QRISK3 (10-Year CVD Risk)

About QRISK3 (10-Year CVD Risk)

QRISK3 is the current generation of the QRISK cardiovascular risk algorithm developed by Hippisley-Cox and colleagues at the University of Nottingham and published in the BMJ in 2017. It is the standard risk calculator used by the National Health Service in England and Wales and is embedded in the major UK primary care electronic health record systems (EMIS Web, SystmOne, Vision). QRISK3 supersedes QRISK2 (2010) by incorporating eight additional risk factors: severe mental illness, rheumatoid arthritis (better characterized than in QRISK2), systemic lupus erythematosus, regular oral steroid use, atypical antipsychotic use, erectile dysfunction (in men), migraine, and systolic blood pressure variability. The algorithm was derived from the QResearch database of over 7.8 million UK patients and validated in a separate cohort of 2.6 million patients.

What the calculator measures

The output is a 10-year probability of a first cardiovascular event (MI, angina, coronary revascularization, ischemic stroke, TIA, subarachnoid hemorrhage, peripheral artery disease, abdominal aortic aneurysm, or cardiovascular death). The NICE NG181 lipid modification guideline recommends statin therapy for individuals with a QRISK3 score of 10% or greater, after a clinician-patient discussion of the benefits and risks.

How the formula works

QRISK3 is a Cox proportional hazards model with fractional polynomial transformations of continuous predictors (age, BMI, total/HDL cholesterol ratio, systolic blood pressure) to allow nonlinear dose-response relationships. The model includes 16 risk factors: age, sex, self-assigned ethnicity (9 categories), Townsend deprivation score, systolic blood pressure and its variability (standard deviation across measurements), BMI, total/HDL cholesterol ratio, smoking status (5 categories), type 1 and type 2 diabetes (separate indicators), family history of early CHD in a first-degree relative <60, chronic kidney disease (GFR <60 or proteinuria), atrial fibrillation, migraine, rheumatoid arthritis, SLE, severe mental illness (schizophrenia or bipolar disorder), erectile dysfunction (men only), systemic corticosteroid use, atypical antipsychotic use, and hypertension treatment. The published coefficients are extensive; the simplified JavaScript implementation here uses representative coefficient values that approximate the full QRISK3 algorithm for educational use. For clinical decisions, the official QRISK3 online calculator (qrisk.org) is the validated tool.

Interpretation

NICE recommends statin therapy (atorvastatin 20 mg daily) for adults with a QRISK3 score of 10% or greater. The 10% threshold was chosen because, at that level, approximately 100 people would need to be treated for 10 years to prevent one cardiovascular event, a number that NICE judged cost-effective given the low cost of generic atorvastatin. Patients with established CVD, familial hypercholesterolemia, or a QRISK3 >10% who decline statins should be offered lifestyle advice and re-evaluated periodically. Patients with QRISK3 <10% should receive lifestyle advice and risk factor monitoring, with re-screening every 5 years (or more often if risk factors change).

Clinical use

QRISK3 is the primary tool for cardiovascular primary prevention in the UK NHS and is also widely used in other countries with national health systems (Ireland, parts of the Middle East, Australia in some settings). It is uniquely valuable for patients whose risk is influenced by conditions that other scores ignore, chronic inflammatory disease (rheumatoid arthritis, SLE), severe mental illness, chronic steroid use, atypical antipsychotic use, and erectile dysfunction. Each of these factors independently increases CVD risk by 15-50% and can reclassify patients from low to high risk.

Limitations

QRISK3 was derived in a UK population and may not be calibrated for non-UK populations; international recalibration factors exist for some countries (e.g. New Zealand, Saudi Arabia). The model requires multiple inputs (16 risk factors), which can be cumbersome in clinical practice but is feasible when integrated into the electronic health record. The model uses Townsend deprivation, a UK-specific measure of socioeconomic status; in other countries, alternative deprivation indices must be substituted. The model excludes patients with established CVD and patients younger than 25 or older than 84.

Comparison with ASCVD and Framingham

QRISK3 includes more variables and more contemporary risk factors than the ASCVD PCE (which has 8-9 inputs) or Framingham (which has 7-8). It generally gives somewhat lower absolute risk estimates than ASCVD for white patients because of the lower baseline CVD rate in the UK QResearch cohort compared with the U.S. cohorts that derived the PCE. QRISK3's strength is its inclusivity: it captures risk in patients with chronic illness, severe mental illness, inflammatory disease, and socioeconomic disadvantage, populations whose risk is systematically underestimated by simpler scores.

How to use the result

A QRISK3 score of 12% means approximately 12 out of 100 people with similar risk factors will have a cardiovascular event in the next decade. Per NICE, this score indicates statin therapy after a clinician-patient discussion of the absolute benefit (about 1 in 100 chance of preventing an event over 10 years), the risks (myalgia in ~5%, diabetes in ~1%, liver enzyme elevation in <1%), and the alternatives (intensive lifestyle therapy, blood pressure control, smoking cessation).