Heart disease risk ‘calculator’ could save lives

risk algorithm
Grant Muller

UK researchers say they have developed an algorithm that is more accurate than current tools at predicting an individual’s risk of developing cardiovascular disease in the next 10 years.

The QR4 algorithm, described in the journal Nature Medicine, can be thought of as a “calculator” that takes into account health data like blood pressure, age, and medical history to give a measure of risk and allow early intervention if needed.

So-called QRISK algorithms are already widely used to help clinicians spot at-risk patients, but the team behind QR4 – led by scientists at the University of Oxford – have layered in new variables that have improved its accuracy and could identify candidates for preventive and treatment measures that others miss.

That includes seven new risk factors that apply to all adults, including chronic obstructive pulmonary disease (COPD), learning disabilities, Down syndrome, and four cancer types – blood, lung, oral, and brain – showing how other diseases can have an impact on heart health.

Two new risk factors that apply only to women - namely, complications from high blood pressure during pregnancy and postnatal depression - have also been included for the first time.

Current tools don’t include these factors, so may underestimate risk, which means patients may miss out on interventions that could improve their chances of survival, as well as the opportunity to take part in clinical trials.

Data on millions of UK patients was used to train and validate the algorithm and, according to the new paper, it was found to be more effective than the latest QRISK algorithm (QRISK3) at predicting cardiovascular risk in both men and women and was also more accurate than other tools, including the ASCVD and SCORE2 scales. QRISK3 is currently recommended for NHS use by health technology assessment (HTA) agency NICE.

“While traditional cardiovascular risk factors such as smoking and high cholesterol are well-recognised, our latest research identifies less obvious, yet crucial, risk indicators,” said lead author Professor Julia Hippisley-Cox of Oxford’s Nuffield Department of Primary Care Health Sciences.

“QR4 also helps to address inequalities in health, particularly around learning disabilities, which often coincide with barriers to effective health management and access, leading to an increased risk,” she added.

The associations with cancer were “particularly striking”, even after accounting for reduced life expectancy, and have implications for the care of cancer survivors, especially at younger ages, according to the paper. Meanwhile, some other diseases that the scientists thought may raise risk, such as asthma, did not show a strong correlation.

The team has published the clinical code groups used to derive the model, which should help with further research into QR4; for example, to see how accurate it is in countries other than the UK.

According to Professor Bryan Williams, chief scientific and medical officer at the British Heart Foundation (BHF) charity, the tool will allow clinicians to build the clearest picture yet of individuals’ risk of developing cardiovascular diseases.

“Informed by large amounts of data from diverse populations, the new and improved algorithm takes vital steps towards improving risk assessment in populations where cardiovascular risk may often go under-detected,” he said.

Image by Grant Muller from Pixabay