Risk-based methods for fish and terrestrial animal disease surveillance.

Over recent years there have been considerable methodological developments in the field of animal disease surveillance. The principles of risk analysis were conceptually applied to surveillance in order to further develop approaches and tools (scenario tree modelling) to design risk-based surveillance (RBS) programmes. In the terrestrial animal context, examples of risk-based surveillance have demonstrated the substantial potential for cost saving, and a similar benefit is expected also for aquatic animals. RBS approaches are currently largely absent for aquatic animal diseases. A major constraint in developing RBS designs in the aquatic context is the lack of published data to assist in the design of RBS: this applies to data on (i) the relative risk of farm sites becoming infected due to the presence or absence of a given risk factor; (ii) the sensitivity of diagnostic tests (specificity is often addressed by follow-up investigation and re-testing and therefore less of a concern); (iii) data on the variability of prevalence of infection for fish within a holding unit, between holding units and at farm level. Another constraint is that some of the most basic data for planning surveillance are missing, e.g. data on farm location and animal movements. In Europe, registration or authorisation of fish farms has only recently become a requirement under EU Directive 2006/88. Additionally, the definition of the epidemiological unit (at site or area level) in the context of aquaculture is a challenge due to the often high level of connectedness (mainly via water) of aquaculture facilities with the aquatic environment. This paper provides a review of the principles, methods and examples of RBS in terrestrial, farmed and wild animals. It discusses the special challenges associated with surveillance for aquatic animal diseases (e.g. accessibility of animals for inspection and sampling, complexity of rearing systems) and provides an overview of current developments relevant for the design of RBS for fish diseases. Suggestions are provided on how the current constraints to applying RBS to fish diseases can be overcome.