Our Research

Data sourced from nationally representative surveys add complexity to statistical analyses. A single major demographic group can overshadow and dominate pattern details. Unequal probabilities of selection and response inherent in the sample design make standard statistical methods inadequate for smaller demographic subgroups. Bayesian nonparametrics provide an efficient and stable solution to manage the complexity of high-dimensional exposures and overwhelmingly large sample sizes, while preserving model stability. Our research is working to develop extensions to commonly used Bayesian supervised and unsupervised methods to account for survey design and sampling variability. This integration of survey methodology and Bayesian statistics will provide researchers tools to improve population-based inference for representative survey data.

These are a set of projects focused on methods development and applications tailored for exposure data that has greatest impact on underrepresented and/or vulnerable populations. Our lab has focused on data that relies on a wide set of high-dimensional and correlated data to identify underlying patterns within a study population and assess its relationship to various health outcomes. Our lab is actively working on both unsupervised and supervised techniques that will account for population heterogeneity and various data complexities present in both cross-sectional and longitudinal cohort studies.

Our lab has focused on applying these methods to study populations at greatest risk for chronic disease and disparities (e.g. women, low-income, racial and ethnic minorities, pregnant persons, children), and has been applied in areas of nutrition, environmental health, reproductive health, and cancer disparities.

Social determinants of health refer to the non-medical factors that can contribute to a person’s overall health and well-being. These factors can include education, financial stability, neighborhood environment, access and quality of health care, and community or social constructs.  The impact of these factors on one’s health do not occur in isolation but are instead intersectional. This collection of projects aims to fully examine how each of these factors contribute collectively towards the widening disparities of certain populations. This research relies on the use multiple data sources including electronic health records, Census data, cancer registry databases, and other databases providing information on neighborhood-level environmental resources. The development and application of novel methods that can integrate and harmonize these different datasets will provide greater insight into the systematic barriers that contribute to poorer outcomes amongst those at greatest risk. Current collaborations have focused the impact of these determinants on the quality of care, access, treatment and survival of patients with cardiovascular disease risk, pancreatic, endometrial, ovarian, and prostate cancers.