ASCVD Risk Calculator
Estimate 10-year risk of a first atherosclerotic cardiovascular disease event with the sex-specific AHA PREVENT base equations now used by the 2026 ACC/AHA lipid guideline.
Enter the prevention profile
What this calculator computes
10-year risk = exp(linear predictor) ÷ [1 + exp(linear predictor)]
The sex-specific linear predictor uses centered age, non-HDL-C, HDL-C, systolic blood pressure below and above equation knots, diabetes, current smoking, eGFR below and above its knot, antihypertensive use, statin use, treated-pressure and treated-lipid terms, and age interactions. Coefficients come from the published PREVENT base 10-year ASCVD model.
Total and HDL cholesterol are converted from mg/dL to mmol/L internally with 0.02586. Non-HDL-C equals total cholesterol minus HDL-C. LDL-C is not a coefficient in this particular equation, but the 2026 guideline uses LDL-C 70–189 mg/dL to define this primary-prevention risk-assessment group, so the calculator requires it as an eligibility gate.
The calculator implements only the base ASCVD endpoint. It does not produce total CVD, heart failure, coronary heart disease, stroke, or 30-year risk and does not add optional urine albumin-to-creatinine ratio, hemoglobin A1c, or social deprivation index terms.
Published profile cross-check
The default profile is a 50-year-old woman with total cholesterol 240 mg/dL, HDL-C 55, systolic blood pressure 160 on antihypertensive medication, no statin, no diabetes, no current smoking, BMI 35, and eGFR 90. LDL-C 130 is entered to satisfy the calculator’s guideline eligibility check.
Applying the female base PREVENT-ASCVD coefficients gives a linear predictor of approximately −3.2768. The logistic transformation produces 0.03638, displayed as 3.6%. The original PREVENT paper reports 3.6% for the same profile. Changing only the smoking input to current smoking yields approximately 5.99%, matching the paper’s 6.0% example.
That cross-check tests implementation fidelity. It does not mean smoking initiation would cause exactly a 2.3-point change or cessation would instantly remove it. Model inputs describe observed populations; one-variable re-estimates are not randomized treatment effects or personal forecasts.
2026 guideline risk categories
Low
Less than 3% estimated 10-year ASCVD risk.
Borderline
3% to less than 5%.
Intermediate
5% to less than 10%.
High
10% or greater.
A category is a starting point, not an automatic prescription. The 2026 guideline combines estimated risk with LDL-C, age, diabetes, chronic kidney disease, HIV, pregnancy-related history, family history, inflammatory conditions, triglycerides, lipoprotein(a), apolipoprotein B, coronary artery calcium when appropriate, patient preferences, likely benefit, adverse effects, and other clinical factors.
Risk thresholds can inform the intensity of discussion, but a value of 4.9% and 5.0% should not be treated as biologically different people. Measurement uncertainty, day-to-day blood-pressure variation, laboratory variation, and model uncertainty can move a result around a boundary.
Who fits this calculator’s eligibility gate
This implementation is designed for U.S. primary-prevention adults ages 30 through 79, without known clinical ASCVD and without known subclinical atherosclerosis, whose LDL-C is 70 through 189 mg/dL. The calculator blocks entries outside those boundaries so an attractive number is not mistaken for a guideline-applicable result.
Clinical ASCVD includes prior myocardial infarction, acute coronary syndrome, ischemic stroke, transient ischemic attack of atherosclerotic origin, symptomatic peripheral artery disease, or arterial revascularization in the relevant context. Known coronary plaque or coronary artery calcium can represent subclinical atherosclerosis and changes how risk is interpreted.
People with LDL-C 190 mg/dL or higher, familial hypercholesterolemia, existing ASCVD, or other guideline-defined high-risk conditions may have management pathways that do not depend on a PREVENT threshold. A clinician should determine the applicable pathway rather than altering an input until the calculator runs.
Measure inputs consistently
Use a current lipid panel and blood pressure that a clinician considers representative. Total cholesterol must exceed HDL-C. LDL-C may be directly measured or calculated by the laboratory; use the reported value and note whether the sample was fasting if relevant. Do not combine lipid values from different dates.
Systolic pressure is the upper blood-pressure number. A single hurried office reading or home-device error may not represent usual pressure. Proper cuff size, rest, body position, repeated measurements, and validated equipment matter. Mark antihypertensive use according to the equation’s intended current-treatment status, not whether pressure happens to be controlled.
eGFR should come from a laboratory report in mL/min/1.73 m². Do not enter serum creatinine in that field. Diabetes and current smoking are yes/no equation variables; “former smoker” is not current smoking. Statin use reflects current therapy, and self-discontinuing medication to explore a lower model estimate would be unsafe and conceptually wrong.
Why there is no race input
PREVENT was deliberately developed without race as a predictor. The investigators sought equations reflecting cardiovascular, kidney, and metabolic health while avoiding the use of race as a biological proxy. The model was developed and externally validated in large, diverse U.S. datasets.
Removing race does not erase inequities. Access to care, neighborhood conditions, exposures, stress, discrimination, treatment, income, and other structural factors can affect health and model performance. Optional PREVENT models can incorporate social deprivation information, but this calculator uses the base equation so it can be calculated from common clinical data.
Every prediction model can perform differently across populations. An individual estimate should be interpreted with clinical judgment, and health systems using PREVENT at scale should monitor calibration, missingness, and equitable implementation.
Risk enhancers and coronary artery calcium
A calculated number cannot capture every reason risk may be higher or lower. Family history of premature ASCVD, elevated lipoprotein(a), elevated apolipoprotein B, persistent hypertriglyceridemia, chronic inflammatory disease, pregnancy-associated conditions, chronic kidney disease, and other factors may shape a treatment discussion.
When uncertainty remains, coronary artery calcium scanning may help reclassify selected adults, subject to guideline indications, radiation considerations, age, pregnancy status, cost, and patient preference. A CAC score is not an input to the base PREVENT equation because known subclinical disease belongs outside this calculator’s simple eligibility gate.
Do not order tests solely to change an online category. A clinician should decide whether a test is likely to change management and how a finding affects the whole prevention plan.
Read “3.6%” correctly
An estimate of 3.6% means that among people sufficiently similar to the modeled profile, roughly 36 of 1,000 may experience the defined first ASCVD event over ten years, on average, under the mixture of care and exposures represented in the data. It does not identify which individuals will have an event and does not promise 964 will remain free of every cardiovascular problem.
ASCVD in this endpoint comprises coronary heart disease and stroke outcomes defined by the PREVENT investigators. It is not the probability of heart failure, any heart symptom, all-cause death, or needing a procedure. Competing events and changing future measurements make a decade-long estimate inherently uncertain.
The risk is conditional on data accuracy and model applicability. It should be recorded with inputs, date, equation version, endpoint, time horizon, and reason for calculation. Repeating it after a clinically meaningful change can update a discussion, but small fluctuations should not drive frequent medication changes.
A practical prevention conversation
Verify
Confirm primary-prevention eligibility, current measurements, medications, and diagnoses.
Estimate
Record the PREVENT endpoint, horizon, percentage, and category without overstating certainty.
Enrich
Review risk enhancers, lifestyle, family history, preferences, and CAC only when appropriate.
Decide
Build a clinician-guided plan for lipids, pressure, diabetes, smoking, activity, sleep, and follow-up.
ASCVD estimate checklist
- Confirm age 30–79.
- Confirm no known clinical ASCVD.
- Confirm no known subclinical atherosclerosis.
- Verify LDL-C is 70–189 mg/dL.
- Use one current lipid panel.
- Use representative systolic pressure.
- Enter antihypertensive status accurately.
- Enter current statin use accurately.
- Use laboratory eGFR, not creatinine.
- Distinguish current from former smoking.
- Review diabetes status.
- Discuss risk enhancers and preferences.
- Do not treat a category as a mandate.
- Document equation version and date.
It does not evaluate medical expenses, insurance choices, preventive treatment, or cardiovascular risk.
Frequently asked questions
Is this the old Pooled Cohort Equation calculator?
No. It uses the AHA PREVENT base 10-year ASCVD equations, aligned with the 2026 ACC/AHA lipid guideline.
Why does the calculator ask for LDL-C if the formula uses non-HDL-C?
LDL-C defines the guideline population of 70–189 mg/dL. The equation itself derives non-HDL-C from total and HDL cholesterol.
Does the smoking comparison show the benefit of quitting?
No. It is a one-input model re-estimate, not a causal or immediate treatment effect. Smoking cessation has broad health benefits that require fuller evidence.
Can I use this after a heart attack or stroke?
No. Known ASCVD is secondary prevention and falls outside this calculator’s eligibility gate.
Does a low score mean I need no prevention?
No. Healthy behaviors and management of blood pressure, lipids, diabetes, and smoking remain important, and some conditions change the treatment pathway regardless of score.