NEW YORK – The Family Heart Foundation on Thursday said it has created a machine-learning model that can help doctors manage patients' cardiovascular disease based on elevated lipoprotein(a).
The nonprofit research and advocacy organization has partnered with a group of hospital systems, including Emory Healthcare, OhioHealth, and the Medical University of South Carolina, to implement the model, which is called FIND (Flag, Identify, Network, and Deliver) Lp(a). The goal is to flag heart disease patients likely to have elevated Lp(a), who would benefit from Lp(a) testing.
About 20 percent of people are estimated to have elevated Lp(a). Currently, such testing is infrequent, but growing, in standard practice.
Information on Lp(a) can help doctors decide, for example, whether a patient should aggressively manage other heart disease risk factors, even though there aren't drugs that specifically target Lp(a) on the market yet. Knowing patients' Lp(a) levels can also help explain their early or aggressive cardiovascular disease.
"As a preventive cardiologist, I know how critical it is that we identify individuals with high Lp(a) early," Ijeoma Isiadinso, director of the Emory Center for Heart Disease Prevention, said in a statement. "It is equally important that patients with high Lp(a) are aware of the risk factors for cardiovascular disease and receive aggressive treatment for these risk factors, including controlling blood pressure, diabetes, and cholesterol."
The machine-learning model was created using the Family Heart Foundation's database of medical claims and lab data and has shown a 60 percent success rate in identifying those with elevated Lp(a).
The FIND Lp(a) program is sponsored by Novartis, which is one of a spate of drugmakers currently developing drugs designed to lower Lp(a).