Skip to main content

New screening tool offers cost-effective way to diagnose more HIV-positive adolescents

Hester Phillips

10 October 2018

A simple questionnaire has been developed and validated in Harare, Zimbabwe, to pre-screen adolescents for HIV testing.

School girl in Zimbabwe
Photos are used for illustrative purposes. They do not imply health status or behaviour. Credit: iStock/Artush

Researchers have developed a cost-effective way to increase the number of HIV-positive adolescents who are aware of their status in places where prevalence of the virus is high but testing is under-resourced.

The screening tool, based on four simple questions, has been shown to effectively identify which adolescents are likely to be HIV-positive and should be prioritised for testing.

Many high prevalence countries are currently off-track to meet UNAIDS’ target to diagnose 90% of people living with HIV by 2020. Because facility-based testing relies on someone actually going to a health facility, community-based testing is increasingly being used to reach more HIV-positive people who are unaware of their status. This is a particularly important strategy for adolescents (aged 10-19 years) where undiagnosed HIV is a significant issue. 

While universal testing (when testing is offered to everyone in a given setting) remains the gold standard, many resource-poor settings are struggling to offer this. To respond to this context, researchers previously developed an HIV screening tool to identify which children and adolescents (aged 6–16) attending health clinics in Harare were at risk of being HIV-positive, who were then offered a test. By asking children/adolescents about four topics – previous hospitalisation, orphanhood, poor health status, and recurring skin problems – researchers found the number of children/adolescents needed to test to identify one HIV-positive individual was halved.

To test how effective the same tool was in a community setting, researchers embedded the four question-topics in an HIV prevalence survey, which they carried out on around 5,400 children and adolescents (8-17 years) from randomly selected households in Harare. The median age of participants was 12 years and 46.7% were male.

Participants’ HIV status was then tested using an anonymous, rapid oral HIV test. HIV prevalence was found to be 1.3%. 

Using the screening tool, researchers found they needed to test 34 children/adolescents to diagnose one child/adolescent living with HIV. This was 55% less than universal testing which needed 76 tests to identify one HIV positive child/adolescent. 

This is a lower success rate than when the tool was used in a clinical setting. However, this is because skin problems and poor health are likely to be more prevalent among children/adolescents in the advanced stages of HIV who, in turn, are more likely to attend health facilities.

Importantly, the findings suggest that one child/adolescent would also be falsely classified as ‘not at risk’ for every 176 children screened. This emphasises the importance of offering HIV testing universally if resources are available. 

The results can be generalised to other African settings with similarly high HIV prevalence, although any future use of the screening tool will need to be tailored to the context in which it operates. 

It is important to note that the screening tool was developed with children/adolescents who had become HIV-positive through parental transmission in mind. When pre-screening adolescents who may be HIV-positive due to sexual or injecting transmission, screening questions are likely to be more effective if they include topics relating to schooling and sexual health. 

Crucially, to be effective, the testing strategy will need to be offered in combination with services that link any children and adolescents who test positive to age-appropriate HIV treatment and care.

Get our news and blogs by email

Keep up-to-date with all our latest news stories and blogs by signing up to the Be in the KNOW news digest.

Share this page

Did you find this page useful?
See what data we collect and why