Dataset Information

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

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.

SUBMITTER: Banda JM

PROVIDER: S-EPMC6550268 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

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

Banda Juan M JM Sarraju Ashish A Abbasi Fahim F Parizo Justin J Pariani Mitchel M Ison Hannah H Briskin Elinor E Wand Hannah H Dubois Sebastien S Jung Kenneth K Myers Seth A SA Rader Daniel J DJ Leader Joseph B JB Murray Michael F MF Myers Kelly D KD Wilemon Katherine K Shah Nigam H NH Knowles Joshua W JW

NPJ digital medicine 20190411

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 us ...[more]

PMID: 31304370

Similar Datasets

Project description:One of the most common congenital metabolic disorders is familial hypercholesterolemia. Familial hypercholesterolemia is a condition caused by a type of genetic defect leading to a decreased rate of removal of low-density lipoproteins from the bloodstream and a pronounced increase in the blood level of total cholesterol. This disease leads to the early development of cardiovascular diseases of atherosclerotic etiology. Familial hypercholesterolemia is a monogenic disease that is predominantly autosomal dominant. Rare pathogenic variants in the LDLR gene are present in 75-85 % of cases with an identified molecular genetic cause of the disease, and variants in other genes (APOB, PCSK9, LDLRAP1, ABCG5, ABCG8, and others) occur at a frequency of < 5 % in this group of patients. A negative result of genetic screening for pathogenic variants in genes of the low-density lipoprotein receptor and its ligands does not rule out a diagnosis of familial hypercholesterolemia. In 20-40 % of cases, molecular genetic testing fails to detect changes in the above genes. The aim of this work was to search for new genes associated with the familial hypercholesterolemia phenotype by modern high-tech methods of sequencing and machine learning. On the basis of a group of patients with familial hypercholesterolemia (enrolled according to the Dutch Lipid Clinic Network Criteria and including cases confirmed by molecular genetic analysis), decision trees were constructed, which made it possible to identify cases in the study population that require additional molecular genetic analysis. Five probands were identified as having the severest familial hypercholesterolemia without pathogenic variants in the studied genes and were analyzed by whole-genome sequencing on the HiSeq 1500 platform (Illumina). The whole-genome sequencing revealed rare variants in three out of five analyzed patients: a heterozygous variant (rs760657350) located in a splicing acceptor site in the PLD1 gene (c.2430-1G>A), a previously undescribed single-nucleotide deletion in the SIDT1 gene [c.2426del (p.Leu809CysfsTer2)], new missense variant c.10313C>G (p.Pro3438Arg) in the LRP1B gene, and single-nucleotide deletion variant rs753876598 [c.165del (p.Ser56AlafsTer11)] in the CETP gene. All these variants were found for the first time in patients with a clinical diagnosis of familial hypercholesterolemia. Variants were identified that may influence the formation of the familial hypercholesterolemia phenotype.

Project description:BackgroundPandemics such as the COVID-19 pandemic and other severe health care disruptions endanger individuals to miss essential care. Machine learning models that predict which patients are at greatest risk of missing care visits can help health administrators prioritize retentions efforts towards patients with the most need. Such approaches may be especially useful for efficiently targeting interventions for health systems overburdened during states of emergency.MethodsWe use data on missed health care visits from over 55,500 respondents of the Survey of Health, Ageing and Retirement in Europe (SHARE) COVID-19 surveys (June - August 2020 and June - August 2021) with longitudinal data from waves 1-8 (April 2004 - March 2020). We compare the performance of four machine learning algorithms (stepwise selection, lasso, random forest, and neural networks) to predict missed health care visits during the first COVID-19 survey based on common patient characteristics available to most health care providers. We test the prediction accuracy, sensitivity, and specificity of the selected models for the first COVID-19 survey by employing 5-fold cross-validation, and test the out-of-sample performance of the models by applying them to the data from the second COVID-19 survey.ResultsWithin our sample, 15.5% of the respondents reported any missed essential health care visit due to the COVID-19 pandemic. All four machine learning methods perform similarly in their predictive power. All models have an area under the curve (AUC) of around 0.61, outperforming random prediction. This performance is sustained for data from the second COVID-19 wave one year later, with an AUC of 0.59 for men and 0.61 for women. When classifying all men (women) with a predicted risk of 0.135 (0.170) or higher as being at risk of missing care, the neural network model correctly identifies 59% (58%) of the individuals with missed care visits, and 57% (58%) of the individuals without missed care visits. As the sensitivity and specificity of the models are strongly related to the risk threshold used to classify individuals, the models can be calibrated depending on users' resource constraints and targeting approach.ConclusionsPandemics such as COVID-19 require rapid and efficient responses to reduce disruptions in health care. Based on characteristics available to health administrators or insurance providers, simple machine learning algorithms can be used to efficiently target efforts to reduce missed essential care.

Dataset Information

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

Publications

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

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets