BOSTON ‐ PUPA : A Bayesian Online Spatio‐Temporal Outbreak Detection Framework With Prior Updating and p ‐Value Adaptation
Yanzhao Wang, I. M. L. Nadeesha Jayaweera, Jian Zou · Statistical Analysis and Data Mining The ASA Data Science Journal · 2025
ABSTRACT Early online outbreak detection for an epidemic is vital for disease‐control authorities to make policies for the protection of public health and normal socioeconomic functions. Modern public health streaming surveillance data are often collected from multiple data sources, exhibiting spatio‐temporal interdependence and imbalance issues. To address those issues, we propose a Bayesian online spatio‐temporal outbreak detection with prior updating and p ‐value adaptation (BOSTON‐PUPA) procedure. Using sequential p ‐value combinations, this iterative procedure involves the generalized Poisson distribution (GPD) model and supports synchronous surveillance over multiple locations, with a controlled false detection rate as well as high sensitivity against outbreaks in a wide range of signal‐to‐noise ratios. In the simulation study, we employed and compared several popular combined p ‐value methods in the BOSTON‐PUPA procedure based on sensitivity, specificity, false detection rate and delay before making recommendations. We illustrated our method by detecting the outbreaks in the real COVID‐19 daily case count data in Massachusetts counties in 2020.