Cardiovascular risk: how it is actually calculated

A risk score is a population statement pointed at you

Cardiovascular risk calculators take a handful of your measurements and return something like "14% risk over ten years". What that actually means is: of a hundred people in the reference population who look like you on these variables, about fourteen had an event within ten years.

It does not say which fourteen. It cannot tell you what will happen to you, and no refinement of it ever will.

What goes in

The major calculators — the ACC/AHA pooled cohort equations, the European SCORE2, the UK's QRISK3 — differ in detail but share a core:

• Age, which dominates the result more than anything else
• Sex
• Total and HDL cholesterol
• Systolic blood pressure, and whether it is treated
• Smoking
• Diabetes

QRISK3 adds a longer list — among them chronic kidney disease, atrial fibrillation, rheumatoid arthritis, severe mental illness, and family history. SCORE2 was recalibrated for four European risk regions, because the same inputs predict differently in different populations.

The problem for anyone reading this in Costa Rica

This deserves to be stated plainly rather than buried. These equations were derived in specific populations — largely North American and European cohorts — and their accuracy depends on the population they are applied to.

A calculator built on one population can systematically over‑ or under‑estimate in another, which is precisely why SCORE2 has separate calibrations by region and why several countries have built their own. The WHO has published region‑specific charts, including for Latin America.

We have not been able to establish that a calculator validated specifically in a Costa Rican population exists. That does not make the scores useless here — they still rank risk sensibly — but the absolute percentage should be held loosely.

What age does to the result

Age is the strongest input, and that produces two systematic distortions worth knowing:

• Young people with genuinely bad numbers score low, because ten‑year risk is low at 35 almost regardless. This is why some guidelines use lifetime risk alongside it.
• Older people score high almost automatically, which can make the number uninformative about anything modifiable.

The modifiers, for when the score sits on a line

Where a result falls near a treatment threshold, several things are used to refine it. They are refinements, not replacements:

• Coronary artery calcium score — a CT scan measuring calcified plaque. The strongest reclassifier available: a score of zero in an intermediate‑risk person argues strongly against, a high score strongly for. Involves radiation and cost.
• Lipoprotein(a) — largely genetically determined and measured once in a lifetime. Increasingly recommended at least once.
• ApoB or non‑HDL, where the standard lipid panel is discordant.
• hs‑CRP, for intermediate cases.
• Family history of premature cardiovascular disease.

What the scores do not include

A longer list than most people assume, and it is the reason the number is not the whole story: diet, physical fitness, sleep, visceral fat specifically, insulin resistance before diabetes appears, fatty liver, and most measures of social circumstance.

A fit, lean person and a sedentary one with identical cholesterol, pressure and age receive the same score.

How to use one sensibly

1. Treat it as a prompt for a conversation, not a verdict.
2. Ask which calculator and which population it was built on.
3. Notice whether your inputs are current. A score built on a blood pressure from two years ago is a score about two years ago.
4. Ask what it would take to change it — and notice how much of the movement comes from things you can act on versus from your age.
5. Where it sits near a threshold, ask what would refine it rather than accepting the midpoint.

The boundary

A percentage is an input to a decision about treatment, not the decision. What yours means, which calculator suits you, and whether anything should change rests with a physician who can weigh it against the rest of your history.