Space-Time Smoothing Models for Surveillance and Complex Survey Data

Abstract

Area and time-specific estimates of disease rates, cause-specific mortality rates and other key health indicators are of great interest for health care and policy purposes. Such estimates provide the information needed to identify areas with increased risk, effectively allocate resources, and target interventions. A wide variety of data, such as vital statistics, complex surveys, demographic surveillance sites, and disease registries, are used for these purposes. Unfortunately, the sample size of data available at a granular space-time scale is often too small to provide reliable estimates and uncertainty intervals. Using data from multiple sources and spatial and temporal smoothing is beneficial to alleviate problems of data scarcity. The purpose of the work described herein is to use Bayesian space-time models, to combine data from multiple sources to provide reliable area-based estimates. This work is motivated by estimating rates of health indicators (e.g. diabetes, smoking) by health reporting areas in King County from the Behavioral Risk Factor Surveillance Survey, child mortality by regions in Tanzania from Demographic and Health Surveys and demographic surveillance sites, and cancer-specific incidence and mortality rates in Europe from government data and local registries.