Authors classified Sub-Saharan African recipient countries into three discrete response categories after the January 2025 suspension of United States foreign aid that included PEPFAR funding. Level 1 described minimal or no governmental response. Level 2 described government action that lacked sufficient plans or resources to fully offset funding losses. Level 3 described high governmental response with strategies intended to maintain HIV prevention and treatment services. The cohort in the manuscript comprised 5 countries (17.2%) at Level 1, 10 (34.5%) at Level 2, and 14 (48.3%) at Level 3, numbers provided directly in the source. These counts framed the comparative analysis of national factors and response strength.
The study team reported a clear positive association between a country’s Human Development Index and its classified response level. In univariable proportional-odds ordinal logistic regression, higher HDI was associated with stronger governmental response (OR = 4.05; 95% CI, 1.49–11.04; Holm-adjusted p = 0.0376). That association remained statistically significant after correction for multiple comparisons, according to the authors. The analysis therefore linked broader measures of national development with an increased likelihood of organized, resourced government activity to address the funding shock.
Investigators also evaluated GDP per capita and national HIV prevalence against response level. Both factors showed positive associations in unadjusted models: GDP per capita had an OR of 2.35 (95% CI, 1.00–5.51; unadjusted p = 0.0489) and HIV prevalence had an OR of 3.53 (95% CI, 1.03–12.11; unadjusted p = 0.0452). Those associations did not survive Holm correction for multiple comparisons (adjusted p = 0.2261 for both), the authors noted. The manuscript therefore reports these as signals that merit further study rather than definitive, corrected associations.
The team examined whether prior PEPFAR funding levels, the total population living with HIV, or annual HIV mortality predicted government response category. After adjustment for multiple comparisons, none of those factors reached statistical significance in the reported univariable models. The source explicitly states that historical PEPFAR funding, total people living with HIV, and annual HIV mortality were not significantly associated with response level following correction, and the authors present that null finding in the results.
The authors reported that several countries with minimal documented governmental responses were affected by active conflict or political instability. They observed that such contextual factors may compound vulnerability to HIV treatment interruptions following external funding shocks. The paper does not provide a formal multivariable analysis that isolates conflict from other national indicators; the observation about instability is presented alongside the primary, country-level statistical findings.
Data for country-level variables and final governmental response classifications are provided in Table 1 of the manuscript, the authors stated. Supporting files described in the paper include S1 Appendix (individual reviewer assessments, consensus classifications, and supporting evidence) and S2 Appendix (complete regression results and sensitivity analyses). Publicly available sources used by the investigators included UNAIDS, the United Nations Development Programme Human Development Reports data center, the International Monetary Fund World Economic Outlook datamapper, PEPFAR program information, and World Bank indicators. The primary statistical approach reported was univariable proportional-odds ordinal logistic regression; details beyond that (for example, model covariates in multivariable frameworks) were not reported in the abstract, and readers must consult the supporting appendices in the manuscript for full methodological specifics.
The authors note several constraints inherent in their analysis. The preprint reports univariable regression results and applies Holm correction for multiple comparisons; it therefore does not claim causal effects. The study did not seek formal IRB oversight because it used public, country-level data and did not involve human-subjects research. The manuscript declares no competing interests. The authors emphasize that broader national development appears to correlate with stronger governmental responses to the 2025 funding disruption, and they identify the need to strengthen resilient health systems, diversify HIV financing, and prioritize support to countries at greatest risk of future funding shocks. Specific unanswered questions documented by the source include which programmatic elements of higher-HDI countries enabled sustained services, how active conflict interacts quantitatively with development metrics to influence response, and whether multivariable models adjusting for confounders would change the associations observed. Those items remain open and are set out in the paper’s supporting materials for further scrutiny by readers and reviewers.