Skip to main content
. 2015 Oct 27;5:15284. doi: 10.1038/srep15284

Table 3. Weighted meta-regression of the effects of study settings, methods, and methodological quality on diagnostic accuracy of interferon-γ release assays.

Covariate Number of studies Co-efficient RDOR P-value
Body Fluid
QUADAS score
≥11 8 –0.46 0.63 (0.14–2.92) 0.537
<11 13      
Setting        
 High TB burden 8 –0.41 0.67 (0.14–3.13) 0.591
 Low TB burden 15      
Method        
 ELISPOT 15 –0.44 0.64(0.13–3.09) 0.437
 ELISA 8      
Disease        
 Pleurisy 14 –0.43 0.65 (0.07–6.49) 0.701
 Peritonitis 2 0.54 1.72 (0.05–57.87) 0.751
 Meningitis 4 –0.54 0.58 (0.03–9.75) 0.693
 Pericarditis 1 dropped dropped  
Peripheral Blood
QUADAS score        
≥11 8 0.67 1.95 (0.91–4.18) 0.082
<11 11      
Settings        
 High TB burden 9 –0.23 0.80 (0.38–1.65) 0.519
 Low TB burden 12      
Method        
 ELISPOT 13 –0.32 0.73 (0.34–1.58) 0.402
 ELISA 8      
Disease        
 Pleurisy 12 –1.36 0.26 (0.05–1.32) 0.098
 Peritonitis 2 dropped dropped  
 Meningitis 4 –1.99 0.14 (0.02–0.79) 0.029
 Mixed* 3 –1.61 0.20 (0.04–1.12) 0.066

*Studies contained different extrapulmonary tuberculosis types, including pleurisy, peritonitis, meningitis, and pericarditis.