Key takeaways
- When researchers ran the same 3,785 American adults through the old and new heart risk calculators, average 10-year risk fell from 8.0% to 4.3%, though no one’s blood pressure or cholesterol changed2.
- The 2026 cholesterol guideline moved its cutoffs down to match: “intermediate” risk now starts at 5% instead of 7.5%, and “high” at 10% instead of 20%56.
- Facts the calculator never asks about, such as a parent’s early heart attack, preeclampsia, lupus or a high Lp(a), count as “risk enhancers” that can tip a borderline decision56.
- A low 10-year number can hide a high lifetime risk: among 55-year-old men, 29.6% with two or more major risk factors died of heart or blood vessel disease by 80, versus 4.7% with ideal numbers11.
- If your new score is lower, don’t stop a statin or blood pressure medicine on your own; PREVENT’s lead developer cautioned that the studies comparing the two tools couldn’t, by themselves, guide treatment2.
In June 2024, a team led by University of Pittsburgh physician Timothy Anderson took 3,785 adults from a national health survey and ran each one through two heart risk calculators1. One was the tool U.S. guidelines had relied on since 20132. The other was PREVENT, newer equations from the American Heart Association12.
The people didn’t change between runs. Same blood pressure, same cholesterol, same smoking. Yet their average 10-year risk of a heart attack or stroke dropped from 8.0% to 4.3%2. For Black adults it fell from 10.9% to 5.1%, and for adults 70 to 75, from 22.8% to 10.2%1.
Under the treatment rules then in place, the number of Americans advised to take a statin would have shrunk from 45.4 million to 28.3 million1. By our math, that’s about 17 million people who would no longer be advised to take one, because of a new formula.
So which number is true? Neither is a fact about your arteries. A risk score is a forecast. We walk through how yours is built, how doctors use it and what it leaves out.
A risk score forecasts what happens to people like you
Cardiovascular disease means disease of the heart and blood vessels, including heart attacks, strokes and heart failure. It killed 919,032 Americans in 2023, about 1 in 3 deaths, according to the CDC9. About 1 in 6 of those deaths were in adults younger than 659.
A risk calculator turns a handful of facts about you into a percentage. A 10-year risk of 8% means that out of 100 people with numbers like yours, about 8 would be expected to have a heart attack or stroke within a decade. The other 92 would not.
The calculator can’t say which group you’ll be in. The U.S. Preventive Services Task Force, an independent panel that grades preventive care, reminds clinicians that these predictions are estimates8.
Think of a weather forecast: a 30% chance of rain helps you decide about an umbrella, but it doesn’t promise a dry afternoon.
From one Massachusetts town to 6.6 million adults
Modern heart risk math began in Framingham, Massachusetts, where researchers enrolled 5,209 local men and women in 194810. By 1961 the study had tied high blood pressure and high cholesterol to heart disease, and the term “risk factor” caught on10.
In 1998, the Framingham team published a 10-year heart disease score built from age, sex, smoking, diabetes, blood pressure and cholesterol10.
The 2013 Pooled Cohort Equations, or PCE, came next. They covered adults 40 to 75, predicted heart attacks and strokes, and used race as an input5. They were built from older study groups totaling about 12,500 people5.
When researchers tested the PCE in real populations, many studies found it predicted more events than happened, the Task Force notes8. Northwestern cardiologist Sadiya Khan, who led the team behind PREVENT, put the overestimate at about twofold2. A colleague noted it ran low in some Asian groups2.
PREVENT was designed to fix those problems. Its developers drew on data from 6,612,004 U.S. adults in 46 data sets, using about half to build the equations and half to test them3. The paper appeared in the journal Circulation in February 20243.
| Feature | 2013 Pooled Cohort Equations | PREVENT equations |
|---|---|---|
| Ages covered | 40 to 755 | 30 to 793 |
| What it predicts | Heart attack and stroke5 | Heart attack, stroke and heart failure3 |
| Time frames | 10 years5 | 10 and 30 years56 |
| Race as an input | Yes5 | No; optional social measures instead35 |
| Kidney function | Not included5 | Included3 |
| Optional extras | None5 | A1C, urine albumin and a ZIP-code-based social index3 |
In testing, PREVENT’s predictions tracked what happened closely. Its calibration slope, a measure where 1.0 means predicted events match actual ones, was 1.03 in women and 0.94 in men3.
Why millions of scores fell, and why that isn’t simply good news
If the old tool ran high, a lower score may simply be more accurate, sparing some people a daily pill based on an inflated number. Anderson said people treated under the old tool shouldn’t conclude they were treated wrongly1.
But treatment rules had been written around the old numbers. A second 2024 team, writing in JAMA, ran PREVENT on 7,765 adults from the same national survey, who stood in for about 187 million Americans, and kept the old cutoffs4.
About 53% landed in a lower risk group, and just 0.41% in a higher one4. By our math, that’s roughly 99 million people whose risk group would drop on paper.
Under the old lines, 14.3 million fewer adults would qualify for a statin and 2.62 million fewer for blood pressure medicine4. The authors projected about 107,000 more heart attacks and strokes over 10 years if those people went untreated4. That figure came from a computer model, not from counting real events, and it assumed the cutoffs would stay the same4.
They didn’t. The 2026 cholesterol guideline from the American College of Cardiology, the American Heart Association and other groups adopted PREVENT and redrew the lines5:
| Risk group | Old cutoffs (2013 calculator)5 | 2026 cutoffs (PREVENT)56 |
|---|---|---|
| Low | Under 5% | Under 3% |
| Borderline | 5% to under 7.5% | 3% to under 5% |
| Intermediate | 7.5% to under 20% | 5% to under 10% |
| High | 20% or more | 10% or more |
By our math, the “high” line fell by half, about as much as the average score dropped in Anderson’s study. So a person whose estimate fell from 8.0% to 4.3% didn’t drop out of the conversation. Under the new rules, 4.3% is still “borderline”5.
The 2025 high blood pressure guideline also counts a PREVENT 10-year risk of 7.5% or more as one reason to start medicine at readings of 130/80 or higher7. Our blood pressure article has the details.
Not every expert body has switched. The Task Force’s 2022 statin advice still uses the old equations, with lines at 10% and 7.5% for adults 40 to 758. In fairness, the Task Force flagged the old tool’s tendency to overpredict, and its advice predates PREVENT38. Its page showed no update when we checked in October 20268.
Khan’s caution applies either way: whatever tool or cutoff is used, too few people at risk get preventive treatment2.
How the 2026 cholesterol guideline uses your number
The guideline applies PREVENT to adults 30 to 79 who haven’t had a heart attack or stroke5. Your 10-year group sets the starting point. Our article on aspirin and statins covers the cholesterol goals for each group.
For people in the borderline group, risk enhancers help decide whether medicine makes sense6. When the choice is still a toss-up, a coronary artery calcium scan, a quick CT picture of calcium in the heart’s arteries, can sharpen the estimate6. A later article in this series on heart screening tests explains what each calcium score means.
The score isn’t the whole story for some groups. In adults 40 and older with diabetes, chronic kidney disease at stage 3 or worse, or HIV, cholesterol-lowering treatment has shown benefit even when the estimated risk is lower, the heart association’s guideline summary says6.
The facts your calculator never asks about
PREVENT uses a short list of inputs, and many things that raise heart risk aren’t on it. The guideline calls them risk-improving factors, and clinicians weigh them on top of the score56. They include5:
- Family history: a parent or sibling with early heart disease, before 55 in a man or before 65 in a woman. Our companion article on family history explains why the details matter.
- Ancestry: higher-risk backgrounds, such as South Asian.
- Chronic inflammatory disease: lupus, rheumatoid arthritis, advanced psoriasis or inflammatory arthritis.
- Pregnancy and reproductive history: preeclampsia, gestational diabetes, high blood pressure in pregnancy, preterm delivery or early menopause.
- Lp(a): 125 nmol/L (50 mg/dL) or higher. Our cholesterol panel article explains this particle.
- Inflammation marker: hs-CRP of 2 mg/L or more on more than one test.
- Blood fats: triglycerides that stay high, or very high non-HDL cholesterol or apoB.
- Cardiovascular-kidney-metabolic syndrome, explained in the next section.
Some of these sit in your past, not your chart. A difficult pregnancy 25 years ago may never come up unless you mention it, so bring this list to your visit.
Heart, kidney and metabolic risk now travel together
In October 2023, an AHA advisory led by Johns Hopkins cardiologist Chiadi Ndumele gave the overlap of obesity, type 2 diabetes, kidney disease and heart disease a name: cardiovascular-kidney-metabolic (CKM) syndrome12. Its stages run from 0, no risk factors, to 4, heart disease with symptoms such as a past heart attack or heart failure12.
The advisory also called for a calculator that would start at age 30, add kidney function and heart failure, and give 30-year estimates12. PREVENT answered that call3.
In national survey data from 2011 to 2020, only 10.6% of U.S. adults were at stage 0, researchers reported in JAMA in 202413. About 15% were at stage 3 or 4, by our math13. A later article on metabolic syndrome takes up this overlap in depth.
Why a 40-year-old should ask for the 30-year number
Most U.S. adults rated low risk over 10 years are at high risk over the rest of their lives, researchers wrote in the New England Journal of Medicine in 201211. For younger adults, a 10-year window can miss most of the danger ahead.
That team pooled 18 long-term studies with 257,384 Black and White men and women11. It compared 55-year-olds with ideal numbers, meaning total cholesterol under 180, untreated blood pressure under 120/80, no smoking and no diabetes, with those who had two or more major risk factors11.
| Risk from age 55 | Ideal numbers | Two or more major risk factors |
|---|---|---|
| Men: death from heart or blood vessel disease by 80 | 4.7% | 29.6% |
| Women: death from heart or blood vessel disease by 80 | 6.4% | 20.5% |
| Men: lifetime fatal heart disease or nonfatal heart attack | 3.6% | 37.5% |
| Women: lifetime fatal heart disease or nonfatal heart attack | Under 1% | 18.3% |
Source: Berry and colleagues, 201211.
Put as absolute numbers, out of 1,000 men with two or more risk factors at 55, about 296 died of heart or blood vessel disease by 80. Out of 1,000 with ideal numbers, about 47 did, by our math11.
The 2026 guideline puts this to work. For adults 30 to 59 with a 10-year risk under 3%, the 30-year estimate helps steer the plan6. If it’s under 10% and LDL is below 160, the focus is lifestyle. If 30-year risk is 10% or more, or LDL is 160 to 189, a moderate-strength statin is reasonable to discuss6.
This is where natural-health values and risk science line up best. Many inputs that drive your number respond to how you live: blood pressure, cholesterol, blood sugar and smoking. The Berry data show what decades of good numbers are worth, and for many younger adults, the guideline’s first move is habits, not pills6.
Your risk number, decoded
Use this table, built from the sources above, when you get your score, and ask your clinician to change one input and rerun the calculator.
| Input | How it feeds your estimate | Can you change it? | What to ask |
|---|---|---|---|
| Age and sex | Core inputs; separate equations for men and women3 | No | “What’s my 30-year risk?”6 |
| Top blood pressure number | Higher readings raise risk310 | Yes, with habits and, if needed, medicine | “Was this from one reading or an average?” |
| Cholesterol | Part of every estimate3 | Often | “Should we check my Lp(a)?”5 |
| Smoking | Current smoking raises risk310 | Yes | “How much would quitting change my number?” |
| Diabetes and A1C | Diabetes is a core input; A1C is optional3 | Often | “Do I have prediabetes?” |
| Kidney function (eGFR, from a blood test) and urine albumin | eGFR is core; urine albumin is optional3 | Partly | “Has my urine albumin ever been checked?” |
| Blood pressure or statin medicine | The equations ask whether you take them3 | Not a lever to pull on your own | “How does my treatment affect the estimate?” |
| ZIP code | Optional social index, a stand-in for neighborhood conditions3 | Not directly | “Does adding it change my category?” |
Five questions to ask when you get a risk score:
- Which calculator did you use, PREVENT or the older equations58?
- What’s my 10-year risk, and what’s my 30-year risk6?
- Which risk enhancers apply to me, including my family and pregnancy history5?
- Would a calcium scan change what we do6?
- If I take a medicine, how many people like me out of 100 avoid a heart attack or stroke? Our article on absolute risk shows how to read the answer.
What the evidence shows
| Claim | Evidence rating | What it rests on |
|---|---|---|
| Risk calculators sort large groups by their real risk | PREVENT tested in 3.3 million adults; predicted and actual events closely matched3 | |
| The 2013 calculator overestimated risk in today’s adults | Many validation studies, noted by the Task Force; about twofold, per PREVENT’s lead developer28 | |
| Adding A1C, urine albumin and ZIP code makes most people’s estimate much more accurate | Tested in the PREVENT data: tiny average gain, though urine albumin helped people with high levels3 | |
| Fewer risk factors in midlife go with much lower lifetime risk | 18 pooled studies of 257,384 people; observational11 | |
| Switching calculators without new cutoffs would add about 107,000 events | A model, not observed events; cutoffs have since changed45 | |
| A score tells you whether you will have a heart attack | Scores describe groups; predictions are estimates8 |
Questions readers ask
My new score is half my old one. Should I stop my statin?
Not on your own. When the comparison studies came out, Khan cautioned that they were snapshots that couldn’t guide treatment until new guidelines arrived2. The 2026 guideline has since redrawn the categories, so ask your clinician to rerun your numbers under them and weigh your risk enhancers56.
Why does the calculator want my ZIP code?
It’s optional. The ZIP code feeds a social deprivation index, a rough measure of neighborhood hardship that replaced race in the equations35. On average, the optional extras, including this one, add very little accuracy3.
I’m 35 and my 10-year risk is under 3%. Can I ignore this?
At 35, a low 10-year number partly reflects your age. Ask for your 30-year estimate. Under the 2026 guideline, a 30-year risk of 10% or more opens a discussion about a statin, while a lower one, with LDL under 160, points to habits as the main focus6.
Is this the same as Life’s Essential 8?
No. Life’s Essential 8 is a separate heart association scorecard for daily habits and health numbers, covered in our article on sleep as a prevention habit. A calculator estimates your odds of an event; the scorecard grades each habit.
The bottom line
A heart risk score is a forecast for people with your numbers, not a verdict on you. PREVENT gives many adults lower estimates than the 2013 tool, largely because the old one ran high. The 2026 cholesterol guideline lowered its cutoffs to match, so a lower score doesn’t automatically mean less care.
Know your 10-year and 30-year numbers, bring your family and pregnancy history, and ask which calculator was used. Much of what moves the estimate is yours to change. Make any medicine change with your clinician.
Related reading
About this article
This article was researched using peer-reviewed studies, AHA and ACC guideline materials, U.S. Preventive Services Task Force recommendations and federal health data, all listed below. It was reviewed in October 2026. It is general education, not medical, legal or financial advice.
Sources
- UPMC. “Statins for Heart Disease Prevention Could Be Recommended for Far Fewer Americans if New Risk Equation Is Adopted” (on Anderson TS, Wilson LM, Sussman JB, JAMA Internal Medicine, 2024;184(8):963–970). News release, June 10, 2024. upmc.com
- TCTMD. “Think You Need a Statin? AHA’s Risk Calculator May Disagree.” Cardiovascular Research Foundation, 2024. tctmd.com
- Khan SS, Matsushita K, Sang Y, et al.; Chronic Kidney Disease Prognosis Consortium and the American Heart Association Cardiovascular-Kidney-Metabolic Science Advisory Group. “Development and Validation of the American Heart Association’s PREVENT Equations.” Circulation, 2024;149(6):430–449 (Johns Hopkins University research record). pure.johnshopkins.edu
- Diao JA, Shi I, Murthy VL, et al. “Projected Changes in Statin and Antihypertensive Therapy Eligibility With the AHA PREVENT Cardiovascular Risk Equations.” JAMA, 2024;332(12):989–1000. pubmed.ncbi.nlm.nih.gov
- Eberly L (Gallup Indian Medical Center). “Guideline Essentials: 2026 ACC/AHA Guideline on the Management of Dyslipidemia” (slides adapted from AHA clinical materials, citing Blumenthal RS, Morris PB, et al., Circulation, 2026). Indian Health Service, California Area, 2026. ihs.gov
- Khan SS, Eaton M. “Top Take-Home Messages for Clinicians: Using PREVENT-ASCVD Equations for Risk-Based Lipid Management,” adapted from the 2026 ACC/AHA Guideline on the Management of Dyslipidemia. American Heart Association, 2026. professional.heart.org
- American Heart Association. “Top Things to Know: 2025 High Blood Pressure Guideline.” AHA Professional Heart Daily; guideline published August 14, 2025, in Circulation and Hypertension. professional.heart.org
- U.S. Preventive Services Task Force. “Statin Use for the Primary Prevention of Cardiovascular Disease in Adults: Preventive Medication.” Final recommendation statement, August 23, 2022. uspreventiveservicestaskforce.org
- Centers for Disease Control and Prevention. “Heart Disease Facts.” CDC, updated October 24, 2024. cdc.gov
- National Institutes of Health. “The Framingham Heart Study: Laying the Foundation for Preventive Health Care.” NIH Impact fact sheet (posted January 2025). nih.gov
- Berry JD, Dyer A, Cai X, et al. “Lifetime Risks of Cardiovascular Disease.” New England Journal of Medicine, 2012;366:321–329 (NIH Public Access author manuscript, hosted by the European Society of Hypertension). eshonline.org
- American Heart Association News. “Report Redefines Overlapping Risks of Heart and Kidney Diseases” (on Ndumele CE, et al., presidential advisory, Circulation, 2023). October 9, 2023. heart.org
- Brooks A. “Nearly 90% of US Adults Meet Cardiovascular-Kidney-Metabolic Syndrome Criteria” (on Aggarwal R, Ostrominski JW, Vaduganathan M, JAMA, 2024). HCPLive, May 13, 2024. hcplive.com
- Centers for Disease Control and Prevention. “About Heart Attack Symptoms, Risk, and Recovery.” CDC, updated October 24, 2024. cdc.gov
- Centers for Disease Control and Prevention. “Signs and Symptoms of Stroke.” CDC, updated May 19, 2026. cdc.gov