HIV Self-Tests as Accurate as Tests by Healthcare Workers

Despite doubts, a new study has found that self-administered home HIV tests may give just as accurate results as tests conducted by healthcare workers. Results were published in Lancet HIV.¹

“Self-testers could reliably and accurately do an HIV rapid diagnostic test, whether assistance was provided or not, when compared with a trained health-care worker,” wrote first author Carmen Figueroa, MPH, of the World Health Organization (Geneva, Switzerland), and colleagues.

In HIV self-testing, an individual performs her/his own specimen collection using an oral swab or finger prick, does the test, and interprets the results at home. Self-testing is one way to increase access and overcome barriers to HIV testing due to privacy concerns, stigma, discrimination, and (in some countries) criminalization.

Despite studies showing that HIV testing is highly acceptable for a broad range of people, doubts still exist about whether non-healthcare workers can perform the test and interpret the results correctly on their own.

To inform 2016 WHO recommendations about HIV self-testing, researchers searched PubMed, PopLine, Embase, six conference databases, and the gray literature for studies published between January 1995 and April 2016, without restriction to study type, language, or geographic location. Studies on self-testing included assisted approaches (in-person demonstration of how to do the test), and unassisted approaches (only the manufacturer’s instructions). The analysis included 25 studies, covering up to 5662 participants.

Key results
• Significant agreement between healthcare workers and self-testers, ranging from fair to almost perfect
• Raw proportion of agreement ranged from 85.4% to 100%
• Almost perfect agreement for assisted and unassisted self-test approaches
• Most common error was incorrect specimen collection in self-testers

In studies of self-testers, the sensitivity (true positive rate) ranged from 80% to 100%, and specificity (true negative rate) ranged from 95.1% to 100%. The authors estimated that applying these ranges to a hypothetical group of 100,000 people with an HIV prevalence of 1% would result in 0 to 200 positive HIV cases being missed, and 0 to 4851 negative HIV cases being misdiagnosed as positive.

To improve the accuracy of HIV self-tests, the authors made several recommendations: simple instructions (with adaptations to the local culture/language), clear product labels, easily identifiable kit components, additional user support (such as with videos), and further improvement in the design of self-tests.

In conclusion, they stressed that the instructions should explain the limitations of testing during the window period, and that people with positive or invalid results should have further testing at healthcare facility.

Take-home points
• WHO meta-analysis found fair to almost perfect agreement between HIV self-tests and HIV tests done by healthcare workers
• For studies of HIV self-tests, sensitivity ranged from 80% to 100%; sensitivity ranged from 95.1% to 100%
• There was almost perfect agreement for assisted and unassisted self-test approaches
• The most common error among HIV self-testers was incorrect specimen collection

References:

1. Figueroa C, Johnson C, Ford N, et al. Reliability of HIV rapid diagnostic tests for self-testing compared with testing by health-care workers: a systematic review and meta-analysis. Lancet HIV. 2018 Apr 24. pii: S2352-3018(18)30044-4. doi: 10.1016/S2352-3018(18)30044-4.