Performance and usability of INSTI, a blood-based rapid HIV self test for qualitative detection of HIV antibodies in intended use populations in Kenya
Mwau et al.
Type of approach
Type of assistance
Mixed: General population and people with HIV
UNAIDS HIV prevalence (2017)
5.9 [4.9 - 7.0]
From 22 March to 11 Apr 2017, a usability study was conducted to assess the INSTI HIV Self Test. 354 participated in the label comprehension (usability) study and 91 subjects completed the results interpretation study. All participants were >18 years, able to provide informed consent, and able to read, write and speak English/Swahili. HIV self-test result was compared to fingerprick and venous blood sent to KEMRI Lab for Bioelisa HIV-1+2 Ag/Ab kit (Biokit S.A, Barcelona, Spain). INSTI HIV Self Test results produced by subjects and interpreted and recorded by a health-care worker.
Summary of findings
There were six invalid results. Blood-based HIVST sensitivity was 198/201 = 98.5% (95.7-99.5%) and blood-based HIVST specificity: 267/269 = 99.26% (97.3-99.8%). Using only the instructions-for-use, 98.9% prepared the fingerstick, 83.1% used the lancet with no assistance (16% needed assistance to perform lancet). 99.4% able to do a hanging drop of blood with no assistance, 99.7% able to add diluent and 99.7% correctly completed the INSTI test procedure. 98% used the instructions, 96.9% found the instructions easy to understand, 94.3% considered INSTI easy to perform, 97.7% found results easy to interpret and were willing to use it again, and 98% would recommend INSTI to a partner or family. Weak positives, however, were difficult to interpret with 31/91 (34%) correctly interpreting the result.
97.7% found results easy to interpret, were willing to use it again and 98% would recommend INSTI to a partner or family.
Willingness to pay
Willingness to pay details
Sensitivity 98.5% (95.7-99.5%), specificity 99.26% (97.3-99.8%).
Linkage to prevention, care and treatment