Date of Award

Summer 2025

Document Type

Open Access Thesis

Department

Epidemiology and Biostatistics

First Advisor

Stella Self

Abstract

Group testing involves pooling specimens from multiple individuals and offers an efficient means to surveil low-prevalence pathogens, but poses challenges for spatial cluster detection when only pooled results are observed. In this thesis, we develop a spatial scan statistic tailored to group-testing data with variable pool sizes. The statistic compares a null hypothesis of a homogeneous infection rate across all clusters to an alternative hypothesis that infection probabilities differ inside and outside a candidate cluster, with both models fitted by maximum likelihood estimation. We approximate the null distribution of the maximum likelihood ratio test via Monte Carlo simulation.

Through a comprehensive simulation study in a 46-county setting, we assess performance across two sample sizes (1,380 and 2,760 individuals), three pooling schemes (fixed sizes of 3 or 6, and randomly varying sizes 2–6), and two infection prevalences (low and high). Our results demonstrate that geographically coherent pooling (same county assignment) markedly enhances power compared to random pooling. The method performed best with smaller pool sizes and larger sample sizes, while larger pools and random assignment diluted the spatial signal and reduced detection performance.

We apply the method to tick pools tested for Rickettsia parkeri collected across South Carolina from 2021 to 2024. Ticks were gathered at multiple sites during that period, grouped into pools, and each pool was tested for Rickettsia parkeri infection using a group testing protocol. No statistically significant clusters were identified.

Rights

© 2025, Vincent Onyame

Download

Included in

Biostatistics Commons

Share

COinS