How Often Are HIV Rapid Tests Wrong? A Mozambique District Found Almost Never

A false HIV diagnosis can upend a life: separation from a spouse and family, years of unnecessary treatment, and public distrust of testing. A review of 30 studies in sub-Saharan Africa had suggested a median false-positive rate for rapid HIV testing of 3.1 percent — worrying for a region where rapid tests had enabled treatment for about 15.3 million people with HIV by 2017. CDC and Mozambican researchers checked in one district. The rate there was less than a fifth of that, and it fell to zero.

Where and how

The Chókwè Health and Demographic Surveillance System follows about 100,000 residents of Chókwè District in Gaza Province, southern Mozambique, where an estimated 25.6 percent of people aged 15–59 had HIV in 2017. From 2014 to 2017, staff visited every household in four rounds, offering a survey and HIV testing to residents aged 15–59.

• Of 57,655 residents, 43,496 (75.4 percent) took part at least once; 8,608 reported a previous HIV diagnosis, and 5,568 of those were retested with the national rapid test algorithm.
• More than 99 percent retested positive. Samples from those who tested negative or indeterminate were checked at CDC with laboratory tests, and people found HIV-negative were contacted again.

What they found

Across all rounds, 31 participants were classified as falsely diagnosed — a cumulative rate of 0.56 percent. Applied to the district's roughly 100,421 residents aged 15–64, that suggests about 110 people might ever have received a false diagnosis.

Verification matters

Of 39 people confirmed HIV-negative and contacted again, 12 (31 percent) said they had never actually been diagnosed — the initial report came from interviewer error, misunderstanding, a belief that reporting a diagnosis would get them services, or mental illness. Studies that do not verify self-reported diagnoses, the authors note, may greatly overstate false diagnosis; higher rates elsewhere may also reflect weaker testing strategies, such as using a third rapid test as a tiebreaker to rule infection in.

Those who were misdiagnosed were mostly relieved. All were taken off HIV care, including 16 who had been on antiretroviral therapy, and no loss of trust in testing was reported in the district.

Still worth retesting

The decline may reflect better testing practice and quality management, or more people choosing to retest; routine retesting before treatment, recommended by the World Health Organization, was not yet standard in 2014–2017. The authors call for retesting everyone before starting antiretroviral therapy and for comprehensive quality management of rapid testing — training, supervision, proficiency testing and outside quality checks. Areas with lower HIV prevalence may see higher false-positive rates, since a test's positive predictive value depends partly on how common the disease is.

Sources

Based on "Low and Decreasing Prevalence and Rate of False Positive HIV Diagnosis — Chókwè District, Mozambique, 2014–2017," by Daniel Shodell and colleagues (CDC, Mozambique's National Institute of Health and Ministry of Health, and partners), Morbidity and Mortality Weekly Report, Centers for Disease Control and Prevention; a work of the United States government in the public domain.