Chapel Hill, NC — A new comprehensive scoping review led by researchers at the UNC Institute for Global Health and Infectious Diseases finds definitions of “persistence” in HIV pre‑exposure prophylaxis (PrEP) research vary widely across studies, making it difficult to compare outcomes, evaluate interventions, and inform public health decision‑making.
If taken consistently and as prescribed, PrEP is highly effective in preventing HIV. However, staying on PrEP through periods of HIV vulnerability is critical to achieving its full prevention impact. This concept of continuing PrEP is frequently described as “persistence.” Published in the Journal of the International AIDS Society, the new review shows persistence outcomes are rarely defined consistently across the scientific literature, variation that could have substantial implications for how PrEP programs are evaluated and scaled.
“Our findings show that PrEP persistence is being measured in many different ways, often using entirely different inputs and thresholds,” said Sarah Rutstein, MD, PhD, assistant professor of medicine and lead author of the study. “Without clarity and transparency about how persistence is defined, it becomes extremely challenging to compare results across studies or to draw conclusions that can guide public health policy.”
A Global Review of More Than a Decade of Research
The research team reviewed peer‑reviewed publications from 2012 through January 2026, searching PubMed, Embase, Scopus, and Global Health databases for studies examining longitudinal use of anti‑HIV agents for prevention. After screening more than 1,500 records, 147 studies met inclusion criteria and explicitly defined PrEP persistence.
Among these studies, nearly one‑third provided only qualitative definitions of PrEP persistence (29%), using general terms such as “continued PrEP use” without specifying measurable criteria. Of the remaining studies that used operational definitions (71%), most relied on indicators of engagement in PrEP services, with approximately three‑quarters using prescription refill dates and/or clinic visit attendance to assess persistence (76%), with more than half relying exclusively on these measures (57%). Adherence, such as self‑reported medication use, pill counts, or drug metabolite testing, was incorporated less frequently, appearing in just over one‑quarter of studies with operational definitions (28%). A smaller subset relied solely on adherence to define persistence (11%).
The review recognizes the growing complexity of defining persistence as new PrEP modalities become available, including long‑acting injectable PrEP and vaginal rings. Having been approved since 2012, most research looks at daily oral PrEP, but for long‑acting PrEP given in clinics, simply showing up for care may be a more meaningful measure than tracking how clinic visits for persons receiving daily oral PrEP.
“As new PrEP options like long‑acting injectables become more widely available, the way persistence is measured needs to evolve,” explained William Miller, MD, PhD, senior author and professor of epidemiology. “Clear reporting will be essential to understanding how these tools perform in real‑world settings, with careful consideration of how persistence definitions align with clinical relevance and HIV prevention effectiveness.”
Looking ahead, the need for greater clarity and transparency in how PrEP persistence is defined and reported, particularly as PrEP options and modalities continue to expand, is needed. For example, studies relying solely on clinic visits or prescription refills may overestimate effective protection if individuals are engaged in care but not consistently taking medication, while adherence‑based definitions using biomarkers may be impractical for large‑scale program evaluations. Differences in thresholds—such as allowing gaps of 14 days versus 90 days in pill coverage—can also lead to dramatically different conclusions about whether a program is successful.
Researchers recommend future studies clearly explain how they define persistence, spell out the measures and cutoffs they use, and, when possible, compare results using different definitions. They also stress the importance of recognizing situations where people appropriately stop PrEP because their HIV risk is low, so persistence measures better reflect real‑world care and HIV prevention impact.
Authors also include Grace E. Mulholland, Laura Limarzi-Klyn, Annabelle Gallinek and Nicole Brown.