Table 2.

Accuracy of the different diagnostic tests compared against a composite reference standard in a group of 92 asymptomatic Eritrean refugees

Diagnostic test and location	Outcome	CRS*, n		Diagnostic accuracy, %		Cohen’s κ			McNemar’s P value
		Positive	Negative	Sensitivity	Specificity	κ value	Interpretation†	P value
LUMC, Leiden, the Netherlands
CAA (urine)	Positive	36	4	92	92	0.845	Almost perfect	< 0.001	1.000
	Negative	3	49
CAA (serum)	Positive	37	0	95	100	0.955	Almost perfect	< 0.001	0.500
	Negative	2	53
PCR (stool)‡	Positive	22	1	56	98	0.576	Moderate	< 0.001	< 0.001
	Negative	17	52
POC-CCA§	Positive	24	4	62	92	0.561	Moderate	< 0.001	0.019
	Negative	15	49
Swiss TPH, Basel, Switzerland‖
Microscopy (stool)¶	Positive	22	0	56	100	0.599	Moderate	< 0.001	< 0.001
	Negative	17	53
Serology	Positive	31	8	79	85	0.644	Substantial	< 0.001	1.000
	Negative	8	45
POC-CCA§	Positive	29	10	74	81	0.555	Moderate	< 0.001	1.000
	Negative	10	43

CAA = circulating anodic antigen; CRS = composite reference standard; LUMC = Leiden University Medical Center; PCR = polymerase chain reaction; POC-CCA = point-of-care circulating cathodic antigen; Swiss TPH = Swiss Tropical and Public Health Institute.

^*The CRS was based on the detection of eggs in stool and/or CAA in urine/serum and/or DNA in stool. An individual was considered positive if either microscopy was positive or at least two of the other diagnostic tests were positive.

^†Interpretation of κ coefficient: ≤0, chance; 0.01 to 0.20, slight; 0.21 to 0.40, fair; 0.41 to 0.60, moderate; 0.61 to 0.80, substantial; 0.81 to 0.99, almost perfect.

^‡All urine samples were negative by urine PCR.

^§The same POC-CCA batch was used at LUMC and Swiss TPH (no. 50182), but a different scoring approach was used; see Materials and Methods.

^‖Data available from Swiss TPH.

^¶All urine samples were negative by urine microscopy.