Search HIVST

Evaluation of the practicability and virological performance of finger-stick whole-blood HIV self-testing in French-speaking sub-Saharan Africa

Tonen-Wolyec et al.

Publication year

Type of approach

Type of assistance

Fingerstick/whole blood

Study population
General population: Male and female adult population

Study design

Sample size

UNAIDS HIV prevalence (2017)

Practicability evaluation of a blood/finger-stick HIV self-test comprised a multicenter cross-sectional study performed between March and June 2016 in Kisangani and Bunia, the capital cities of the DRC provinces of Tshopo and Ituri. The evaluation consisted of face-to-face and self-administered questionnaires with participants recruited from five voluntary testing and counselling sites. Participants were patients or visitors to the clinic or healthcare workers. Blood was drawn from each participant and serological HIV testing was conducted. After completing questionnaires on the HIV self-test instructions and interpretation of results, participants were conducted an HIV self-test (HIVST) under supervision from a researcher who recorded each step and any appeal for verbal assistance.

Summary of findings
Overall, 79.6% (n=256) of participants correctly understood the instructions for use, measured by answering at least five questions correctly. Comprehension of the instructions for use was associated with educational level and language use. The age of participants was also associated with comprehension of the instructions for use. Thus, 79.7% (n = 197) of participants between 18 to 29 years correctly interpreted the instructions for use, as did 87.5% (n = 80) and 64.4% (n = 45) of those between 30 to 39 years and 40 years or more, respectively (P < 0.01). Most (96.6%, n=n/a) participants found that the instructions for use were easy to understand. A large majority (85.4%, n=n/a) of participants had a favorable opinion of the possibility to use the instructions for use in vernacular languages. total of 4,186 standardized tests were read and interpreted by the 322 participants; 90.2% (n=3,777) of tests were correctly interpreted, whereas 9.8% (n=409) of tests were misinterpreted. In multivariate logistic regression analysis, higher educational level remained associated with the correct interpretation of positive and invalid test results. When observed conducting an HIV self-test (HIVST), 98.4% (n=317) of participants correctly used the self-test and succeeded in obtaining an interpretable result. The correct use of HIVST was observed in 98.3% (n=n/a) of the general public and 100% (n=n/a) of health care workers (P > 0.99). In total, 79.2% (n=255) of participants carried out the self-test autonomously, whereas 20.8% (n=67) asked for oral assistance, especially regarding use of the lancing. The level of education influenced the need of oral assistance, but did not influence the correct performance of HIVST.


Acceptability details

Willingness to pay

Willingness to pay details




HIV positivity
17% (n=55/322)

Accuracy details
The Cohen's K coefficient between the results of reading by participants and the expected results was 0.84, demonstrating excellent concordance according to Landis and Koch's rank.

Social harm

Linkage to prevention, care and treatment


Study status