guideline, but broadly: under 5% is low, 5–7.5% borderline, 7.5–20% intermediate, and 20% or above high. Statin conversations usually begin in the intermediate band and become firm recommendations above it.
Why your score can look wrong
The models are calibrated for average risk within broad categories. They systematically underestimate risk in people with a strong family history of premature heart disease, familial hypercholesterolaemia, chronic inflammatory disease, HIV, or a history of pre-eclampsia.
They also underperform in populations they were not derived from — a well-documented limitation when applying US or European equations in African, South Asian and other under-represented groups.
The useful part: what changes if you change something
The single best use of a risk calculator is comparative. Re-run it with your systolic pressure 10 mmHg lower, or with smoking set to no, and watch the percentage move. That difference is the honest estimate of what the change is worth to you.
Stopping smoking typically produces the largest single reduction available to anyone who smokes. Blood pressure control is usually next, then LDL lowering.
Key takeaways
- Different models disagree because they were built on different populations and predict different outcomes.
- A percentage describes a group of people like you, not a personal forecast.
- Scores underestimate risk in several specific groups, including strong family history and inflammatory disease.
- Use the calculator comparatively — the change in score is more actionable than the score.
Related reading
Sources
- Goff DC et al. 2013 ACC/AHA Guideline on the Assessment of Cardiovascular Risk
- Hippisley-Cox J et al. Development and validation of QRISK3, BMJ 2017
- SCORE2 working group, European Heart Journal 2021
This article is educational and is not medical advice. Always discuss your own results with a qualified healthcare professional.
Know your risk, take action
Create a free account and get your first risk assessment in under ten minutes. No card required.