Can HIV incidence testing be used for evaluating HIV intervention programs? A reanalysis of the Orange Farm male circumcision trial (ANRS-1265)

Table 5 Comparison of the results given by the BED and the AI assays with observed HIV incidence cases obtained from HIV testing

Visit A
Label	V1	V3	V1	V2
HIV positive cases	127	148	124	124
HIV negative cases	11	35	32	56
Visit B
Label	V2	V4	V3	V4
HIV positive cases	138	183	156	180
¹HIV negative cases	0	0	0	0
From visit A to visit B
Average duration	3.6 m	8.8 m	12.6 m	18.0 m
HIV incident cases	11	35	32	56
BED
Cut-off	0.325	0.90	1.51	2.27
Window period	3.0 m	9.0 m	12.0 m	18.0 m
Tested recent	12	43	54	110
Tested recent not incident case	6	25	21	53
Tested recent and not incident case	6	18	33	57
Not tested recent and incident cases	5	10	11	3
Kappa statistics [95% CI]	0.64 [0.41 to 0.81]	0.55 [0.40 to 0.69]	0.31 [0.16 to 0.46]	0.39 [0.26 to 0.51]
Sensitivity (%) [95% CI]	55 [28 to 79]	71 [56 to 84]	66 [49 to 80]	95 [87 to 99]
Specificity (%) [95% CI]	95 [91 to 98]	88 [82 to 92]	73 [65 to 81]	54 [45 to 63]
AV
Cut-off	31.6	90	94	NC
Window period	3.0 m	9.0 m	12.0 m
Tested recent	11	49	44
Tested recent and incident case	8	28	20
Tested recent and not incident case	3	21	24
Not tested recent and incident cases	3	7	12
Kappa statistics [95% CI]	0.48 [0.31 to 0.64]	0.57 [0.43 to 0.71]	0.38 [0.23 to 0.53]
Sensitivity (%) [95% CI]	73 [45 to 92]	80 [65 to 91]	63 [46 to 78]
Specificity (%) [95% CI]	98 [94 to 99]	86 [79 to 91]	81 [73 to 87]
BED-AI
Cut-off	²0.40 - 41	1.51 - 94	NC	NC
Window period	9.0 m	8.8 m
Tested recent	10	38
Tested recent and incident case	7	27
Tested recent and not incident case	3	11
Not tested recent and incident cases	4	8
Kappa statistics [95% CI]	0.70 [0.54 to 0.87]	0.68 [0.53 to 0.82]
Sensitivity (%) [95% CI]	64 [36 to 86]	77 [62 to 89]
Specificity (%) [95% CI]	98 [94 to 99]	93 [88 to 96]

ISSN: 1471-2334