HIV Burden Continued to Ease, but the Gains Remain Fragile
Global HIV metrics improved across the long arc from 2003 to 2023, yet the scale of the epidemic still demands caution. Annual new infections fell from 2.85 million to 2.06 million, and HIV-related deaths dropped from a peak of 1.71 million in 2004 to 0.83 million in 2023. At the same time, the number of people living with HIV climbed from 26.3 million to 42.4 million, a pattern that reflects a simple clinical truth: antiretroviral therapy is keeping more people alive, but survival also expands the size of the population needing lifelong treatment, monitoring, and adherence support.
This is progress without closure.
The analysis, built on the Global Burden of Diseases, Injuries, and Risk Factors Study 2023 framework, estimated annual HIV incidence, prevalence, and mortality for 204 countries and territories from 1990 through 2023. Data inputs came from national HIV programme reports, population-based serosurveys, clinical and vital registration data, and systematically reviewed literature. Locations were grouped by the completeness of prevalence and mortality data, with tailored modelling approaches applied across settings. The result is a global picture that is both broad and specific enough to reveal where gains have held and where they have not.
Women Remain More Affected by Prevalence, Even as the Incidence Gap Narrows
In 2023, females accounted for 23.4 million people living with HIV, compared with 19.1 million males. The prevalence burden therefore remained higher among women, even though the incidence gap between females and males narrowed substantially over time. The biological and social explanation is not difficult to trace. Sexual exposure patterns, power imbalances in relationships, barriers to timely testing, and delayed access to treatment all shape who acquires infection and who survives with it.
The burden is not evenly distributed within or across countries. Sub-Saharan Africa accounted for the largest declines in both incidence and mortality between 2003 and 2023, with incidence falling 64.3% and mortality falling 74.6%. Those reductions align with the expansion of treatment coverage and the broader consolidation of HIV services across the region. In stark contrast, central Europe, eastern Europe, and central Asia recorded the highest increases in incidence rates, rising 232.1%, alongside a 37.8% increase in mortality rates over the same period. The divergence points to health-system gaps, uneven access to prevention and treatment, and the different trajectory of HIV epidemics outside the regions that saw the greatest ART scale-up.
Intimate Partner Violence Carried a Measurable Share of HIV Mortality in Women
One of the most clinically consequential findings was the estimated contribution of intimate partner violence to HIV-related mortality among females aged 15 years and older. In 2023, IPV was associated with 10.7% of HIV-related deaths globally in that group, equivalent to 40,700 deaths. The highest population attributable fractions were observed in Oceania at 17.4%, central sub-Saharan Africa at 14.2%, and eastern sub-Saharan Africa at 13.2%.
IPV can intensify HIV risk through multiple pathways at once. Violence limits a woman’s ability to negotiate condom use, undermines autonomy around testing and treatment, increases exposure to trauma and coercion, and can disrupt retention in care after diagnosis. The estimate used time series data on IPV prevalence by location and effect-size estimates to generate population attributable fractions, translating a social exposure into a quantifiable share of disease burden. In practical terms, HIV prevention cannot be separated from violence prevention when the same patient population is being pushed toward infection and away from sustained care.
A woman cannot take uninterrupted ART if violence makes disclosure unsafe.
Funding Reductions Could Reopen the Epidemic Curve by 2030
The forecast component carries the clearest warning. Using a non-parametric stochastic frontier approach, the analysis estimated how development assistance for health could influence ART coverage, then folded those changes into future HIV projections. Under a scenario of funding cuts, ART coverage could decline by 8.1% globally between 2025 and 2030. The downstream consequences would be severe: about 1.6 million additional new HIV infections among females and 1.3 million among males, along with 790,600 HIV-related deaths among females and 670,600 among males.
Those projected losses would fall mainly in sub-Saharan Africa, where treatment systems have relied heavily on external financing and where any erosion in service delivery would be felt quickly. Reduced ART coverage does more than increase mortality. It raises viral replication, fuels transmission, and weakens the prevention effect that treatment has delivered across entire communities. The model therefore captures not only a clinical setback for individual patients, but a population-level shift in infectiousness that can undo years of progress.
The Epidemiology Points to Policy, Not Only to Pathogens
The HIV epidemic described here is shaped by more than virology. ART expansion has changed survival, but social instability and financing volatility continue to decide who benefits from that progress. The concentration of IPV-related mortality among women makes gender-responsive programming a clinical requirement, not an optional add-on. Prevention, post-violence support, safe disclosure pathways, and treatment access need to operate together if mortality is to fall further.
The forecasted effects of reduced funding also expose how dependent HIV control remains on stable financial support. Even a modest contraction in ART coverage can cascade into higher incidence and more deaths over a relatively short period. The projected burden through 2030 suggests that the UNAIDS targets remain vulnerable to policy reversals, especially in regions where external assistance still anchors treatment scale-up. Sustaining the gains achieved over the last two decades will require resilient health systems, protected financing, and HIV services that recognize violence as part of the disease burden rather than a separate social problem.