Estimating prevalence and diagnostic test characteristics of bovine cysticercosis in Belgium in the absence of a ‘gold standard’ reference test using a Bayesian approach

[Display omitted] •4 detection tests for bovine cysticercosis (BCC) are combined in a Bayesian analysis.•34% of Belgian cattle are BCC positive for one or more detection techniques.•Feasibility of Ag-ELISA as detection technique at slaughter needs further research. A Bayesian model was developed to...

Full description

Saved in:
Bibliographic Details
Published in:Veterinary parasitology 2018-04, Vol.254, p.142-146
Main Authors: Jansen, Famke, Dorny, Pierre, Gabriël, Sarah, Eichenberger, Ramon Marc, Berkvens, Dirk
Format: Article
Language:English
Subjects:
Bayesian model
Belgium
Cattle
Taenia saginata
True prevalence
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:[Display omitted] •4 detection tests for bovine cysticercosis (BCC) are combined in a Bayesian analysis.•34% of Belgian cattle are BCC positive for one or more detection techniques.•Feasibility of Ag-ELISA as detection technique at slaughter needs further research. A Bayesian model was developed to estimate values for the prevalence and diagnostic test characteristics of bovine cysticercosis (Taenia saginata) by combining results of four imperfect tests. Samples of 612 bovine carcases that were found negative for cysticercosis during routine meat inspection collected at three Belgian slaughterhouses, underwent enhanced meat inspection (additional incisions in the heart), dissection of the predilection sites, B158/B60 Ag-ELISA and ES Ab-ELISA. This Bayesian approach allows for the combination of prior expert opinion with experimental data to estimate the true prevalence of bovine cysticercosis in the absence of a gold standard test. A first model (based on a multinomial distribution and including all possible interactions between the individual tests) required estimation of 31 parameters, while only allowing for 15 parameters to be estimated. Including prior expert information about specificity and sensitivity resulted in an optimal model with a reduction of the number of parameters to be estimated to 8. The estimated bovine cysticercosis prevalence was 33.9% (95% credibility interval: 27.7–44.4%), while apparent prevalence based on meat inspection is only 0.23%. The test performances were estimated as follows (sensitivity (Se) – specificity (Sp)): enhanced meat inspection (Se 2.87% – Sp 100%), dissection of predilection sites (Se 69.8% – Sp 100%), Ag-ELISA (Se 26.9% – Sp 99.4%), Ab-ELISA (Se 13.8% – Sp 92.9%).
ISSN:0304-4017
1873-2550
DOI:10.1016/j.vetpar.2018.03.013