Heart Age Calculator — Your Cardiovascular Age from PREVENT

Methodology

Heart age turns an abstract percentage — your predicted 10-year cardiovascular risk — into something intuitive: an age. It is the age at which a person with ideal risk factors would carry the same risk you do. A heart age above your real age means your blood pressure, cholesterol and health history are ageing your cardiovascular system faster than the calendar.

The risk model

We use the American Heart Association's PREVENT equations (Khan et al., Circulation, 2024) — the current standard for cardiovascular risk in US adults aged 30–79, and the model behind the direction of the 2026 ACC/AHA guidelines. Specifically, we compute the base model for 10-year total cardiovascular disease (heart attack, stroke and heart failure combined), using the exact published sex-specific coefficients.

Each predictor is entered into a logistic model:

risk = 1 / (1 + e^−(β·predictors + constant))

The predictors are age, non-HDL cholesterol (total minus HDL), HDL, systolic blood pressure, diabetes, current smoking, kidney function (eGFR), and whether you take blood-pressure or statin medication — plus the interaction terms the model specifies. Cholesterol is handled in mmol/L (converted from mg/dL at ÷38.67).

We transcribed the coefficients from the open-source preventr implementation, which reproduces the AHA supplement and is validated against the AHA's own online calculator, and we locked our implementation to a reproduced worked example (a 50-year-old woman's profile returning 14.7%, matching to the digit) before shipping.

Note on BMI. PREVENT's total-CVD model gives body-mass index a coefficient of exactly zero — BMI drives its separate heart-failure sub-model, not this one — so we don't ask for it here rather than collect an input that wouldn't move your result.

How heart age is derived

We compute your 10-year risk, then find the age at which someone with an optimal reference profile of the same sex has that same risk. Our reference profile is: total cholesterol 170 mg/dL, HDL 55 mg/dL, systolic blood pressure 110 mmHg untreated, non-smoking, non-diabetic, normal kidney function, no statin. If your risk falls outside the model's validated 30–79 window, we say so rather than extrapolate.

ApoB and Lp(a): shown as context, never in the maths

ApoB and lipoprotein(a) are increasingly discussed, and rightly so — but they are not inputs to the PREVENT equations. Folding them into the risk score would be inventing a number the model was never built to produce. So if you enter them, we show them separately, against widely-cited reference thresholds, purely as context to raise with your doctor:

• Lp(a): below ~75 nmol/L is generally lower-risk; ≥125 nmol/L (≈50 mg/dL) is considered high risk (NLA 2024 focused update). Lp(a) is largely genetic and worth measuring once in a lifetime.
• ApoB: a level below ~90 mg/dL is a common desirable reference for primary prevention, with lower targets at higher risk. It reflects the number of atherogenic particles, which can be informative when standard cholesterol looks reassuring.

These are general references, not treatment targets, and they don't change your heart-age figure.

Interpreting your result

A heart age below your real age is the favourable direction. But this is a population-level estimate, not a personal forecast — two people with the same heart age can have very different outcomes. The most useful things about it are the size of any gap and what's driving it: blood pressure and smoking, in particular, move the number a lot and are among the most modifiable.

Treat the result as a conversation-starter with a clinician. Never start, stop or change any medication on the basis of it.

Limitations

• PREVENT is validated for adults aged 30–79; outside that range it does not apply.
• It was developed in US population data; performance can differ elsewhere.
• Total-CVD risk is not the same as the ASCVD figure some other tools report, so numbers won't line up one-to-one across calculators.
• This is a research-grade risk estimate, not a diagnosis or a medical test. Any decision about your health belongs with a qualified clinician.

Sources

Pull the receipts — 4 cited sources

FitTools · sources

1. 01 Khan SS, et al. Development and Validation of the American Heart Association's PREVENT Equations. Circulation 2024;149:430–449
2. 02 Mayer M. preventr — open-source implementation of the PREVENT equations (source of the exact coefficients used here; validated against the AHA calculator)
3. 03 Koschinsky ML, et al. A focused update to the 2019 NLA scientific statement on use of lipoprotein(a) in clinical practice. J Clin Lipidol 2024 (Lp(a) risk thresholds)
4. 04 American College of Cardiology. An Update on Lipoprotein(a): Testing, Treatment, and Guideline Recommendations, 2023 (Lp(a) and ApoB context)

Every formula cited ✓