From Noise to Characterization Tool: Assessing Biases in Influenza Surveillance Methods Using a Bayesian Hierarchical Model

Ying Zhang, Ali Arab, Michael A. Stoto, Bejamin J. Cowling


Infectious disease surveillance is a process, the product of which reflects both real illness and public awareness of the disease. To develop a statistical framework to characterize influenza surveillance systems, Bayesian hierarchical model was applied to estimate the statistical relationships between influenza surveillance data and information environment (e.g. HealthMap, Google search volume,etc.) The model identified characteristics of surveillance systems that are more resistant to the information environment (percentage data, broad case definition and the senior population). General practitioner (%ILI-visit) and Laboratory (%positive) seem to capture the true infection at a constant proportion, and are less influenced by information environment.

Full Text:



Online Journal of Public Health Informatics * ISSN 1947-2579 *