HIV Self-Testing: A Decade of Delay Leaves Millions Unaware of Their Status

More than a decade after the World Health Organization recommended HIV self-testing as a way to reach people conventional services were failing, a new analysis in PLOS Medicine argues that the technology’s promise has been squandered on a global scale. Writing in the journal, Reuben Granich and colleagues reconstruct what happened after 2016, when WHO gave its formal endorsement to a simple idea: let people test themselves for HIV in private, with an oral fluid swab or finger-prick blood test, and then connect to care if the result is reactive. The verdict is stark. Self-testing was deployed at negligible scale for roughly ten years, and the diagnostic gap it was designed to close has narrowed mainly through conventional facility-based and outreach testing instead.

The scale of the original problem is worth recalling. In 2015, a survey-based analysis across 16 countries in sub-Saharan Africa estimated that nearly half of people living with HIV in those countries did not know their status. Extrapolated across the region, that implied roughly 11 million people standing entirely outside the treatment cascade, unaware they needed antiretroviral therapy and capable of onward transmission at rates far higher than people whose infection was suppressed by medication. Self-testing, the authors contend, was precisely the technology positioned to reach this population, because it removed many of the barriers that keep people away from clinics: stigma, distance, waiting times, fear of being seen entering an HIV service, and the loss of a day’s income.

The epidemiological logic behind self-testing is straightforward. Antiretroviral therapy dramatically reduces infectiousness, and people who know their status can start treatment and achieve viral suppression. Every undiagnosed infection therefore represents both a personal health risk and a transmission risk to others. The UNAIDS 95-95-95 framework, which targets 95 percent of people living with HIV knowing their status, 95 percent of those diagnosed on treatment, and 95 percent of those on treatment virally suppressed, depends on the first 95 more than any other, because no one can be treated if they have not been found. Yet in 2025, an estimated 2.6 million people in sub-Saharan Africa and 4.9 million worldwide still did not know their HIV status.

The authors quantify the shortfall in procurement terms. The need for HIV self-testing in low- and middle-income countries was estimated at 177 million self-tests in 2020, rising to 192 million by 2025. Confirmed procurement over that same period reached only 21 million kits in total, an average of 5.3 million per year. That is roughly 3 percent of the estimated annual need. In other words, for every hundred self-tests that modeling suggested the region required, fewer than three were purchased. The gap between need and supply was not marginal; it was nearly total.

Why did a cheap, simple, WHO-endorsed technology fail to reach the market at scale? The analysis identifies four interlocking structural failures. The first is what the authors call research capture. Instead of population-scale deployment, the field invested heavily in demonstration projects, small studies designed to show that self-testing could work. Such studies accumulated evidence of feasibility and acceptability year after year, but they substituted for, rather than preceded, real distribution. A technology that had already been endorsed by WHO did not need another pilot; it needed procurement, supply chains, and delivery systems. The demonstration-project cycle, the authors argue, consumed a decade that should have been spent scaling.

The second failure is a paradox embedded in early implementation. Many programs introduced self-testing in supervised or provider-assisted forms, in which a health worker remained present while the client performed the test. This arrangement, intended to reassure regulators and funders, effectively reconstituted the very clinic barriers self-testing was designed to bypass. If a person must still travel to a facility, wait for a provider, and test under observation, the private, convenient, stigma-free experience that defines self-testing disappears. Supervised delivery, the authors suggest, blunted the technology’s core advantage during the years when scale was most urgently needed.

The third failure concerns procurement conservatism. Donors and national programs were slow to commit volume, and without predictable demand, manufacturers had little incentive to invest in production capacity or price reduction. Procurement behaves like a market signal: guaranteed volumes attract suppliers, drive competition, and lower unit costs, while hesitant, fragmented purchasing keeps prices high and supply chains thin. By buying cautiously, the global health system suppressed the very market development that would have made self-testing cheap and abundant. The result was a self-reinforcing loop of low demand, high prices, and continued hesitation.

The fourth failure is one of accountability. The authors found no publicly accessible, Africa-level series of HIV self-testing commodity volumes, which means the scale of the missed opportunity cannot even be measured from published data. Without transparent reporting of how many kits are bought and distributed, no one can hold programs, donors, or manufacturers to account, and no one can detect shortfalls in time to correct them. The absence of a simple volume series, the authors argue, is itself a policy failure, because what is not measured cannot be managed, and what cannot be measured cannot be missed publicly.

The consequences of the delay, while not formally modeled in the analysis, are likely to be severe. Each year that millions of people remained undiagnosed translated into untreated infection, avoidable illness and death, and onward transmission. The authors characterize the burden as potentially millions of illnesses, deaths, and transmissions that could have been avoided had self-testing been deployed at the scale the need estimates demanded. They are careful to note that this counterfactual has not been quantified with formal modeling, but the arithmetic of the treatment cascade makes the direction of the effect unambiguous: people outside the cascade are sicker and more infectious than people within it.

The paper closes with four concrete prescriptions. First, mandatory public reporting of HIV self-testing commodity volumes, so that procurement and distribution become visible and auditable. Second, a dedicated Africa-specific distribution target of at least 50 million kits per year, embedded within the UNAIDS 95-95-95 framework, to convert the first 95 from an aspiration into a supply commitment. Third, community-based distribution as the default delivery model for the general population, moving self-tests out of clinics and into the places where undiagnosed people actually live, work, and socialize. Fourth, explicit self-testing volume commitments written into the 2026 to 2030 memoranda of understanding and work plans of the Global Health Sector Strategies on HIV, so that the next decade begins with binding obligations rather than another round of pilots. Whether the global response absorbs these lessons, the authors suggest, will determine whether the remaining millions finally learn their status, or whether the diagnostic gap closes as slowly and as late as it opened.

Subject of Research: HIV self-testing scale-up and the global undiagnosed HIV gap in sub-Saharan Africa

Article Title: HIV self-testing and the global diagnostic gap: Addressing a decade of missed opportunity

Article References: Granich, R., Gupta, S., Munyangaju, I., & Ruffner, M. (2026). HIV self-testing and the global diagnostic gap: Addressing a decade of missed opportunity. PLOS Medicine, 23(9), e1005241. https://doi.org/10.1371/journal.pmed.1005241

Image Credits: AI Generated

DOI: 10.1371/journal.pmed.1005241

Keywords: HIV self-testing, HIV diagnosis, sub-Saharan Africa, UNAIDS 95-95-95, procurement, global health, PLOS Medicine, undiagnosed HIV, community-based distribution, World Health Organization, antiretroviral therapy, diagnostic gap