BioMed Research International / 2019 / Article / Tab 7

Research Article

Comparison of Two Bayesian Methods in Evaluation of the Absence of the Gold Standard Diagnostic Tests

Table 7

The posterior estimation of four models using the TB data under conditional dependence situation.

SituationMethodModelKnotMeanSDMedian95% Bayesian CI
(P25-P75)

No prior constraintsBayesian probabilistic constraint modelNP0.6260.2180.6650.516-0.773
0.3690.2170.3300.222-0.476
0.4900.1670.4950.391-0.585
0.5080.1670.5030.414-0.616
0.4980.2480.4970.305-0.692
Conditional
Covariance
Bayesian model
NC0.6900.1830.7080.622-0.807
0.4350.2260.3900.287-0.568
0.5470.1970.5350.455-0.657
0.5710.2130.5620.444-0.722
0.5530.2540.5780.366-0.755

Prior constraintsBayesian probabilistic constraint modelPP0.7130.0360.7140.689-0.738
0.4230.0510.4210.387-0.456
0.5350.0250.5350.518-0.552
0.5370.0220.5370.522-0.552
0.5380.0420.5390.510-0.567
Conditional
Covariance
Bayesian model
PC0.9040.0370.9070.881-0.931
0.7960.1190.8140.715-0.892
0.5880.0550.5800.551-0.614
0.6770.0680.6770.631-0.723
0.6490.0760.6620.608-0.701

: sensitivity of T-SPOT; : specificity of T-SPOT; : sensitivity of KD38; : specificity of KD38;: prevalence within the patients in the study; SD: standard deviation; CI: confidence interval.