Quantitative Health Sciences Calendar

Intersectionality theory, which posits that social categorizations (e.g., those based on race, gender, and class) are interconnected and create overlapping systems of discrimination or disadvantage, has become increasingly influential in population health research. However, translating the core tenets of intersectionality into quantitative analyses presents significant methodological challenges. This presentation will introduce Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (MAIHDA) as a novel approach for quantitative intersectionality research that addresses and overcomes many of these challenges. I will start with a brief overview of intersectionality theory and its application in population health research, highlighting the limitations of traditional statistical methods in capturing the complexity of intersecting social positions. I will then introduce the MAIHDA framework, explaining its core components and how it addresses these limitations. A case example will illustrate the application of MAIHDA, demonstrating how it can reveal patterns of health inequities that might be obscured by conventional approaches. Finally, we will review recent extensions of the MAIHDA methodology, including the incorporation of survey weights, applications to longitudinal data, and the inclusion of exposure variables, expanding its potential for investigating the complex interplay of social factors and health outcomes. Throughout the presentation, resources and opportunities for discussion will be provided to promote both the understanding and critical evaluation of MAIHDA as a tool for quantitative intersectionality research.  

Bio: Dr. Ariel Beccia is an Instructor in the Division of Adolescent and Young Adult Medicine at Boston Children’s Hospital, the Department of Pediatrics at Harvard Medical School, and the Department of Epidemiology at the Harvard T.H. Chan School of Public Health. Her research focuses on identifying the structural drivers of inequities in mental health, especially eating disorder-related outcomes; she also has an interest in developing and applying methods to better incorporate critical social theories into quantitative population health analyses. She earned her PhD in epidemiology from the Clinical and Population Health Research program at the University of Massachusetts Chan Medical School.  

Click here to join or call 1 301 715 8592, Meeting ID: 948 2951 6040 password: 202286

We develop Bayesian two-part regression models for estimating treatment effect in longitudinal randomized clinical trials with zero-inflated data. We model data coming from two randomized controlled trials, one measuring the effect of an intervention to reduce levels of substance abuse and the other measuring the effect of an intervention to improve linkage and engagement in medical care among previously incarcerated people with HIV. To estimate the effect of a screening, brief intervention, and referral to treatment (SBIRT) intervention on days of substance we develop a longitudinal zero-inflated binomial model. We also develop a longitudinal zero-inflated Poisson model for modeling counts of medical visits with unbalanced observation times. Time varying random effects with a parameterized covariance matrix are used to model within-individual correlation over time and are shown to improve fit over standard models. Unbalanced observation times are accounted for using partially observed latent time effects and regression equation offsets.

Bio: Benjamin Rogers is a Research Assistant Professor in the Department of Biostatistics at UMass Amherst and a member of the Reich Lab. He received his Ph.D. in biostatistics from UCLA in 2022 for work on Bayesian longitudinal models. As a member of the Reich lab, he has worked on disease forecasting models, forecast scoring approaches and helped launch the United States SARS-CoV-2 Variant Nowcast Hub.

Click here to join or call 1 301 715 8592, Meeting ID: 948 2951 6040 password: 202286