Table 2 Bayesian diagnostic accuracy parameter estimates for various combinations of the biomarkers without covariate adjustment.

From: Application of latent class analysis in diagnosis of graft-versus-host disease by serum markers after allogeneic haematopoietic stem cell transplantation

Composite test

cut-off

Accuracy Parameters

Mean

SD

Median

95% CrI

LDH and Cr

31

     
  

cSensitivity

0.57

0.11

0.58

(0.53–0.64)

  

cSpecificity

0.43

0.12

0.46

(0.39–0.52)

  

cAUC

0.55

0.14

0.56

(0.50–0.65)

LDH and UA

11

     
  

cSensitivity

0.56

0.10

0.55

(0.51–0.60)

  

cSpecificity

0.41

0.14

0.43

(0.37–0.52)

  

cAUC

0.54

0.13

0.55

(0.48–0.62)

UA and Cr

10

     
  

cSensitivity

0.54

0.14

0.58

(0.50–0.65)

  

cSpecificity

0.46

0.11

0.47

(0.41–0.53)

  

cAUC

0.56

0.13

0.55

(0.51–0.65)

UA and Cr and LDH

32

     
  

cSensitivity

0.58

0.12

0.57

(0.53–0.65)

  

cSpecificity

0.49

0.10

0.48

(0.43–0.52)

  

cAUC

0.61

0.11

0.62

(0.57–0.68)

  1. The calculation of cut-off points for the markers’ combinations is as the same as one marker. In fact, by combining the markers’ results, only one composite test is obtained. Notes: SD: standard deviation; CrI: credible interval; LDH: lactate dehydrogenase; UA: uric acid; Cr: creatinine; cSensitivity: combined Sensitivity; cSpecificity: combined Specificity; cAUC: combined area under receiver operating characteristic curve.
Back to article page