- Finding missed cases of familial hypercholesterolemia in health systems using machine learning NPJ DIGITAL MEDICINE 2019; 2
EVALUATION OF TRENDS IN READMISSION AND MORTALITY RATES AFTER HEART FAILURE HOSPITALIZATION IN THE VETERANS AFFAIRS HEALTH CARE SYSTEM BETWEEN 2007 AND 2017
ELSEVIER SCIENCE INC. 2019: 737
View details for Web of Science ID 000460565900737
Finding missed cases of familial hypercholesterolemia in health systems using machine learning.
NPJ digital medicine
2019; 2: 23
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.
View details for DOI 10.1038/s41746-019-0101-5
View details for PubMedID 31304370
View details for PubMedCentralID PMC6550268
Validity of Performance and Outcome Measures for Heart Failure.
Circulation. Heart failure
2018; 11 (9): e005035
Background Numerous quality metrics for heart failure (HF) care now exist based on process and outcome. What remains unclear, however, is if the correct quality metrics are being emphasized. To determine the validity of certain measures, we compared correlations between measures and reliability over time. Measures assessed include guideline-recommended beta-blocker (BB), any BB, angiotensin-converting enzyme (ACE) inhibitor or angiotensin receptor blocker, mineralocorticoid receptor antagonist, and hydralazine/isosorbide dinitrate (in blacks) use among candidates, 30-day mortality, 1-year mortality, and 30-day readmission. Methods and Results This was an observational cohort analysis using chart review and electronic resources for 55735 patients from 102 Veterans Affairs medical centers hospitalized with HF from 2008 to 2013. Assessments of convergent validity and reliability were performed. Significant correlations were found between in-hospital rates of ACE inhibitor use and the following measures: BB use, 30-day mortality, and 1-year mortality. Guideline-recommended BB use was also significantly correlated with mineralocorticoid receptor antagonists, 30-day mortality, and 1-year mortality. There was no correlation between 30-day readmission rates and any therapy or mortality. Measure reliability over time was seen for guideline-recommended BBs ( r=0.57), mineralocorticoid receptor antagonists ( r=0.50), 30-day mortality ( r=0.29), and 1-year mortality ( r=0.31). ACE inhibitor and readmission rates were not reliable measures over time. Conclusions BB use, ACE inhibitor use, mortality, and mineralocorticoid receptor antagonist use are valid measures of HF quality. Thirty-day readmission rate did not seem to be a valid measure of HF quality of care. If the goal is to identify high-quality HF care, the emphasis on decreasing readmission rates might be better directed towards improving usage of the recommended therapies.
View details for PubMedID 30354367
- Association Between Offering Limited Left Ventricular Ejection Fraction Echocardiograms and Overall Use of Echocardiography JAMA INTERNAL MEDICINE 2018; 178 (9): 1270-+
- Validity of Performance and Outcome Measures for Heart Failure CIRCULATION-HEART FAILURE 2018; 11 (9)
Association Between Offering Limited Left Ventricular Ejection Fraction Echocardiograms and Overall Use of Echocardiography.
JAMA internal medicine
View details for PubMedID 30039163
Novel Therapies for Familial Hypercholesterolemia.
Current treatment options in cardiovascular medicine
2016; 18 (11): 64-?
Both HeFH and HoFH require dietary and lifestyle modification. Pharmacotherapy of adult HeFH patients is largely driven by the American Heart Association (AHA) algorithm. A high-potency statin is started initially with a goal low-density lipoprotein cholesterol (LDL-C) reduction of >50 %. The LDL-C target is adjusted to <100 or <70 mg/dL in subjects with coronary artery disease (CAD) with ezetimibe being second line. If necessary, a third adjunctive therapy, such as a PSCK9 inhibitor (not yet approved in children) or bile acid-binding resin, can be added. Finally, LDL-C apheresis can be considered in patients with LDL-C >300 mg/dL (or >200 mg/dL with significant CAD, although now approved for LDL-C as low as 160 mg/dL with CAD). Due to the early, severe LDL-C elevation in HoFH patients, concerning natural history, rarity of the condition, and nuances of treatment, all HoFH patients should be treated at a pediatric or adult center with HoFH experience. LDL-C apheresis should be considered as early as 5 years of age. However, apheresis availability and tolerability is limited and pharmacotherapy is required. Generally, the AHA algorithm with reference to the European Atherosclerosis Society Consensus Panel recommendations is reasonable with all patients initiated on high-dose, high-potency statin, ezetimibe, and bile acid-binding resins. In most, additional LDL-C lowering is required with PCSK9 inhibitors and/or lomitapide or mipomersen. Liver transplantation can also be considered at experienced centers as a last resort.
View details for DOI 10.1007/s11936-016-0486-2
View details for PubMedID 27620638
Spatiotemporal Analysis of Malaria in Urban Ahmedabad (Gujarat), India: Identification of Hot Spots and Risk Factors for Targeted Intervention.
The American journal of tropical medicine and hygiene
2016; 95 (3): 595-603
The world population, especially in developing countries, has experienced a rapid progression of urbanization over the last half century. Urbanization has been accompanied by a rise in cases of urban infectious diseases, such as malaria. The complexity and heterogeneity of the urban environment has made study of specific urban centers vital for urban malaria control programs, whereas more generalizable risk factor identification also remains essential. Ahmedabad city, India, is a large urban center located in the state of Gujarat, which has experienced a significant Plasmodium vivax and Plasmodium falciparum disease burden. Therefore, a targeted analysis of malaria in Ahmedabad city was undertaken to identify spatiotemporal patterns of malaria, risk factors, and methods of predicting future malaria cases. Malaria incidence in Ahmedabad city was found to be spatially heterogeneous, but temporally stable, with high spatial correlation between species. Because of this stability, a prediction method utilizing historic cases from prior years and seasons was used successfully to predict which areas of Ahmedabad city would experience the highest malaria burden and could be used to prospectively target interventions. Finally, spatial analysis showed that normalized difference vegetation index, proximity to water sources, and location within Ahmedabad city relative to the dense urban core were the best predictors of malaria incidence. Because of the heterogeneity of urban environments and urban malaria itself, the study of specific large urban centers is vital to assist in allocating resources and informing future urban planning.
View details for DOI 10.4269/ajtmh.16-0108
View details for PubMedID 27382081
View details for PubMedCentralID PMC5014265