A stochastic model studying interactions between breastfeeding and vaccination in providing rotavirus immunity
Open Access
Author:
Mattiace, John Michael
Graduate Program:
Statistics
Degree:
Master of Science
Document Type:
Master Thesis
Date of Defense:
December 25, 2024
Committee Members:
Murali Haran, Thesis Advisor/Co-Advisor Stephen Berg, Committee Member Murali Haran, Committee Member Runze Li, Program Head/Chair
Keywords:
rotavirus approximate bayesian computation bayesian ordinary differential equations sir
Abstract:
It is well known that breastfeeding bolsters infant immune systems, but whether breastfeeding impacts long-term immunity acquired via vaccination is less well-studied. Of particular interest is whether breastfeeding actively assists or detracts from vaccine uptake in otherwise susceptible children. We consider data from the National Immunization Survey as well as the Healthcare Cost and Utilization Project and develop a discrete-time, stochastic compartmental model, explicitly accounting for the number of breastfed children, with the end goal of investigating the interaction between rates of breastfeeding, rates of infection, and rates of successful vaccination. Likelihood-based inference for this model is computationally challenging, so we resort to an approach based on Approximate Bayesian Computation Sequential Monte Carlo (ABCSMC) methodology to estimate parameters driving the model dynamics. We present preliminary findings from fitting our new stochastic model to these data.
