How to Read the Latest PEPFAR Data

What the topline numbers show—and what they miss

The latest PEPFAR data is starting to give us a clearer picture of what actually happened during the 2025 foreign aid disruption. At first glance, it doesn’t look catastrophic. As the State Department noted in its official release on April 17, 2026, the U.S. government still supported HIV treatment for 20.6 million people globally by the end of FY25. It also pointed to progress on country ownership and highlighted a sharp increase in PrEP use among pregnant and breastfeeding women.

It is important to note that the State Department release reflects fourth-quarter results, which offer a snapshot rather than a full-year view. Broader analyses of 2024–2025 data show sharper declines across testing and prevention than what is visible in Q4 alone.

Overall, two things seem to be true. First, these are impressive achievements and should be acknowledged. Second—and just as important—they do not fully capture what is happening within the system.

When you look underneath the topline numbers, the trends shift quickly. HIV testing declined substantially over the year. Diagnoses fell across settings, with the steepest drops in community-based programs.

PEPFAR-supported testing in Q4 2025 dropped to 17.2 million people, compared to 21.9 million a year earlier. Diagnoses followed the same trend, declining to 307,000 from 385,000—consistent with lower testing levels.

PrEP initiations also declined overall, even as targeted increases were observed among pregnant and breastfeeding women. At the same time, the workforce supporting service delivery contracted. These patterns emerge from full-year comparisons and facility-level analyses, which provide a more complete picture than quarterly snapshots alone, particularly in a year marked by reporting disruptions.

What is emerging is not a system that has failed, but one that is beginning to thin at the margins. In the HIV response, those margins are not negligible—they are where people are first tested, where they enter care, and where prevention reaches populations at highest risk of HIV acquisition. Changes in these areas often precede shifts in more visible indicators.

PEPFAR has been tested in similar ways before. Across past crises, it has functioned less like a narrow disease program and more like a broader health platform capable of absorbing shocks. During COVID-19, treatment continuity largely held, while testing and case finding declined. A similar pattern was observed during the Ebola outbreak, where treatment and PMTCT services proved more durable than upstream functions. These experiences suggest that the system has an inherent capacity to preserve core services when under stress.

This pattern also helps explain what we are seeing now. Following the PEPFAR funding freeze—and the subsequent waiver limited to a narrow set of lifesaving services—the program has effectively shifted into preservation mode. Treatment has been prioritized and largely protected, but many of the activities that identify and bring new people into care have been scaled back or outright cancelled. That includes programs like DREAMS, prevention for key populations, voluntary medical male circumcision, and much of the community outreach that drives testing and linkage. As a result, the system is maintaining continuity for those already on treatment, while the front end of the response is weakening. Over time, this creates a divergence between what treatment numbers suggest and what is actually happening across the broader HIV response. This is concerning because it risks reversing progress on case finding and prevention, ultimately increasing the likelihood of new infections and undermining long-term epidemic control.

One reason treatment appears so stable—even in the face of disruption—is that PEPFAR has deliberately built resilience into how treatment is delivered. Over the past several years, especially during COVID-19, programs scaled up multi-month dispensing, meaning many patients now receive two, three, or even up to six months of antiretroviral medication at a time. That shift reduced the need for frequent clinic visits and created a buffer against short-term disruptions. When funding pauses or service delivery is interrupted, people do not immediately fall out of care because they often still have medication in hand. At the same time, many patients have established relationships with providers, community health workers, or clinics, which allows them to stay connected even when systems are under strain. These features of the system are not incidental—they reflect years of programmatic evolution grounded in evidence and experience. But they also explain why treatment can appear stable even as other parts of the system begin to weaken.

The challenge is that HIV control programs do not run on treatment alone. They depend on a continuous flow of people entering care through testing, diagnosis, and linkage services. When testing declines and fewer people are diagnosed, that pipeline slows. The data suggest that this is already happening. Declines in testing and prevention are not immediately visible in treatment numbers, but they shape what those numbers will look like in the future. This is why treatment is a lagging indicator. By the time changes show up there, the underlying dynamics have already shifted.

A related issue is what might be described as the scaffolding of the response. Over time, PEPFAR investments have built systems that extend well beyond clinical care—community outreach, peer navigation, and civil society partnerships. These systems are particularly important for reaching populations that are not well served by formal health facilities, because they are often the entry point into care. The current data suggest that these parts of the system are where the greatest strain is occurring. Declines are most pronounced in community-based services and in sites with less consistent support, which raises concerns about how effectively the system is continuing to reach those populations—often the very groups at highest risk of HIV acquisition.

This becomes especially important when thinking about prevention. Reductions in testing and PrEP are not just short-term service gaps; they have implications for future transmission dynamics. When fewer people are tested, fewer diagnoses are made. When fewer people initiate PrEP, fewer infections are prevented. The State Department notes that more than 103,000 pregnant and breastfeeding women were newly enrolled on PrEP, which is an important achievement. But beyond that, there is limited visibility into other populations being reached or prioritized.

Historically, PEPFAR has shown that prevention programs are most effective when resources are explicitly targeted and sustained for specific populations at highest risk. Without that focus, coverage becomes diffuse and impact declines. These effects accumulate over time and are often only visible later, when incidence begins to change. In other words, cutting prevention is a ticking time bomb—the consequences are delayed, but they will come.

Looking ahead

The State Department’s release emphasizes improvements in Q4 data completeness and signals a transition toward reporting that is more closely aligned with national systems and bilateral agreements. While this shift is consistent with a broader emphasis on country ownership, it also introduces important tradeoffs. Historically, PEPFAR has provided detailed, continuous, publicly accessible data that allowed for granular analysis across countries and programs. That level of visibility has been central to how the program has been managed and held accountable.

As reporting evolves, it is likely that both the frequency and the granularity of available data will change. Greater reliance on national reporting systems may mean less standardized data and, in some cases, reduced public access. While this transition may improve alignment with country systems, it also introduces variability in how indicators are defined, collected, and shared, complicating cross-country comparisons and independent analysis.

Looking ahead, this shift in how data are generated and shared will require a corresponding shift in how oversight is conducted. As the U.S. moves toward a more integrated, bilateral framework, expectations around transparency, data sharing, and accountability will need to be clearly defined within those arrangements. Otherwise, there is a risk that emerging challenges—particularly in areas like prevention and community engagement—will be less visible until they are more difficult to address. More fundamentally, this transition signals a different kind of HIV response—one I have written about previously—where the tools we have relied on to track progress, identify gaps, and course-correct in real time may no longer function in the same way.

The PEPFAR Panorama Spotlight dashboard provides a high-level snapshot of program performance in the final quarter of FY25, highlighting key indicators across testing, treatment, viral suppression, and prevention. It shows the scale of services delivered—including tens of millions of people tested and on treatment, high levels of viral suppression, and targeted prevention efforts such as PrEP for pregnant and breastfeeding women. While these figures underscore the continued reach of the program, they represent a point-in-time view and should be interpreted within the broader context of year-over-year trends and shifts in program delivery.

There appears to be a discrepancy in the headline treatment figures that warrants clarification. The State Department press release reports 20.6 million people on treatment, but when I look through the publicly available dashboards, the total comes closer to 20.3 million. See screenshot below. This may reflect differences in methodology, reporting periods, or data inclusion—but without clear documentation, it raises avoidable confusion. Given the stakes, the State Department should reconcile these figures and release the underlying data and definitions to enable full, independent interpretation.

Technical note on data interpretation: Some changes in reported results may reflect both real service disruptions and shifts in what is captured within PEPFAR-supported systems. In some cases, services may continue even as they fall out of formal reporting structures, making trends more difficult to interpret.

---

References

• Harris TG, Rabkin M, El-Sadr WM. Achieving the HIV care continuum during the COVID-19 pandemic: PEPFAR program adaptations. MMWR Morb Mortal Wkly Rep. 2022;71(12):1-7. Available at: https://www.cdc.gov/mmwr/volumes/71/wr/mm7112a2.htm

• Lecher SL, Ellenberger D, Kim AA, et al. Scale-up of HIV viral load monitoring — PEPFAR-supported countries, 2019–2020. MMWR Morb Mortal Wkly Rep. 2021;70(21):794-800. Available at: https://www.cdc.gov/mmwr/volumes/70/wr/mm7021a3.htm

• Ly J, Sathananthan V, Griffith B, et al. Facility-based delivery of HIV services during the Ebola virus disease outbreak in Liberia: an analysis of service disruption and recovery. PLoS Med. 2016;13(8):e1002096. doi:10.1371/journal.pmed.1002096

• Ratevosian J, Beyrer C. Fundamental PEPFAR reform risks a period of structural vulnerability in the HIV response. Lancet. 2026;407(10533):1035-1038. doi:10.1016/S0140-6736(26)00369-7