Improving Diagnosis and Clinical Management of Familial Hypercholesterolemia Through Integrated Machine Learning, Implementation Science, and Behavioral Economics
Abstract
This project addresses the PRMRP FY20 Topic Area of Familial Hypercholesterolemia (FH) and the following areas of encouragement: (1) studies to systematically identify individuals at risk for FH using machine-learning tools, (2) research to improve early diagnosis of FH and the implementation of diagnostic tools, and (3) research to understand approaches to clinical management to treat FH patients at higher risk for progressing to clinical atherosclerotic cardiovascular disease (ASCVD). FH is an inherited disorder that affects more than about 1 in every 250 American men, women, and children of all races and ethnicities, including approximately 84,000 active and reserve Service Members, civilians working in the Department of Defense (DoD), and Veterans. People with FH have high levels of low-density lipoprotein cholesterol (LDL-C), also known as “bad” cholesterol. Buildup of LDL-C in artery walls can harden arteries, which is called atherosclerosis, and can lead to problems such as heart attacks and strokes in young adults and even children. FH can be diagnosed in children as young as age 2, and treatment can begin at age 8 to prevent or delay the onset of heart disease. Although lifestyle and diet are important factors to staying heart healthy, people with FH need medication to reduce the chance of early heart disease and heart attacks. Unfortunately, nearly 9 out of 10 individuals with FH never find out they have this genetic disorder, and therefore do not get the necessary treatment. This is because there is low public awareness of FH, and even many doctors fail to link very high cholesterol with a genetic cause. In addition, many individuals with FH show no visible symptoms and do not know they have the disorder until they suffer a major health event such as a heart attack. The following individual echoes the needs and frustrations of many others with FH: “I was told by my primary care doctor I just had high cholesterol for years and I ‘just need to watch my diet.’ I completely changed my diet to mostly plant-based, but my LDL-C [level] hardly moved…When I was 38, I had almost complete blockage and had bypass surgery. Since then, I was diagnosed with FH. Before the diagnosis, I had no idea what FH was and was very frustrated with myself thinking of how I could have prevented this! After seeing my specialist and understanding more about FH, I have been on a good combination of medications and my LDL-C is now [low]! I suffered for a long time and almost died because of not being diagnosed. That has to change.” The good news is that FH can be diagnosed with a simple blood test and a reported family history of early heart disease. Genetic testing can be used to confirm the presence of FH. Early treatment markedly reduces the risk for heart attacks and other consequences of heart disease. With early diagnosis and treatment, individuals with FH can live longer, healthier lives. This project aims to improve the diagnosis and care of people with FH. This research team of doctors, scientists, and FH specialists is a partnership of the University of Pennsylvania Health System (UPHS) and the FH Foundation, a patient-centered organization dedicated to research, advocacy, and education. Together, the team plans to develop an improved way for doctors to identify their patients who are at high risk for having FH and to help diagnosed patients and their doctors manage the condition. To do this, we will use a machine-learning model, or tool, called Flag Identify Network and Deliver FH (FIND FH), developed by the FH Foundation, that can scan a large database of patient health records to look for clues that identify people at high risk of FH. We will then test different ways that doctors can contact these individuals, get them tested for FH, and decide on FH treatments and how to best ensure that people stay on these medications. This part of the project will take into consideration how people make decisions and choices i
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Dec 05, 2021
- Source ID
- W81XWH2110421
Entities
People
- Kevin Volpp
Organizations
- United States Army
- University of Pennsylvania