News
Better, more targeted use of electrocardiograms could help ease underdiagnosis rates, new research has found.
Women’s lives could be saved by introducing sex-specific risk criteria for heart disease detection, according to American researchers calling for a modelling makeover.
A new paper out of Stanford University found the current methods of measuring heart disease risk are leaving many women diagnosed later than men and with more symptoms, or left undiagnosed completely.
Using artificial intelligence, the team tested thousands of possible factors to improve detection, ultimately finding electrocardiograms (EKGs) were likely to have the most impact.
Stanford University researcher Skyler St. Pierre said sex-neutral criteria is failing to diagnose women adequately, and if criteria were overhauled, underdiagnosis would be less severe.
‘While traditional clinical models are easy to use, we can now use machine learning to comb through thousands of other possible factors to find new, meaningful features that could significantly improve early detection of disease,’ they said.
‘While sex-specific medicine is one step in the right direction, patient-specific medicine would provide the best outcomes for everyone.’
The research team found women are being underdiagnosed for first degree atrioventricular block and dilated cardiomyopathy twice and 1.4 times more than men, respectively.
Using the AI model, the team was able to identify similarities and differences between men and women when diagnosing hypertension; for example, blood pressure, body mass index, age, and cholesterol ranked highly as risks for both sexes, but diabetes status was an important feature for men only.
‘Only total cholesterol, not HDL cholesterol, ranked in the top 10 for men, while both appeared for women,’ the researchers said.
‘For ischemic diseases, age, HDL and total cholesterol, and body mass index were in the top 10 for both women and men.
‘[Whereas] for conduction disorders, for women, age and HDL cholesterol were in the top features, while for men, the age and body mass index appeared.’
In a bid to combat account for these differences, the research tested for four additional metrics not traditionally measured: cardiac magnetic resonance imaging, pulse wave analysis, EKGs, and carotid ultrasounds.
Using data from more than 20,000 patients, researchers concluded that of the tested metrics, EKGs were most effective at improving the detection of cardiovascular disease.
However, they stressed that it should be used in conjunction with traditional risk models, such as the Framingham Risk Score, rather than as a replacement.
The results come as heart disease remains Australia’s, and the world’s, number one killer, accounting for 10% of all deaths.
The US researchers concluded that their findings are the first step towards rethinking risk factors for heart disease and using new technologies to improve predictions.
‘Unarguably, there is an urgent need for sex-specific diagnostic criteria,’ they said.
‘A more accurate individualised risk prediction of cardiovascular diseases would enable personalised treatment and prevention strategies, a more effective allocation of medical resources, and an early and precise identification of high-risk individuals.’
Last year, a new Australian CVD Risk Calculator was launched to help GPs to better identify high-risk people in need of treatment.
Log in below to join the conversation.
cardiovascular disease CVD preventive health
newsGP weekly poll
Which RACGP request would you most like the Government to fund in the upcoming Federal Budget?