Statistical analysis
Data were entered into Excel and analysed using SAS version 9.4 (SAS
Institute Inc). The results of the three tests were compared using
cross-tabulations. ELISA indeterminate results (OD 0.9−0.99 nm) were
considered “not positive”. Sensitivity (positive result in a true
positive case (TP)), specificity (negative result in a true negative
case (TN)), positive and negative predictive values (PPV, NPV) (the
probability of positive or negative results in true positive or negative
cases, respectively) and positive and negative likelihood ratios (LRP
and LRN) were calculated for the rapid test and Platelia ELISA using the
Wantai ELISA as the reference test, which has shown superior performance
in previous evaluations against RT-PCR confirmed samples [3,4].
Standard formulae were used: sensitivity (TP)/(TP + FN) (FN = false
negative); specificity (TN)/(TN + FP) (FP = false positive; positive
predictive value (TP/TP + FP); negative predictive value (TN/TN + FN);
and positive and negative likelihood ratios (LR+, LR-) (probability that
a person with or without the disease tested positive respectively; and
probability that a person with or without the disease tested negative
respectively). Results were calculated within 95% confidence intervals.
The Kappa coefficient of agreement between the rapid test and the Wantai
ELISA, and between the Platelia ELISA and the Wantai ELISA, were
interpreted according to the criteria of Cohen (values ≤0 indicating no
agreement; 0.01–0.20 as none to slight agreement; 0.21–0.40 as fair
agreement; 0.41–0.60 as moderate agreement; 0.61–0.80 as substantial
agreement; 0.81–1.00 as almost perfect agreement) [5]. The results
of the total population were compared with those of the non-vaccinated
population.