Table 2 |
||||||
| Agreement between three serological tests used for the detection of H5 antibodies in ducks experimentally infected with H5N1 | ||||||
| Test comparison | 8 dpi | 15 dpi | 22 dpi | 29 dpi | 34 dpi | Overall Kappa |
| HI-CRBC and HI-HRBC: | ||||||
| Kappa for dichotomised results (95%-CI)1 | 0.44 (0.01-0.88) | 1 (1.00-1.00) | 0.74 (0.27-1.00) | 1 (1.00-1.00) | 1 (1.00-1.00) | 0.86 (0.73-0.99) |
| Weighted kappa for log titres (95%-CI)2 | 0.56 (0.30-0.83) | 0.85 (0.65-1.00) | 0.61 (0.32-0.77) | 0.85 (0.70-1.00) | 0.66 (0.49-0.84) | 0.8 (0.71-0.85) |
| HI-CRBC and VN: | ||||||
| Kappa for dichotomised results (95%-CI)1 | 0.55 (0.04-1.00) | 1 (1.00-1.00) | 0.38 (−0.33-1.00) | 0.74 (0.27-1.00) | 0.41 (−0.18-1.00) | 0.76 (0.60-0.93) |
| Weighted kappa for log titres (95%-CI)2 | 0.49 (0.27-0.71) | 0.76 (0.46-0.95) | 0.50 (0.25-0.66) | 0.36 (0.02-0.52) | 0.41 (0.24-0.65) | 0.68 (0.58-0.74) |
| HI-HRBC and VN: | ||||||
| Kappa for dichotomised results (95%-CI)1 | 0.19 (−0.10-0.49) | 1 (1.00-1.00) | 0.74 (0.27-1.00) | 0.74 (0.27-1.00) | 0.41 (−0.18-1.00) | 0.69 (0.50-0.88) |
| Weighted kappa for log titres (95%-CI)2 | 0.19 (0.01-0.46) | 0.71 (0.41-0.93) | 0.41 (0.12-0.61) | 0.37 (0.17-0.58) | 0.36 (0.12-0.63) | 0.55 (0.42-0.63) |
Ten ducks were sampled before inoculation with H5N1 virus and at 8, 15, 22, 29 and 34 days post inoculation (dpi). The kappa statistic (95% confidence interval1) was used to compare dichotomised test results (positive versus negative) and a weighted kappa statistic (95% confidence interval2) was used to compare log titers at each sampling period and across all sampling periods.
1The analytical method was used to calculate confidence intervals for dichotomised test results [35].
2The bootstrap method with 100 bootstrap replications was used to calculate confidence intervals for the log titers [36,37].
Abbreviations: HI-CRBC (Haemagglutination inhibition test using chicken red blood cells), HI-HRBC (Haemagglutination inhibition test using horse red blood cells), VN (Virus neutralisation test), CI (confidence interval).
Wibawa et al. BMC Veterinary Research 2012 8:117 doi:10.1186/1746-6148-8-117
