Validation of a model for ranking aquaculture facilities for risk-based disease surveillance

Abstract

A semi-quantitative model for risk ranking of aquaculture facilities in Switzerland with regard to the introduction and spread of Viral Haemorrhagic Septicaemia (VHS) and Infectious Haematopoietic Necrosis (IHN) was developed in a previous study (Diserens et al., 2013). The objective of the present study was to validate this model using data collected during field visits on aquaculture sites in four Swiss cantons compared to data collected through a questionnaire in the previous study. A discrepancy between the values obtained with the two different methods was found in 32.8% of the parameters, resulting in a significant difference (p<0.001) in the risk classification of the facilities. As data gathered exclusively by means of a questionnaire are not of sufficient quality to perform a risk-based surveillance of aquaculture facilities a combination of questionnaires and farm inspections is proposed. A web-based reporting system could be advantageous for the factors which were identified as being more likely to vary over time, in particular for factors considering fish movements, which showed a marginally significant difference in their risk scores (p≥0.1) within a six- month period. Nevertheless, the model proved to be stable over the considered period of time as no substantial fluctuations in the risk categorisation were observed (Kappa agreement of 0.77).Finally, the model proved to be suitable to deliver a reliable risk ranking of Swiss aquaculture facilities according to their risk of getting infected with or spreading of VHS and IHN, as the five facilities that tested positive for these diseases in the last ten years were ranked as medium or high risk. Moreover, because the seven fish farms that were infected with Infectious Pancreatic Necrosis (IPN) during the same period also belonged to the risk categories medium and high, the classification appeared to correlate with the occurrence of this third viral fish disease.

An Integrated GIS-Based Reinforcement Learning Approach for Efficient Prediction of Disease Transmission in Aquaculture

“…In [5], modifications to a proposed "risk of flooding" project were proposed based on whether the farm has experienced flooding before. The initial categorization of flood farms ("null" risk and "low" risk) was revised to take into account the flooding hazard maps of each Swiss canton.…”