Research & Publications

Improving guideline adherence for diverse populations

We aim to advance the quality of patient care by studying racial/ethnic minority representation in cardiovascular guidelines and increasing guideline adherence for diverse populations.

Selected projects:

Statin use in elderly patients with stable atherosclerotic cardiovascular disease

Authors: Gabriela Spencer-Bonilla, Sukyung Chung, Summer Ngo, Paul Heidenreich, Latha Palaniappan, Fatima Rodriguez

Objective: Elderly patients (>75 years of age) are underrepresented in clinical trials that guide our evidence base for secondary prevention and cholesterol management. The 2018 multi-society cholesterol guidelines recommend use of at least moderate intensity statins for elderly patients with atherosclerotic cardiovascular disease (ASCVD). This study aims to understand statin prescribing patterns and clinical outcomes for elderly patients with ASCVD.

Under-reporting, under-representation, and disaggregation of racial/ethnic minorities in high-impact blood cholesterol clinical trials

Authors: Ashish Sarraju, Andrew Ward, Sukyung Chung, David Scheinker, Fatima Rodriguez

Objective: Evaluate contemporary gaps in the reporting, representation, and disaggregation of racial/ethnic minority participants in high-impact blood cholesterol clinical trials by examining trials cited in the 2018 American Multisociety Guideline on the Management of Blood Cholesterol.

Understanding health disparities

We strive to improve health outcomes for vulnerable populations by identifying racial, ethnic, gender, and socioeconomic disparities in cardiovascular disease prevention and treatment, and developing interventions for these disparities.

Selected projects:

County-level factors associated with cardiovascular mortality disaggregated by race/ethnicity

Authors: Bongeka Zuma, Justin T. Parizo, Areli Valencia, Gabriela Spencer-Bonilla, Manuel Blum, David Scheinker, Fatima Rodriguez

Objective: The interplay between race/ethnicity and county-level characteristics in explaining disparities in CVD mortality is poorly understood. Using national 2017 mortality data linked with county-level characteristics, we studied the association between county-level factors (demographic, regional, socioeconomic, cardiovascular risk factors, and healthcare access) with race/ethnicity-specific CVD mortality patterns. 

To read the full article, click here
The Hispanic Paradox in County-Level Obesity Prevalence

Authors: Areli Valencia, Bongeka Zuma, Gabriela Spencer-Bonilla, Lenny López, David Scheinker, Fatima Rodriguez

Objective: The Hispanic paradox refers to an epidemiological phenomenon whereby Hispanics experience better than expected health outcomes despite having adverse sociodemographic factors. Prior work has shown that percentage of Hispanics in a county is associated with a lower prevalence of county-level obesity. Using data extracted from the 2018 Robert Wood Johnson Foundation County Health Rankings, we sought to determine if the association between percentage of Hispanics in a county and county-level obesity prevalence is explained by county-level characteristics by stratifying counties by Hispanic ethnic proportion.

To read the full article, click here

Enhancing cardiovascular risk prediction

We construct novel strategies for improving cardiovascular risk stratification to help guide clinical decision-making.

Selected projects:

Predicting recurrent atherosclerotic cardiovascular disease risk in secondary prevention patients using machine learning

Authors: Ashish Sarraju, Andrew Ward, Sukyung Chung, David Scheinker, Fatima Rodriguez

Objective: Patients with prior atherosclerotic cardiovascular disease (ASCVD) are at high risk for subsequent ASCVD events despite the use of guideline-directed preventive therapies. Yet, our ability to risk stratify these patients to guide management decisions is limited, due to a paucity of validated predictive algorithms for subsequent ASCVD risk, which is a notable gap in secondary prevention. We sought to build and test ML models trained on EHR data to facilitate recurrent ASCVD risk prediction in a diverse, real-world, secondary prevention population.

Identification of Factors Associated With Variation in US County-Level Obesity Prevalence Rates Using Epidemiologic vs Machine Learning Models

Authors: David Scheinker, Areli Valencia, Fatima Rodriguez

Objective: Obesity is a leading cause of high health care expenditures, disability, and premature mortality. Previous studies have documented geographic disparities in obesity prevalence. This cross-sectional study identified county-level factors associated with obesity using traditional epidemiologic and machine learning methods.

To read the full article, click here