Skip to main content

Table 3 Laboratory evaluation of the Deki Reader comparing device versus SOC results

From: Continuous quality monitoring in the field: an evaluation of the performance of the Fio Deki Reader™ for rapid HIV testing in South Africa

Deki Reader Version

RDT brand

Result compared to SOC

aConcordance n (%)

HIV positive

HIV negative

aSensitivity

% (95%CI)

aSpecificity

% (95%CI)

Deki & HCW invalid (true)

Deki invalid (false)

True

False

True

False

bV100

First Response® HIV-1-2-0

(n = 253)

128

118

100

(97.2–100)

100

(96.9–100)

7

253 (100)

Determine™ HIV-1/2 Ag/Ab Combo (n = 259)

135

c 2

94

d 1

99.3

(96.0–100)

97.9

(92.7–99.8)

27

256 (98.8)

e Uni-Gold HIV

(n = 262)

132

127

d 1

99.3

(95.9–100)

100

(97.1–100)

259 (99.6)

Overall (n = 772)

395

2

339

2

99.5

(98.2–99.9)

99.4

(97.9–99.9)

34

768 (99.5)

V200

First Response® HIV-1-2-0 (n = 158)

113

2

43

100

(96.8–100)

95.6

(84.9–99.5)

156 (98.7)

One Step Anti-HIV (1&2) (n = 158)

109

2

40

100

(96.7–100)

95.2

(83.8–99.4)

1

f 6

150 (94.9)

HIV 1/2/0 Tri-Line

(n = 158)

114

5

39

100

(96.8–100)

88.6

(75.4–96.2)

153 (96.8)

Overall (n = 474)

336

9

122

100

(98.9–100)

93.1

(87.4–96.8)

1

6

459 (96.8)

  1. aConcordance includes invalid specimens, but excludes errors and specimens that were not processed. Sensitivity and specificity only includes specimens that gave a valid positive or negative result
  2. bDeki Reader data was adjudicated by Fio as part of the process to optimise the output data (software optimisation)
  3. cAir bubble over control line, linked to RDT
  4. dLine < 50% of width, software adjustments required
  5. eDeki Reader v100 error (n = 2)
  6. fDeki Reader v200 invalid control line