Heart disease risk: why the calculator is just the start

Clinical Rounds, the GPAtlas podcast. Cardiovascular. 20 minutes. Published 18 September 2026. Free to listen.

From our Ocean topic Cardiovascular Disease (CVD) Risk Assessment and Management, reviewed August 2026.

Transcript

Sarah: Hello, and welcome to GPAtlas Clinical Rounds. I'm Sarah, and this is the podcast where we take one thing you'll genuinely see this week and work out what to do with it.

Ben: And I'm Ben. Today it's cardiovascular risk, and the ten-year score we all print off and read out to patients. The number isn't the answer, it's where the consultation starts.

Sarah: So. Amar is forty-two, and his parents came here from Lahore. He's booked in with his Health Check letter in his hand, and his ten-year score is eight, which the letter calls lower risk. He stopped smoking five months ago, and his dad had a heart attack at fifty- two. He sits down, smiles, and says, so I'm fine then, yeah? And on a Tuesday with three extras, the easy answer is yes: under ten, keep walking, see you in five years.

Ben: Three things today. Who shouldn't be on the calculator at all. Who the calculator lets down, and what you do instead of trusting the number. And the conversation and the follow- up that turn a score into a genuinely lower risk.

Sarah: So before I answer him, here's what struck me. He only got a score because he turned forty, and if he'd been thirty-eight, nobody would have invited him. With a dad who had a heart attack at fifty-two, that bothers me.

Ben: And our article names that as a common pitfall, because people conflate two different age windows. The Health Check invites everyone aged forty to seventy-four without established cardiovascular disease, diabetes or chronic kidney disease, every five years. But the calculator itself, Q-risk three, is validated from twenty-five to eighty-four. So a thirty-year-old with the right history is fair game.

Sarah: And the guideline doesn't want us waiting for people to turn up with a letter, it wants them found systematically. So who goes to the top of that search?

Ben: Our article lists them. A first-degree relative with cardiovascular disease before sixty, a smoking history, hypertension, obesity or central adiposity, and South Asian family origin. Then two that get forgotten: a chronic inflammatory condition like rheumatoid arthritis or lupus, and severe mental illness. Amar ticks three of those before you've even touched a cuff.

Sarah: The Health Check is forty to seventy-four, but the calculator works from twenty-five to eighty-four. Go looking for high-risk people on purpose.

Ben: But before you trust any score, our article asks a blunter question: should this person be on the calculator at all? Some groups are high risk by definition, and it names three. Type one diabetes. Chronic kidney disease, CKD, meaning an estimated glomerular filtration rate, or e-GFR, under sixty, or albuminuria. And familial hypercholesterolaemia, FH. In all three, you don't run the score, you offer a statin directly.

Sarah: And I've seen that go wrong. Someone with CKD gets a score run opportunistically, it comes back at six, and they're told they're low risk. But that number should never have existed, because they were high risk before anyone typed a thing.

Ben: And each one has its own route. In CKD it's atorvastatin twenty milligrams, with renal advice if the e-GFR is under thirty. In type one diabetes it's atorvastatin twenty milligrams if they're over forty, diabetic for over ten years, or have nephropathy or other risk factors. And eighty-five and over is high risk through age alone, so you consider atorvastatin twenty milligrams, weighing frailty and comorbidity.

Sarah: For Amar, his kidney function's normal and there's no diabetes. But his dad still nagged at me, so I looked at his total cholesterol, which was six point eight. Then I got his socks off and felt his Achilles tendons, and he thought I'd lost the plot.

Ben: You hadn't, because that's possible FH you're ruling out. Our article's trigger is a total cholesterol above seven point five plus a family history of premature coronary heart disease, or tendon xanthomata, which are specific for FH. Either one means the lipid clinic, not the calculator. There, treatment is specialist-guided, aiming to more than halve the low-density lipoprotein cholesterol, the LDL.

Sarah: Type one diabetes, CKD or FH: don't run the score. A young family history means check the cholesterol and feel the tendons.

Ben: So he's a calculator patient, and the next question is whether his number was built on the right inputs. Because some of what goes into the score never gets asked about in a busy Health Check.

Sarah: And this is the one I nearly didn't ask. I said, any trouble with erections? There was a long pause, and then, actually, yes, for about a year. He'd not told anyone, not even his wife, and it wasn't in his score.

Ben: And our article is clear that erectile dysfunction is an independent predictor of cardiovascular disease, and a Q-risk three input. Regular corticosteroids and atypical antipsychotics raise risk and are captured by the score too. Ethnicity matters as well, because South Asian origin raises risk and adjusts the number. Get those boxes wrong and the number's wrong.

Ben: Ask about erections, steroids and antipsychotics, and check the ethnicity. A missing input is a wrong score.

Sarah: So I re-ran it, and it came out just over nine. Still under ten. Then I did the symptom questions, because they're easy to forget in a prevention appointment. Chest pain, breathlessness, palpitations, ankle swelling, leg pain on walking that eases with rest, and any brief weakness, speech trouble or loss of vision.

Ben: Because if any of that is active, it isn't prevention any more. Rest or unstable angina, or a recent transient ischaemic attack or stroke, means urgent same-day assessment or admission. And with signs of heart failure, so breathlessness, peripheral oedema or a raised jugular venous pressure, you arrange assessment.

Sarah: He had none of it. Then blood pressure in both arms, as our article asks at a first assessment, and it was one thirty-eight over eighty-six with a regular pulse. But if I'd got a hundred and eighty over a hundred and twenty, I'd have been looking in his eyes.

Ben: Right, because a hundred and eighty over a hundred and twenty with retinal haemorrhage or papilloedema, or with life-threatening symptoms like new confusion, chest pain, heart failure or acute kidney injury, is accelerated hypertension. That's same-day specialist assessment. So is suspected phaeochromocytoma: labile or postural blood pressure with headache, palpitations, pallor and sweating. And the pulse matters, because it's a chance to pick up atrial fibrillation.

Sarah: Ask the symptom questions every time. Active ischaemia, accelerated hypertension or possible phaeochromocytoma means same day, not a statin chat.

Ben: Now the bit that matters most for Amar. Even with the right inputs, our article lists the people the calculator underestimates. That's people on treatment for human immunodeficiency virus, HIV, severe mental illness, autoimmune or inflammatory disease, and anyone already on blood pressure or lipid-lowering drugs. And the recently stopped smoker.

Sarah: Five months off cigarettes, so that's him. Just over nine, in a recently stopped smoker whose dad had a heart attack at fifty-two, is not the reassurance it looks like. And I'd been about thirty seconds from telling him he was fine.

Ben: And our article is blunt that clinical judgement overrides the number. Under ten doesn't rule out a statin. Atorvastatin twenty milligrams stays on the table if there's an informed preference, or if the risk may be underestimated. Ten is where you offer, but it isn't a wall.

Ben: A low score in the wrong patient isn't reassurance. Under ten doesn't rule out a statin.

Sarah: And here's what actually moved him. Nine in a hundred over ten years sounds like nothing when you're forty-two. He's hearing that ninety-one times out of a hundred he'll be fine. So I didn't argue with the ten-year number, I changed the time frame.

Ben: That's the clinical pearl in our article, that a ten-year score under-sells risk in the young. Their example is a thirty-five-year-old smoker with a strong family history, whose ten-year figure looks reassuringly low while the lifetime risk is alarming. So you reach for a lifetime-risk tool, because it communicates risk far better than a single ten-year number.

Sarah: And it did. His lifetime figure was a different conversation, and he went quiet, then said, so it's not if, it's when. That isn't right either, and I said so. But for the first time we were talking about his heart, not his letter.

Ben: In the young, the ten-year score under-sells. A lifetime-risk tool gets you the honest conversation.

Sarah: So then he asked the obvious thing: would a statin actually help me? This is where I used to say it lowers your risk and leave it there. Which tells him nothing about what it would do for him.

Ben: Our article wants absolute risk and absolute benefit, and it gives you the numbers. Over ten years, at a risk of five in a hundred, you treat about fifty-four people to prevent one event. At ten in a hundred it's twenty-five, and at twenty in a hundred it's fourteen. That's what you show the patient.

Sarah: So at just over nine, he's close to the twenty-five mark, and that's if the score is right, which in a recently stopped smoker it may not be. Then I asked what worried him about a tablet every day. His uncle had muscle aches on one, which is a fair worry.

Ben: And two things to say plainly. Aspirin isn't used for primary prevention. And if he'd said no to the statin, that's fine, because our article says you simply offer reassessment later. Either way, you document the discussion and the decision, because it's a shared decision, not an automatic prescription.

Sarah: Give absolute numbers: twenty-five people treated for ten years, at ten in a hundred. No aspirin, and write it down.

Ben: Whatever the score, lifestyle advice is for everyone, and our article is clear that stopping smoking is the single most effective intervention. Then there's diet, a Mediterranean-style pattern. That's olive or rapeseed oil instead of saturated fat, five or more fruit and vegetables a day, oily fish and less salt.

Sarah: He's already done the hardest bit, so my job is keeping him stopped. I offered the stop smoking service and medication support, and he took both. He says he walks the dog. Then he asked about the cholesterol spread his wife buys, the one with plant sterols in it.

Ben: And our article says don't recommend plant sterols or stanols, fish oil or vitamin supplements. Put the time into what works instead. That's a hundred and fifty minutes of moderate activity a week, or seventy-five vigorous, plus muscle-strengthening on two or more days. And alcohol at fourteen units a week or less, spread over three or more days.

Sarah: Lifestyle for everyone, whatever the number, and stopping smoking beats everything. Leave the sterols and the fish oil on the shelf.

Sarah: So he chose the statin, atorvastatin twenty milligrams, and it's bloods first. I'll confess that for years I booked everyone a fasting lipid test, out of pure habit. He works night shifts at a warehouse, so fasting would have cost him a morning's sleep.

Ben: And our article says you don't need it. It's a full lipid profile, non-fasting, with a fasting sample only if triglycerides are above four point five. Then watch for two results that change the route. Total cholesterol above nine, or non-high-density-lipoprotein cholesterol, the non-HDL, above seven point five, means specialist assessment whatever the family history. And triglycerides above twenty, not explained by alcohol or poor glycaemic control, need urgent specialist review for pancreatitis risk.

Sarah: Nothing in his lipids changed the route. I also sent kidney and liver function, thyroid, and a glycated haemoglobin, the HbA1c. And because of his uncle, he asked for a muscle test, but he's got no muscle symptoms himself.

Ben: Then he doesn't need it. Creatine kinase, or CK, is only for unexplained muscle symptoms before you start. Over five times the upper limit of normal, you don't start, and if it's raised but under five times, you start a lower dose. The thyroid test is there to exclude hypothyroidism as a secondary cause of dyslipidaemia, and the liver tests give you a baseline.

Sarah: No fasting for lipids unless triglycerides are over four point five. And only check CK for unexplained muscle symptoms.

Sarah: He came back at three months, and his non-HDL was down, but not by much. When I asked properly, he'd been missing it most weekends. Then he said, and my sugar's gone up, hasn't it? He'd read that statins cause diabetes, and he wanted to stop.

Ben: Our article covers both of those. The target at three months is a fall in non-HDL of more than two fifths, and if it's not there, you check adherence before you escalate. And a small rise in glucose, or even new type two diabetes, is not a reason to stop the statin.

Sarah: So we didn't touch the dose. We talked about the weekends, and he set an alarm on his phone. Then his liver tests were slightly up, and he'd been told statins damage the liver, so that was his next question.

Ben: Liver tests are repeated at three and twelve months, and you stop only if transaminases are persistently above three times the upper limit of normal. If he's taking it properly and still isn't at target, or can't tolerate a statin, there's ezetimibe ten milligrams, as an add-on or alone, or bempedoic acid with ezetimibe. Inclisiran, evolocumab and alirocumab are specialist options. And he gets an annual medication review.

Sarah: At three months, check adherence before you escalate. A small sugar rise is not a reason to stop the statin.

Ben: Right, exam corner. A thirty-eight-year-old man comes in about a cholesterol result. His total cholesterol is eight point one, and his brother had a heart attack at fifty-five. Someone has already run his ten-year score, and it's four. What's the most appropriate next step? A, lifestyle advice and reassess in five years. B, a fasting lipid profile before deciding. C, refer to the lipid clinic as possible familial hypercholesterolaemia. Or D, aspirin and lifestyle advice.

Ben: C. A total cholesterol above seven point five with a family history of premature coronary heart disease is possible FH, so it's the lipid clinic, and the calculator shouldn't have been used. A is the tempting one, because four looks so reassuring, but that number should never have been run. B is wrong because a fasting sample only matters if triglycerides are above four point five. And aspirin isn't used for primary prevention, so not D.

Sarah: So. Three things for Monday. One. Before you trust a score, check the person belongs on the calculator, because type one diabetes, CKD, FH and the over-eighty-fives don't. Two. When the number's low, ask who it underestimates: the recently stopped smoker, inflammatory disease, severe mental illness, HIV treatment, anyone already on blood pressure or lipid drugs. And ask about erections. And three. When you start a statin, book the three-month lipids, and check adherence before you change anything.

Sarah: One thing to reflect on, if you're logging this. When you last reassured someone with a score under ten, were they in a group the calculator underestimates? Check your local guidance before you act on it. The full transcript and the sources are on the episode page. Ben and I are natural AI voices, and what we've said comes from our reviewed Ocean topic. Check your local guidance before you act on it. See you next week.

More from Clinical Rounds

Every episode of Clinical Rounds, free to listen with a full transcript.