Table 3.

Performances and differences in performances among study sites and RDTs

Test parameter	Site	SD Bioline HRP2		CareStart HRP2		CareStart pLDH		SD Bioline HRP2 vs CareStart HRP2		SD Bioline HRP2 vs CareStart pLDH		CareStart HRP2 vs CareStart pLDH
		%	[95 % CI]	%	[95 % CI]	%	[95 % CI]	Diff %	[95 % CI]	Diff  %	[95 % CI]	Diff  %	[95 % CI]
Sensitivity	Mbarara	98.4	[96.0 to 99.6]	98.8	[96.6 to 99.8]	96.1	[92.9 to 98.1]	−0.4	[−1.5 to 0.8]	2.3	[−0.2 to 4.9]	2.7	[0.3 to 5.1]*
	Kazo	99.2	[97.3 to 99.9]	98.9	[96.7 to 99.8]	94.7	[91.2 to 97.0]	0.4	[−0.7 to 1.5]	4.6	[1.7 to 7.5]***	4.2	[1.2 to 7.2]**
	Diff	−0.8	[−2.6 to 1.0]	0.0	[−1.9 to 1.8]	1.4	[−2.2 to 5.0]
Specificity	Mbarara	98.9	[98.5 to 99.2]	98.8	[98.5 to 99.1]	99.8	[99.6 to 99.9]	0.0	[−0.1 to 0.2]	−0.9	[−1.2 to −0.6]***	−0.9	[−1.2 to −0.6]***
	Kazo	79.7	[73.4 to 85.1]	80.7	[74.5 to 86.0]	93.9	[89.6 to 96.8]	−1.0	[−2.9 to 0.9]	−14.2	[−19.8 to −8.6]***	−13.2	[−18.6 to −7.8]***
	Diff	19.2	[13.6 to 24.8]$$$	18.1	[12.6 to 23.6]$$$	5.8	[2.5 to 9.2]$$$
Misclassification	Mbarara	1.1	[0.9 to 1.5]	1.2	[0.9 to 1.5]	0.4	[0.3 to 0.7]	0.0	[−0.1 to 0.1]	0.7	[0.4 to 1.0]***	0.7	[0.4 to 1.0]***
	Kazo	9.2	[6.7 to 12.2]	8.9	[6.5 to 11.9]	5.7	[3.7 to 8.2]	0.2	[−0.7 to 1.2]	3.5	[0.5 to 6.5]*	3.3	[0.3 to 6.2]*
	Diff	−8.0	[−10.7 to −5.4]$$$	−7.8	[−10.4 to −5.1]$$$	−5.2	[−7.4 to −3.1]$$$
Positive predictive value	Mbarara	83.2	[78.5 to 87.2]	82.7	[70.0 to 86.7]	95.7	[92.5 to 97.8]	–		–		–
	Kazo	86.7	[82.3 to 90.3]	87.2	[82.9 to 90.8]	95.4	[92.1 to 97.6]	–		–		–
	Diff	−3.5	[−9.2 to 2.2]	−4.5	[−10.2 to 1.2]	0.3	[−3.2 to 3.9]
Negative predictive value	Mbarara	99.9	[99.8 to 100.0]	99.9	[99.8 to 100.0]	99.8	[99.6 to 99.9]	–		–		–
	Kazo	98.7	[95.5 to 99.8]	98.1	[94.7 to 99.6]	93.0	[88.5 to 96.1]	–		–		–
	Diff	1.2	[−0.6 to 2.9]$	1.8	[−0.3 to 3.9]$	6.8	[3.3 to 10.4]$$$

* Exact McNemar’s p value <0.05

** Exact McNemar’s p value <0.01

*** Exact McNemar’s p value <0.001

^$ Fisher’s exact p value <0.05

^$$ Fisher’s exact p value <0.01

^$$$ Fisher’s exact p value <0.001