Modelling of Infectious Disease CEntre
Imperfect Gold Standard Models
Subscribe :
Register
|
Forget Password
MODEL106 : 4-tests in 1-population Model (Advance Interface)
Home
Choose Models
Job list
Publication
Contact us
General Description
|
Model Designation
|
Use Model on Web
|
Download model
|
Publication and Other Information
Please
regsiter for subscribe
before submit parameter and run the model
If already subscribe please enter user name and password on top of the page
Data input
Test A
Test B
Test C
Test D
Frequency observed
Positive
Positive
Positive
Positive
Positive
Positive
Positive
Negative
Positive
Positive
Negative
Positive
Positive
Positive
Negative
Negative
Positive
Negative
Positive
Positive
Positive
Negative
Positive
Negative
Positive
Negative
Negative
Positive
Positive
Negative
Negative
Negative
Negative
Positive
Positive
Positive
Negative
Positive
Positive
Negative
Negative
Positive
Negative
Positive
Negative
Positive
Negative
Negative
Negative
Negative
Positive
Positive
Negative
Negative
Positive
Negative
Negative
Negative
Negative
Positive
Negative
Negative
Negative
Negative
Correlation between diagnostic tests
(Using example value)
No correlation between diagnostic tests
Correlation between test A and test B in diseased population
Correlation between test A and test B AND Correlation between test C and test D in diseased population
Correlation between test A and test B in non-diseased population
Correlation between test A and test B AND Correlation between test C and test D in non-diseased population
In diseased population
In non-diseased population
Correlation between test A and test C
Correlation between test A and test D
Correlation between test B and test C
Correlation between test B and test D
Correlation between test C and test D
Correlation between test A, test B and test C
Correlation between test A, test B and test D
Correlation between test A, test C and test D
Correlation between test B, test C and test D
Correlation between test A and test C
Correlation between test A and test D
Correlation between test B and test C
Correlation between test B and test D
Correlation between test C and test D
Correlation between test A, test B and test C
Correlation between test A, test B and test D
Correlation between test A, test C and test D
Correlation between test B, test C and test D
Prior distributions
(Using example value)
Parameters
Options
Values
Prevalence
Default
Adjust manually
Beta distribution
Beta distribution (0.5,0.5)
Range (0,1)
Beta distribution
Range
(
,
)
(
,
)
Fix the value
(
)
Sensitivity of test A
Default
Adjust manually
Beta distribution
Fix the value
Beta distribution (0.5,0.5)
Range (0,1)
Beta distribution
Range
(
,
)
(
,
)
Fix the value
(
)
Sensitivity of test B
Default
Adjust manually
Beta distribution
Fix the value
Beta distribution (0.5,0.5)
Range (0,1)
Beta distribution
Range
(
,
)
(
,
)
Fix the value
(
)
Sensitivity of test C
Default
Adjust manually
Beta distribution
Fix the value
Beta distribution (0.5,0.5)
Range (0,1)
Beta distribution
Range
(
,
)
(
,
)
Fix the value
(
)
Sensitivity of test D
Default
Adjust manually
Beta distribution
Fix the value
Beta distribution (0.5,0.5)
Range (0,1)
Beta distribution
Range
(
,
)
(
,
)
Fix the value
(
)
Specificity of test A
Adjust manually
Beta distribution
Fix the value
Beta distribution (0.5,0.5)
Range (0.4,1)
Beta distribution
Range
(
,
)
(
,
)
Fix the value
(
)
Specificity of test B
Default
Adjust manually
Beta distribution
Fix the value
Beta distribution (0.5,0.5)
Range (0.4,1)
Beta distribution
Range
(
,
)
(
,
)
Fix the value
(
)
Specificity of test C
Default
Adjust manually
Beta distribution
Fix the value
Beta distribution (0.5,0.5)
Range (0.4,1)
Beta distribution
Range
(
,
)
(
,
)
Fix the value
(
)
Specificity of test D
Default
Adjust manually
Beta distribution
Fix the value
Beta distribution (0.5,0.5)
Range (0.4,1)
Beta distribution
Range
(
,
)
(
,
)
Fix the value
(
)
Note!
Model default value of specificity of Test A is fixed to 100%. The dataset is taken from a study of diagnostic tests for melioidosis. Test A is culture, and a positive culture of B. pseudomallei is a definite indication of melioidosis infection. Please refer to
Limmathurotsakul D, et al. PLoS One 2010;5:e12485
for more details.
Initial values
(Using example value)
Note!
The
prevalance
is set to a fixed value in the above section; therefore, its initial will not be taken into account during Bayesian LCM analysis.
The
sensitivity of test A
is set to a fixed value in the above section; therefore, its initial will not be taken into account during Bayesian LCM analysis.
The
sensitivity of test B
is set to a fixed value in the above section; therefore, its initial will not be taken into account during Bayesian LCM analysis.
The
sensitivity of test C
is set to a fixed value in the above section; therefore, its initial will not be taken into account during Bayesian LCM analysis.
The
sensitivity of test D
is set to a fixed value in the above section; therefore, its initial will not be taken into account during Bayesian LCM analysis.
The
specificity of test A
is set to a fixed value in the above section; therefore, its initial will not be taken into account during Bayesian LCM analysis.
The
specificity of test B
is set to a fixed value in the above section; therefore, its initial will not be taken into account during Bayesian LCM analysis.
The
specificity of test C
is set to a fixed value in the above section; therefore, its initial will not be taken into account during Bayesian LCM analysis.
The
specificity of test D
is set to a fixed value in the above section; therefore, its initial will not be taken into account during Bayesian LCM analysis.
Parameters
Options
Values
Prevalence
Default
Adjust manually
Initial 1 (0.9)
Initial 2 (0.3)
Initial 1
Initial 2
(
)
(
)
Sensitivity of test A
Default
Adjust manually
Initial 1 (0.9)
Initial 2 (0.7)
Initial 1
Initial 2
(
)
(
)
Sensitivity of test B
Default
Adjust manually
Initial 1 (0.9)
Initial 2 (0.7)
Initial 1
Initial 2
(
)
(
)
Sensitivity of test C
Default
Adjust manually
Initial 1 (0.9)
Initial 2 (0.7)
Initial 1
Initial 2
(
)
(
)
Sensitivity of test D
Default
Adjust manually
Initial 1 (0.9)
Initial 2 (0.7)
Initial 1
Initial 2
(
)
(
)
Specificity of test A
Default
Adjust manually
Initial 1 (0.9)
Initial 2 (0.99)
Initial 1
Initial 2
(
)
(
)
Specificity of test B
Default
Adjust manually
Initial 1 (0.9)
Initial 2 (0.99)
Initial 1
Initial 2
(
)
(
)
Specificity of test C
Default
Adjust manually
Initial 1 (0.9)
Initial 2 (0.99)
Initial 1
Initial 2
(
)
(
)
Specificity of test D
Default
Adjust manually
Initial 1 (0.9)
Initial 2 (0.99)
Initial 1
Initial 2
(
)
(
)
Covalence in test A and B
Default
Adjust manually
Initial 1 (0.00001)
Initial 2 (0.00001)
Initial 1
Initial 2
(
)
(
)
Covalence in test C and D
Default
Adjust manually
Initial 1 (0.00001)
Initial 2 (0.00001)
Initial 1
Initial 2
(
)
(
)
Settings
(Using example value)
Parameters
Options
Values
Iterations
Default
Adjust manually
Burn-in iterations (5000)
MCMC Iterations (20000)
Burn-in iterations
MCMC Iterations
(
)
(
)
Thin
Default
Adjust manually
thin (10)
thin
(
)
Conventional method to be compared with the imperfect gold standard model
(Using example value)
Test A was assumed as a perfect gold standard
Test B was assumed as a perfect gold standard
Test C was assumed as a perfect gold standard
Test D was assumed as a perfect gold standard
Positive results of either test A or B was assumed as perfect gold standard
Positive results of either test A or C was assumed as perfect gold standard
Positive results of either test A or D was assumed as perfect gold standard
Positive results of either test B or C was assumed as perfect gold standard
Positive results of either test B or D was assumed as perfect gold standard
Positive results of either test C or D was assumed as perfect gold standard
Positive results of both test A and B was assumed as perfect gold standard
Positive results of both test A and C was assumed as perfect gold standard
Positive results of both test A and D was assumed as perfect gold standard
Positive results of both test B and C was assumed as perfect gold standard
Positive results of both test B and D was assumed as perfect gold standard
Positive results of both test C and D was assumed as perfect gold standard
MICE is funded by Li Ka Shing and Wellcome trust, and initiated under collaboration between Mahidol-Oxford Tropical Medicine Research Unit (MORU) and Faculty of Tropical Medicine, Mahidol University, Thailand.
The content of this site is © 2011 MORU but may be used under the terms of the Creative Commons Attribution Non-Commercial Licence which permits unrestricted use, distribution, and reproduction for non-commercial purposes and provided the original work is properly cited.