Estimating the force of infection from current-status data.

Project Lead: Dennis Heisey (USGS)

We have developed novel Bayesian techniques to analyze disease data. The primary challenge is to infer how observed patterns are driven by the underlying transmission process, and how that process is a function of space, time, age, and cohort (group). We combined age-cohort-period models with Bayesian spatial frailty models into a single framework that allows us to use observed current-status data to estimate the spatio-temporal disease transmission process. In deer, we showed that infection hazards were highest in 2-3yr olds, and that the spatial hazard for females was rougher (more heterogenous) than for males reflecting their more localized use of space. We believe this method also has utility for the analysis of spatially-reference survival data.


Citations:
Heisey, D.M., E.E. Osnas, P.C. Cross, D.O. Joly, J.A. Langenberg & M.W. Miller. 2010. Linking process to pattern: estimating spatiotemporal dynamics of a wildlife epidemic from cross-sectional data. Ecological Monographs: Vol. 80, No. 2, pp. 221-240. doi: 10.1890/09-0052.1

Heisey, DM, EE Osnas, P.C. Cross, DO Joly, JA Langenberg & MW Miller. In press. Rejoinder - Shifting through model space. Ecological Monographs.



Return to Disease Ecology Research Projects