Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2019 Apr 11:2:23.
doi: 10.1038/s41746-019-0101-5. eCollection 2019.

Finding missed cases of familial hypercholesterolemia in health systems using machine learning

Juan M Banda  1   2 , Ashish Sarraju  3 , Fahim Abbasi  3 , Justin Parizo  3 , Mitchel Pariani  3 , Hannah Ison  3 , Elinor Briskin  3 , Hannah Wand  3 , Sebastien Dubois  1 , Kenneth Jung  1 , Seth A Myers  4 , Daniel J Rader  5   6 , Joseph B Leader  7 , Michael F Murray  8 , Kelly D Myers  4   6 , Katherine Wilemon  6 , Nigam H Shah  1 , Joshua W Knowles  3   6   9
Affiliations

Finding missed cases of familial hypercholesterolemia in health systems using machine learning

Juan M Banda et al. NPJ Digit Med. .

Abstract

Familial hypercholesterolemia (FH) is an underdiagnosed dominant genetic condition affecting approximately 0.4% of the population and has up to a 20-fold increased risk of coronary artery disease if untreated. Simple screening strategies have false positive rates greater than 95%. As part of the FH Foundation's FIND FH initiative, we developed a classifier to identify potential FH patients using electronic health record (EHR) data at Stanford Health Care. We trained a random forest classifier using data from known patients (n = 197) and matched non-cases (n = 6590). Our classifier obtained a positive predictive value (PPV) of 0.88 and sensitivity of 0.75 on a held-out test-set. We evaluated the accuracy of the classifier's predictions by chart review of 100 patients at risk of FH not included in the original dataset. The classifier correctly flagged 84% of patients at the highest probability threshold, with decreasing performance as the threshold lowers. In external validation on 466 FH patients (236 with genetically proven FH) and 5000 matched non-cases from the Geisinger Healthcare System our FH classifier achieved a PPV of 0.85. Our EHR-derived FH classifier is effective in finding candidate patients for further FH screening. Such machine learning guided strategies can lead to effective identification of the highest risk patients for enhanced management strategies.

Keywords: Health care; Translational research.

PubMed Disclaimer

Conflict of interest statement

Competing interestsS.M. is an employee of Atomo LLC. M.M. has received personal fees from InVitae and Merck, and grant funding from Regeneron, outside the submitted work. K.M. is the founder of Atomo LLC. The remaining authors declare no competing interests. The FH Foundation (EIN 45-4597425) is a 501(c)3 public charity and research advocacy organization receiving funding from a diverse set of program sponsors including Amgen, Inc.

Figures

Fig. 1
Fig. 1
Classifier building followed by internal and external evaluation as well as evaluation via chart review for EHR-based FH case identification
Fig. 2
Fig. 2
Distribution of FH cases according to probability assigned by the random forest classifier as arbitrated by independent chart review at Stanford
Fig. 3
Fig. 3
Learning and testing setup for the Stanford FH classifier. * One comorbidity from the following: hypertension, coronary atherosclerosis (CAD), dyslipidemia, myocardial infarction; and had no history of nephrotic syndrome, or obstructive (cholestatic) liver disease
See this image and copyright information in PMC

References

    1. Gidding SS, et al. The agenda for familial hypercholesterolemia: a scientific statement from the american heart association. Circulation. 2015;132:2167–2192. doi: 10.1161/CIR.0000000000000297. - DOI - PubMed
    1. Abul-Husn NS, et al. Genetic identification of familial hypercholesterolemia within a single U.S. health care system. Science. 2016;354:aaf7000. doi: 10.1126/science.aaf7000. - DOI - PubMed
    1. Khera AV, et al. Diagnostic yield and clinical utility of sequencing familial hypercholesterolemia genes in patients with severe hypercholesterolemia. J. Am. Coll. Cardiol. 2016;67:2578–2589. doi: 10.1016/j.jacc.2016.03.520. - DOI - PMC - PubMed
    1. Besseling J, et al. Selection of individuals for genetic testing for familial hypercholesterolaemia: development and external validation of a prediction model for the presence of a mutation causing familial hypercholesterolaemia. Eur. Heart J. 2017;38:565–573. - PubMed
    1. Nordestgaard BG, et al. Familial hypercholesterolaemia is underdiagnosed and undertreated in the general population: guidance for clinicians to prevent coronary heart disease: consensus statement of the European Atherosclerosis Society. Eur. Heart J. 2013;34:3478–90a. doi: 10.1093/eurheartj/eht273. - DOI - PMC - PubMed