Skip to main content

Table 4 Influence of the prior distributions of sensitivities on posterior distributions

From: A two-stage Bayesian method for estimating accuracy and disease prevalence for two dependent dichotomous screening tests when the status of individuals who are negative on both tests is unverified

 

Sei ~ U(0, 1)

Sei ~ U(0.1, U(0.2, 0.99))

Sei ~ U(0.8, U(0.9, 0.99))

Median (95% BCI)

Median (95% BCI)

Median (95% BCI)

Ï€

0.0102 (0.0044, 0.1793)

0.0130 (0.0056, 0.0323)

0.0055 (0.0038, 0.0075)

Se1

0.3053 (0.0075, 0.9666)

0.2461 (0.1065, 0.1065)

0.8691 (0.803, 0.9612)

Se2

0.2658 (0.0055, 0.9397)

0.2414 (0.1057, 0.7268)

0.8675 (0.8032, 0.9596)

Sp1

0.9805 (0.9740, 0.9851)

0.9806 (0.9762, 0.9848)

0.9818 (0.9781, 0.9853)

Sp2

0.9540 (0.9434, 0.9595)

0.9545 (0.9494, 0.9594)

0.9561 (0.9516, 0.9604)

SeJE

0.5162 (0.0293, 0.9861)

0.3996 (0.1703, 0.8295)

0.9466 (0.8561, 0.9899)

SpJE

0.9398 (0.9276, 0.9445)

0.9401 (0.9354, 0.9446)

0.9405 (0.9360, 0.9449)

  1. Note: U, uniform distribution; BCI, Bayesian credible interval.