Open AccessValidation of a spatial liver fluke model under field conditions in Ireland
Author(s) -
Amalia Naranjo Lucena,
María Pia Munita,
Ana María Martínez-Ibeas,
Guy McGrath,
Ríona Sayers,
Grace Mulcahy,
Annetta Zintl
Publication year - 2018
Publication title -
geospatial health
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.545
H-Index - 36
eISSN - 1970-7096
pISSN - 1827-1987
DOI - 10.4081/gh.2018.641
Subject(s) - fasciola hepatica , liver fluke , fasciolosis , livestock , risk assessment , veterinary medicine , herd , hepatica , predictive modelling , geography , cluster (spacecraft) , predictive value , biology , statistics , ecology , medicine , helminths , computer science , zoology , mathematics , computer security , programming language
Fasciola hepatica is the causative agent of fasciolosis, a global disease of a wide range of mammals, particularly sheep and cattle. Liver fluke infection causes annual losses estimated at around €2.5 billion to livestock and food industries worldwide. Various models have been developed to define risk factors and predict exposure to this liver fluke in ruminants in European countries, most of them based exclusively on data from dairy herds. The aim of this study was to validate a published theoretical baseline risk map of liver fluke exposure and cluster maps in Ireland, by including further explanatory variables and additional herd types that are spatially more widespread. Three approaches were employed: i) comparison of predicted and actual exposure; ii) comparison of cluster distribution of hotspots and coldspots; and iii) development of a new model to compare predicted spatial distribution and risk factors. Based on new survey data, the published baseline predictive map was found to have a sensitivity of 94.7%, a specificity of 5%, a positive predictive value of 60% and a negative predictive value of 38.2%. In agreement with the original model, our validation highlighted temperature and rainfall among the main risk factors. In addition, we identified vegetation indices as important risk factors. Both the previously published and our new model predict that exposure to Fasciola is higher in the western parts of Ireland. However, foci of high probability do not match completely, nor do the location of clusters of hotspots and coldspots.