Abstract 16078: Risk Classification of Familial Hypercholesterolemia Through Screening of the Electronic Health Record Within a Large Healthcare System

Abstract

Introduction: Familial hypercholesterolemia (FH) is a genetic disorder characterized by very high low-density lipoprotein cholesterol (LDL-c) levels. The heterogeneity of clinical risk within FH may lead to differential treatment, management, and outcomes. Screening the electronic health record (EHR) permits risk classification within FH at scale. We leveraged the Georgetown/MedStar Health System (MHS) EHR to perform risk classification while describing other characteristics of suspected FH patients. Methods: The Dutch Lipid Clinical Network (DLCN) remains a common scoring algorithm for FH risk classification. Applying the DLCN algorithm to the EHR of MHS enabled risk classification of approximately 1.6 million patients. The classifiers included ‘definite,’ ‘probable,’ and ‘possible’ heterozygous FH (HeFH). ‘Possible’ homozygous FH (HoFH) was determined by LDL-c > 500 mg/dL. Of these suspected FH patients, current age, gender, and race as well as abnormal lipid profile data (high-density lipoprotein cholesterol (HDL-c) cholesterol < 20 mg/dL, triglyceride (TG) > 300 mg/dL) were descriptively described. Results: Of 1.6 million patients, the number of definite, probable, and possible HeFH patients were 83 (1:19722), 690 (1:2318), and 5945 (1:266), respectively. The number of possible HoFH patients was 8 (1:200000). The highest reported median LDL-C (IQR) for the three HeFH subpopulations was 377 (346, 430 mg/dL), 262 (251, 282 mg/dL), and 203 (mg/dL), and the majority of all subpopulations were female and non-white. Definite HeFH patients were more likely to present with very low HDL-c and approximately 1:10 suspected HeFH patients presented with very high triglycerides. Conclusion: The application of DLCN to the EHR demonstrates risk classification for FH in a large healthcare system. The implications of stratifying suspected FH patients by classifier within the EHR may optimize clinical care as well as registry management.