Ending the HIV Epidemic

“Ending the HIV epidemic” is now a defined and time-bound goal. The operational endpoint set by UNAIDS and adopted by every major HIV programme is the end of AIDS as a public-health threat by 2030: a 90% reduction in new infections and AIDS-related deaths against the 2010 baseline, with continued decline thereafter to sustain the gain. The world is not yet on track, but the path is no longer notional: the framework, the targets, the tools and the evidence that those tools work are all in place.

The global pandemic in numbers

Around 40 million people are living with HIV worldwide. Eastern and southern Africa carry just over half of that total, approximately 21 million people living with HIV (PLHIV), and remain the regional epicentre. Western and central Africa, Asia, the Americas and Europe make up the remainder.

The long arc since 2010 is a substantial decline in both new infections and AIDS-related mortality. In eastern and southern Africa, new infections have fallen by more than half and AIDS-related deaths by almost 60%. The combination of antiretroviral treatment scale-up, voluntary medical male circumcision, prevention of vertical transmission and biomedical prevention has produced what is, by any historical measure, an extraordinary public-health response. Several southern and eastern African countries (Botswana, Eswatini, Lesotho, Namibia, Rwanda, Zambia and Zimbabwe) have already achieved the 95-95-95 cascade target. South Africa, the country with the largest single national burden, is close on two of three.

Three caveats sit alongside the progress. First, the decline has plateaued globally: at around 1.3 million new infections per year, the world remains well above the UNAIDS interim target of 370,000 by 2025. Second, the picture is regionally heterogeneous: incidence is rising in the Middle East and North Africa, in Latin America and in eastern Europe and central Asia, even as it falls across eastern and southern Africa. Third, many populations (adolescent girls and young women, men who have sex with men, sex workers, people who inject drugs, transgender people, prisoners and migrants) remain disproportionately affected and underserved worldwide.

The 95-95-95 framework

The dominant operational framework of the modern HIV response is the 95-95-95 treatment cascade: 95% of people living with HIV diagnosed, 95% of those diagnosed on antiretroviral therapy (ART), and 95% of those on therapy virally suppressed. Achieving all three in sequence delivers population-level viral suppression of approximately 86%, the threshold above which onward transmission collapses and the epidemic begins to end as a public-health threat. The mechanism is treatment as prevention applied at scale: a person with a sustained viral load below the limit of detection does not transmit HIV sexually (the principle of U=U, undetectable equals untransmittable), and across a population a sufficiently high proportion of suppressed PLHIV drives incidence toward zero.

The framework originated as 90-90-90 in 2014 and was upgraded to 95-95-95 when the original 2020 milestone came into view and the 2030 horizon needed a more ambitious anchor. Global cascade progress sits in the mid-80s across all three pillars, with substantial regional and population variation beneath the averages. Eastern and southern Africa sits at 93-91-95 as a region, closer to the target than any other, though several countries remain well behind on the first 95 (men, youth and key populations being the structural gap).

The UNAIDS 2026 to 2031 strategic horizon

The UNAIDS 2026 to 2031 Global AIDS Strategy: Towards Ending AIDS sets out the framework for the next five years of the response. Its overarching goal is the end of AIDS as a public-health threat by 2030, defined as a 90% reduction in new infections and AIDS-related deaths against the 2010 baseline. Three top-line outcome targets sit alongside the cascade:

• 40 million PLHIV on treatment and virally suppressed by 2030.
• 20 million people accessing antiretroviral-based prevention, pre-exposure prophylaxis (PrEP) and post-exposure prophylaxis (PEP) at scale.
• All people accessing stigma- and discrimination-free HIV services.

Three strategic shifts deserve flagging.

Long-acting biomedical prevention has been elevated to a central priority. The twice-yearly subcutaneous capsid inhibitor lenacapavir showed 100% efficacy in cisgender women in southern Africa in the Purpose 1 trial and 96% efficacy in cisgender men and gender-diverse populations in Purpose 2, the most dramatic single advance in HIV prevention in a decade. Long-acting injectable cabotegravir (intramuscular every eight weeks after a loading dose) is also part of the strategic emphasis on agents that turn adherence from a daily-tablet problem into a clinic-visit problem. Broadly neutralising antibodies and continued vaccine development sit further out.

Triple elimination of vertical transmission, covering HIV, syphilis and hepatitis B, is integrated into a single antenatal and postnatal framework rather than three parallel ones. Programmatic integration with tuberculosis (TB), sexual and reproductive health, family planning and non-communicable diseases is a structural emphasis of the new Strategy.

The financing environment is under significant pressure, with external development assistance for health having fallen substantially. The Strategy is explicit about the need for sustainable, country-led financing and the integration of HIV services into universal health coverage and primary-health-care systems. The specific financing landscape will shift over the strategy period; what is durable is the recognition that the next phase of the response must be operationally lighter and structurally more resilient than the last.

The South African epidemic

South Africa carries the largest single national HIV burden in the world, an estimated 7.8 million people living with HIV, and runs one of the largest medication programmes on earth. The most recent comprehensive national survey, the Sixth South African National HIV Prevalence, Incidence, Behaviour and Communication Survey (SABSSM VI, fieldwork 2022), gives the most authoritative current picture.

Headline numbers (SABSSM VI):

• National HIV prevalence 12.7%, down from 14.0% in 2017.
• The sex differential remains marked: females 16.4% versus males 8.8%.
• Estimated annual incidence 0.44%, around 232,000 new infections per year (down from approximately 250,000 in 2017).
• Provincial range from 17.4% in Mpumalanga and 16.0% in KwaZulu-Natal at the high end to 8.9% in the Northern Cape and 7.4% in the Western Cape at the low end. The Northern Cape was the only province where prevalence rose between SABSSM V and VI.

The 95-95-95 cascade in South Africa (SABSSM VI, PLHIV aged 15 and older): 89.6-90.7-93.9. The second and third targets are effectively achieved. The persistent gap is the first 95, undiagnosed PLHIV, concentrated in men (85% diagnosed versus 92% in women) and in youth aged 15 to 24 (73%). Translated to all PLHIV, around 77% are virally suppressed, approaching but not yet at the 86% population threshold. This shapes where the South African programme places its emphasis: male-targeted services, youth engagement, and the standardised use of HIV self-screening from the second PrEP-continuation visit and from re-engagement.

The adolescent girls and young women (AGYW) success story. HIV incidence in girls and young women aged 15 to 24 fell approximately three-fold between 2017 and 2022, from 1.17% to 0.39%, roughly 19,000 new AGYW infections per year now, down from approximately 72,000. This is one of the most substantial population-level prevention successes anywhere in the global HIV response, and the reason South Africa hosted the Purpose 1 trial of twice-yearly lenacapavir PrEP, which produced 100% efficacy in the same population.

The ageing epidemic. The shape of the South African HIV burden has shifted right. Peak prevalence is now in females aged 35 to 39 (34.2%) and males aged 45 to 49 (27.1%), and prevalence in the over-50s is rising. As successful antiretroviral therapy continues to convert HIV into a chronic condition, the future of South African HIV care will be an older, multimorbid population, managing the cardiovascular, metabolic, renal and oncological consequences of long-term HIV alongside the virus itself.

The TLD transition and transmitted resistance. Approximately 73% of ART-treated adults are on a dolutegravir-containing regimen (TLD, tenofovir, lamivudine and dolutegravir); the transition is well advanced but incomplete, with provincial heterogeneity from under 40% in some areas to nearly 80% in others. At the same time, drug-resistance mutations in viraemic samples have risen from 27.4% (2017) to 36.2% (2022), predominantly NNRTI (non-nucleoside reverse transcriptase inhibitor)-related, the legacy of the long efavirenz-based first-line era. Both findings reinforce the rationale for the dolutegravir anchor and for routine pre-treatment HIV drug-resistance surveillance.

South Africa within the region. South Africa is unusual in eastern and southern Africa for the substantial domestic share of HIV financing: while many regional countries remain heavily dependent on external assistance, South Africa has a meaningful and increasing domestic-funding base. This is a strategic advantage in the current funding environment, and a reason the programme is well-placed to sustain the response even as the external-aid landscape shifts.

References and further reading

• Human Sciences Research Council. South African National HIV Prevalence, Incidence, Behaviour and Communication Survey VI (SABSSM VI): Executive Report. 2024. The primary source for current South African epidemiology.
• UNAIDS. 2026 to 2031 Global AIDS Strategy: Towards Ending AIDS. 2026. The global strategic framework, including the three new top-line targets, the elevation of long-acting prevention, and the triple-elimination integration.
• UNAIDS dashboards, THEMBISA model outputs, National Institute for Communicable Diseases (NICD) surveillance reports, and National Department of Health dashboards for current programmatic data.