Coronary artery calcium scans have become easy to book and easy to pay for out of pocket. Over the weekend of 13 September 2026, science media picked up a Northwestern Medicine analysis published in JAMA that asks a blunt question: does the scan actually tell you anything your doctor does not already know? The ten-year answer reframes the whole conversation. For most adults, the numbers already sitting in a routine laboratory report carry most of the predictive weight. The scan changes the picture for one specific group, and blood work is what tells you whether you belong to it.
What the new JAMA analysis found
Researchers followed 6,098 adults aged 45 to 79 for a decade inside the Multi-Ethnic Study of Atherosclerosis. Each participant had a coronary calcium score measured at the start, and each had a ten-year cardiovascular risk estimate produced by PREVENT, the American Heart Association risk equations. Over those ten years, 366 people, roughly 6 in 100, had a heart attack or a stroke.
Adding the calcium score on top of the PREVENT estimate improved the accuracy of the prediction only slightly across the group as a whole. In some risk categories it did not improve it at all. For a test that involves a CT scanner, a small dose of radiation and a bill, that is a meaningful finding.
The exception is the interesting part. Among people whose PREVENT estimate put them in the borderline band, the calcium score sorted them convincingly into higher and lower real-world risk.
Ten-year event rates in the borderline group
| Calcium score | Heart attack or stroke within 10 years |
|---|---|
| 0 | About 2 people in 100 |
| Above 0 but under 100 | About 4 people in 100 |
| 100 to under 300 | About 7 people in 100 |
| 300 or above | About 14 people in 100 |
Same starting estimate, a sevenfold spread in what actually happened. That is why the 2026 dyslipidaemia guideline from the American College of Cardiology, the American Heart Association and partner societies recommends the scan selectively, for people whose estimated ten-year risk sits between 3 per cent and just under 10 per cent, rather than for everyone who asks.
Your risk estimate starts in the laboratory
PREVENT is not an imaging tool. It is built almost entirely out of measurements a laboratory produces, plus your age, sex, blood pressure and smoking status. Our team explains the underlying measurements in a guide to the complete lipid panel.
| Input | Where it comes from |
|---|---|
| Total cholesterol and HDL cholesterol | Lipid panel, blood draw |
| Estimated glomerular filtration rate (eGFR) | Creatinine, blood draw |
| Diabetes status | Glucose and HbA1c history |
| Systolic blood pressure, age, sex, smoking, current medication | Clinic visit |
| Optional refinements: HbA1c, urine albumin-to-creatinine ratio | Blood and urine samples |
Two things follow from that list. First, a calcium scan cannot be interpreted sensibly until the blood work exists, because the blood work is what places you in a risk band. Second, PREVENT deliberately includes kidney and metabolic health, which is why creatinine and urine albumin are no longer side notes in a cardiovascular workup. Our team covers that shift in an article on kidney function tests for heart patients.
Where the scan earns its place, and where it does not
The logic is about whether the result could change anything.
- Low estimated risk: a scan mostly confirms what the laboratory already suggested, whilst adding radiation, cost and a chance of incidental findings that lead to further testing.
- Borderline to intermediate estimated risk: this is the genuine grey zone, where a score of zero and a score above 300 point towards different conversations about preventive treatment.
- High estimated risk: preventive treatment is generally recommended whatever the scan shows, so the scan rarely changes the plan.
The study authors made the same point directly: scanning people who are already clearly low risk or clearly high risk tends to generate testing rather than answers.
Reading your own results before booking anything
If you are wondering whether a scan is for you, the honest first step is your most recent laboratory report, not a booking page.
- Find your total cholesterol and HDL cholesterol values, and the date they were measured.
- Check whether LDL cholesterol was calculated or measured directly. Our team details how the number behaves in a guide to LDL cholesterol levels and ranges.
- Look for creatinine and eGFR. If eGFR is missing, ask for it, because PREVENT needs it.
- Note your blood pressure, whether you smoke, and whether you already take a statin or a blood pressure medication.
- Bring all of it to the appointment and ask for your estimated ten-year risk before discussing imaging.
If several values look unfamiliar, our team walks through the broader context in an overview of high cholesterol and what to do about it.
Latest scientific advances
According to PubMed and recent peer-reviewed literature, the JAMA result fits a pattern rather than contradicting one.
Earlier work comparing the calcium score with a genetic risk score in two large population cohorts, one American and one Dutch, found that the calcium score was the stronger of the two at sorting people by future coronary risk, and that it did move some people into a more accurate risk category. A 2024 European study reached a similar conclusion and added a practical note: the gain was clearest in people whose starting estimate was already in the middle of the range. What that means for you is simple. The scan is a refinement tool, not a screening tool, and it refines best when there is genuine uncertainty to resolve.
A 2026 review of cardiovascular risk assessment made the broader point. A growing list of extras, including apolipoprotein B, lipoprotein(a), genetic scores and calcium scores, can each reclassify some people, but none of them has yet been shown to improve outcomes when used routinely. The review's conclusion, in plain terms: these should guide a conversation in selected patients, not replace the basic assessment for everyone. Research in this field is ongoing, and none of it changes the fact that the decision belongs to you and your doctor.
When to talk to a doctor
Book an appointment rather than a scan if you have a family history of early heart disease, if your cholesterol results have shifted noticeably, if your eGFR has fallen, or if you have simply never had your cardiovascular risk formally estimated. Seek urgent care for chest pain or pressure, sudden shortness of breath, or sudden weakness or difficulty speaking, which need emergency assessment and not a risk calculator.
Glossary
- Coronary artery calcium score: a number from a CT scan reflecting how much calcified plaque sits in the heart’s arteries.
- PREVENT equations: the American Heart Association tool that estimates ten-year and thirty-year cardiovascular risk from routine clinical and laboratory data.
- eGFR: estimated glomerular filtration rate, a calculation based on blood creatinine that reflects how well the kidneys filter.
- Borderline risk: in the 2026 categories, an estimated ten-year risk of 3 per cent to under 5 per cent.
- Reclassification: moving a person into a different risk category once new information is added.
Frequently asked questions
Does a calcium score of zero mean I am safe?
It means your ten-year risk is lower than your starting estimate suggested, not that it is zero. In the borderline group studied, about 2 people in 100 with a score of zero still had an event within ten years. A score of zero also says nothing about non-calcified plaque, which does not show up on this scan.
Should I ask for a calcium scan at my next check-up?
Ask for your risk estimate first. If it falls in the borderline to intermediate band, the scan is a reasonable topic. If it falls clearly below or above that band, the scan is unlikely to change what happens next.
Can a blood test replace the scan entirely?
No, they measure different things. Blood tests measure circulating risk factors, whilst the scan looks at damage that has already accumulated. The study simply shows that for most people the blood-based estimate is already doing the heavy lifting.
Do I need to fast before the blood tests used in the risk estimate?
Guidance varies by laboratory and by which tests are ordered. Many lipid panels are now accepted without fasting, but follow the instructions your laboratory or doctor gives you.
How often should the risk estimate be repeated?
For most adults without known cardiovascular disease, every few years is typical, and sooner if results change or a new risk factor appears. Your doctor sets the interval based on your profile.
Sources
- Predictive Utility of Coronary Artery Calcium Score Added to the PREVENT Atherosclerotic Cardiovascular Disease Equations, JAMA, 2026
- American Heart Association: about the PREVENT calculator
- Northwestern University via ScienceDaily: who really needs a heart calcium scan?
- Coronary Artery Calcium Score and Polygenic Risk Score for the Prediction of Coronary Heart Disease Events, JAMA, 2023
- Predictive improvement of adding coronary calcium score and a genetic risk score to a traditional risk model, European Journal of Preventive Cardiology, 2024
- Cardiovascular Risk Assessment: A Review of Current Models and Emerging Determinants, Current Cardiology Reviews, 2026
Further reading
- Lipid panel explained: cholesterol, LDL, HDL and triglycerides
- LDL cholesterol: levels, ranges and risks
- Kidney function test for heart patients: what the new guidelines change
- High cholesterol: symptoms, causes, numbers and treatment
Understand your lab results with AI DiagMe
Your cholesterol, creatinine and eGFR values decide whether a calcium scan would tell you anything new. If those numbers mean little to you as they stand, upload your report and get a clear, plain-language reading of every marker, reviewed by a medical committee. Start with AI DiagMe.
