Knot selection for low-rank kriging models of spatial risk in case-control studies. (June 2022)

• Record Type: Journal Article
• Authors: Boyle, Joseph; Wheeler, David C.
• Is Part Of: Spatial and spatio-temporal epidemiology. Volume 41(2022)
• Publication Date: 2022-06
• Subjects: Case-control study -- Low-rank kriging -- Simulation study -- Bayesian -- Generalized additive model; Epidemiology -- Statistical methods -- Periodicals; Epidemiology -- Periodicals; 614.4072
• DOI: 10.1016/j.sste.2022.100483
• Languages: English
• ISSNs: 1877-5845

Abstract

Highlights: Proposed two new approaches for knot selection in low-rank kriging models. Compared three knot selection approaches for low-rank kriging models. The commonly-used space-filling algorithm is suboptimal for case-control studies. Proposed methods increase the sensitivity and power to detect regions of risk. It is advised to use a number of knots approaching the number of cases in the study. Abstract: Many spatial analysis methods have been used to identify potential geographic clusters of disease in case-control studies. Low-rank kriging (LRK) models reduce the computational burden in generalized additive models by using a set of knot locations instead of the observed subject locations for estimating spatial risk. However, there is little guidance regarding selection of the number and location of the knots in case-control studies. We perform an extensive simulation study that compares a commonly-used method of knot selection in LRK models with two proposed methods and varies the number of knots. We find the commonly-used method is vastly outperformed by those that consider the locations of cases. We find that the Teitz and Bart heuristic allows the highest spatial sensitivity and power to detect zones of elevated risk, and recommend its use with a number of knots as close to the number of case locations as computation time will allow.