Estimating a patient’s ten-year risk of cardiovascular disease turns a lipid panel and a few risk factors into a single number that guides shared decisions about statins and lifestyle. This calculator implements the widely cited Framingham general CVD model (D’Agostino et al., Circulation 2008).
How it works
The model is a sex-specific Cox proportional hazards equation. Each continuous variable is log-transformed and multiplied by a published coefficient, and binary terms add a fixed amount for treated blood pressure, smoking and diabetes:
sum = b_age*ln(age) + b_tc*ln(TC) + b_hdl*ln(HDL)
+ b_sbp*ln(SBP) + b_smoke*smoker + b_dm*diabetes
risk = 1 - S0 ^ exp(sum - meanSum)S0 is the baseline ten-year survival and meanSum is the mean of the linear predictor in the original cohort. Separate coefficients apply for treated versus untreated systolic pressure, because treatment marks a higher underlying risk. All coefficients come from D’Agostino et al., Circulation 2008.
What each input contributes to risk
Age is the strongest driver in the model. For both men and women, the coefficient on age dominates because risk rises steeply with every decade — a 60-year-old with the same lipid and blood pressure values as a 40-year-old will score substantially higher.
Total cholesterol and HDL cholesterol are not interchangeable. TC contributes positively to risk; HDL contributes negatively. A high HDL meaningfully reduces the score even when TC is elevated, which reflects the real-world benefit of HDL in reverse cholesterol transport. Units matter: enter both in mg/dL. To convert from mmol/L, multiply by 38.67.
Systolic blood pressure is entered separately for treated versus untreated values. The model uses a higher coefficient for treated SBP because treatment implies a history of hypertension — the underlying risk is higher than the current reading suggests.
Smoking (current cigarette smoking) is a binary variable. Former smoking is not included in this model; the risk reduction from quitting does not appear here but is well documented clinically.
Diabetes is also binary. In people with diabetes, the 2008 general CVD model treats it as a binary risk factor. Separate diabetes-specific risk tools may be more precise.
Understanding the risk bands
The result is a percentage — for example, 14% — which means that in 100 people with this risk profile, approximately 14 would be expected to experience a cardiovascular event (coronary heart disease, stroke, peripheral arterial disease, or heart failure) within ten years, compared with a risk-matched population.
- Below 10% — generally considered low risk; lifestyle counselling is first-line.
- 10–19% — intermediate risk; the range where the statin discussion becomes most nuanced, often guided by additional factors (coronary calcium score, family history).
- 20% or above — high risk; most guidelines recommend statin therapy unless contraindicated.
These thresholds reflect common clinical practice but vary by country. UK NICE guidance uses 10% as the treatment threshold; ACC/AHA in the USA uses the pooled-cohort equations rather than this model.
Comparing this model to alternatives
The 2008 Framingham general CVD model is widely cited in research because it predicts a broad composite endpoint and is validated on a large US cohort. In clinical practice:
- QRISK3 (UK) adds deprivation, ethnicity, chronic kidney disease, atrial fibrillation, systemic lupus, severe mental illness, and erectile dysfunction — making it more appropriate for a diverse population.
- ACC/AHA Pooled Cohort Equations (USA) are the standard for US primary care and incorporate race as a variable.
- SCORE2 (Europe) is calibrated for European populations and predicts fatal cardiovascular events.
Use this calculator when you want to understand the Framingham framework or when QRISK3 or PCE are unavailable. For actual clinical decision-making, use whichever tool your local guideline specifies.
Tips and notes
Enter cholesterol in mg/dL; multiply mmol/L values by 38.67 first. The tool is valid only for primary prevention in adults aged 30 to 79 and should not be used in people who already have cardiovascular disease, where risk is already high by definition. A result at or above 10 percent typically opens a statin discussion, and 20 percent or more is firmly high risk, but treatment thresholds vary by country. Treat this as an educational estimate to support, not replace, a clinician’s judgement.