Machine learning models that incorporate social determinants of health (SDH) do a better job of accurately predicting cardiovascular disease (CVD) and guiding treatment strategies, according to new research out of New York University’s School of Global Public Health and Tandon School of Engineering.
The analysis, published in the American Journal of Preventive Medicine, focused on 48 different peer-reviewed studies published from 1995 to 2020.
Overall, the group found, machine learning models performed better when SDH were included than when they were not.
“Cardiovascular disease is increasing, particularly in low- and middle-income countries and among communities of color in places like the United States,” Rumi Chunara, associate professor of biostatistics at NYU School of Global Public Health and associate professor of computer science and engineering at NYU Tandon School of Engineering, said in a prepared statement. “Because these changes are happening over such a short period of time, it is well known that our changing social and environmental factors, such as increased processed foods, are driving this change, as opposed to genetic factors which would change over much longer time scales.”
Gender, race/ethnicity, marital status, occupation and income, were the most frequently counted SDH variables.
The authors also found that the studies they found were primarily from North America, Europe, and China, thereby limiting the diversity of the included populations and variance in SDH.
“If you only do research in places like the United States or Europe, you’ll miss how social determinants and other environmental factors related to cardiovascular risk interact in different settings and the knowledge generated will be limited,” Chunara said.
Chunara and colleagues concluded that their research underscores the need for studies that include more diverse populations improve CVD prediction in diverse settings, especially in regions where disease risk is increasing.
“Social and environmental factors have complex, non-linear interactions with CVD,” Chunara said. "Machine learning can be particularly useful in capturing these intricate relationships.”
Read the full study here.