A5018886198- Obesity, Physical Activity, Diet
- Statistical Methods and Bayesian Inference
- Bayesian Methods and Mixture Models
- Statistical Methods and Inference
- Health disparities and outcomes
- Child Nutrition and Feeding Issues
- Food Security and Health in Diverse Populations
- Birth, Development, and Health
- COVID-19 epidemiological studies
- Data-Driven Disease Surveillance
- Child Nutrition and Water Access
- Advanced Causal Inference Techniques
- Healthcare Policy and Management
- Urban Transport and Accessibility
- Gestational Diabetes Research and Management
- Health Systems, Economic Evaluations, Quality of Life
- Air Quality and Health Impacts
- Spatial and Panel Data Analysis
- Infant Development and Preterm Care
- Breastfeeding Practices and Influences
- Children's Physical and Motor Development
- Pharmaceutical Practices and Patient Outcomes
- Gene expression and cancer classification
- Medication Adherence and Compliance
- Global Maternal and Child Health
Medical University of South Carolina
2016-2025
Ralph H. Johnson VA Medical Center
2016-2025
MUSC Hollings Cancer Center
2024
Duke University
2010-2019
Johns Hopkins University
2016
Duke University Hospital
2012-2014
Duke Medical Center
2012-2014
Durham VA Health Care System
2014
Durham VA Medical Center
2014
Health Services Research & Development
2014
The response to the COVID-19 pandemic became increasingly politicized in U.S., and political affiliation of state leaders may contribute policies affecting spread disease. This study examines differences infection, death, testing by governor party across 50 U.S. states District Columbia.A longitudinal analysis was conducted December 2020 examining incidence, testing, test positivity rates from March 15, through 2020. A Bayesian negative binomial model fit estimate daily risk ratios posterior...
Few assessment instruments have examined the nutrition and physical activity environments in child care, none are self-administered. Given emerging focus on care settings as a target for intervention, valid reliable measure of environment is needed.To inter-rater reliability, 59 center directors 109 staff completed self-assessment concurrently, but independently. Three weeks later, repeat was by sub-sample 38 to assess test-retest reliability. To criterion validity, researcher-administered...
In applications involving count data, it is common to encounter an excess number of zeros. For example, in the study outpatient service utilization, utilization days will take on integer values, with many subjects having no (zero values). Mixed distribution models, such as zero-inflated Poisson and negative binomial, are often used fit data. A more general class mixture called hurdle can be model zero deflation well inflation. Several authors have proposed frequentist approaches fitting...
We examine the use of fixed-effects and random-effects moment-based meta-analytic methods for analysis binary adverse-event data. Special attention is paid to case rare adverse events that are commonly encountered in routine practice. study estimation model parameters between-study heterogeneity. In addition, we traditional approaches hypothesis testing average treatment effect detection heterogeneity across studies. derive three new methods, a simple (unweighted) estimator, parametric...
Health services data often contain a high proportion of zeros. In studies examining patient hospitalization rates, for instance, many patients will have no hospitalizations, resulting in count zero. When the number zeros is greater or less than expected under standard model, are said to be zero modified relative model. A similar phenomenon arises with semicontinuous data, which characterized by spike at followed continuous distribution positive support. analyzing zero‐modified and flexible...
Socially vulnerable communities may be at higher risk for COVID-19 outbreaks in the US. However, no prior studies examined temporal trends and differential effects of social vulnerability on incidence death rates. Therefore, we among counties with high low to quantify disparities over time.We conducted a longitudinal analysis examining rates from March 15 December 31, 2020, each US county using data USAFacts. We classified Social Vulnerability Index (SVI), percentile-based measure Centers...
Multiplex imaging platforms have enabled the identification of spatial organization different types cells in complex tissue or tumor microenvironment. Exploring potential variations co-occurrence colocalization cell across distinct disease classes can provide significant pathological insights, paving way for intervention strategies. However, existing methods this context either rely on stringent statistical assumptions suffer from a lack generalizability. We present highly powerful method to...
In many applications, the mean of a response variable can be assumed to nondecreasing function continuous predictor, controlling for covariates. such cases, interest often focuses on estimating regression function, while also assessing evidence an association. This article proposes new framework Bayesian isotonic and order-restricted inference. Approximating with high-dimensional piecewise linear model, constraint is incorporated through prior distribution slopes consisting product mixture...
In health services research, it is common to encounter semicontinuous data characterized by a point mass at zero followed right‐skewed continuous distribution with positive support. Examples include expenditures, in which the zeros represent subpopulation of patients who do not use services, while describes level expenditures among users. Semicontinuous are typically analyzed using two‐part mixture models that separately model probability and However, because second part conditions on...
Despite the rapid growth of electronic health data, most data systems do not connect individual patient records to sets from outside care delivery system. These isolated cannot support efforts recognize or address how physical and environmental context each influences choices outcomes. In this article we describe a geographic information system in Durham, North Carolina, links social via shared geography provide multidimensional understanding community status vulnerabilities. Geographic can...
We develop a spatial Poisson hurdle model to explore geographic variation in emergency department (ED) visits while accounting for zero inflation. The consists of two components: Bernoulli component that models the probability any ED use (i.e., at least one visit per year), and truncated number given use. Together, these components address both abundance zeros right-skewed nature nonzero counts. has hierarchical structure incorporates patient- area-level covariates, as well spatially...
Background: Adverse childhood experiences (ACEs) show a graded association with the development of substance use disorders (SUDs) and engagement in risky behaviors. Women are overrepresented among individuals more severe adversity (≥4 types ACEs) may be at particular risk for aberrant use.Objectives: To assess prevalence ACEs men women cannabis, opioid, cocaine, tobacco disorders.Methods: Non-treatment-seeking participating clinical addiction research single site completed ACE questionnaire...
In biomedical studies, there is often interest in assessing the association between one or more ordered categorical predictors and an outcome variable, adjusting for covariates. For a k-level predictor, typically uses either k-1 degree of freedom (df) test single df trend test, which requires scores different levels predictor. absence knowledge parametric form response function, can incorporate monotonicity constraints to improve efficiency tests association. This article proposes general...
Summary In studies involving functional data, it is commonly of interest to model the impact predictors on distribution curves, allowing flexible effects not only mean curve but also about mean. Characterizing for each subject as a linear combination high‐dimensional set potential basis functions, we place sparse latent factor regression coefficients. We induce selection by choosing shrinkage prior that allows many loadings be close zero. The number factors treated unknown through...
In health services research, it is common to encounter semicontinuous data, characterized by a point mass at zero followed right-skewed continuous distribution with positive support. Examples include expenditures, in which the zeros represent subpopulation of patients who do not use services, while describes level expenditures among users. Longitudinal data are typically analyzed using two-part random-effect mixtures one component that models probability use, and second log-scale However,...
Motivated by a study examining spatiotemporal patterns in inpatient hospitalizations, we propose an efficient Bayesian approach for fitting zero-inflated negative binomial models. To facilitate posterior sampling, introduce set of latent variables that are represented as scale mixtures normals, where the precision terms follow independent Pólya-Gamma distributions. Conditional on variables, inference proceeds via straightforward Gibbs sampling. For fixed-effects models, our is comparable to...
This article is the second installment of a two-part tutorial on analysis zero-modified count and semicontinuous data. Part 1, which appears as companion piece in this issue Statistics Medicine, provides general background overview topic, with particular emphasis applications to health services research. Here, we present three case studies highlighting various approaches for The first study describes methods analyzing zero-inflated longitudinal Case 2 considers use hurdle models...
Abstract Background Early care and education (ECE) is an important setting for influencing young children’s dietary intake. There are several factors associated with barriers to healthy eating in ECE, recent evidence suggests that participation the Child Adult Care Food Program (CACFP), primary national food assistance program may be fewer serving healthier foods. However, no prior studies have examined differences between CACFP participants non-participants across a large, multi-state...
Numerous previous studies have reported positive associations between exposure to greenspace and children's physical activity, but in high-income countries only. Prior also examined obesity children, these yielded inconsistent results focused mostly on older children. The purpose of this study was assess time children spent as the primary our outcomes interest, including 1) minutes 2) body mass index (BMI) z-score. Our sample 102 ages 3 5 years living Ensenada Tijuana, Mexico. We fit linear...