Structural determinants of outcomes related to the prevention of vertical HIV transmission: A cross‐country ecological analysis

OBJECTIVE: This study examined the longitudinal associations of epidemiological, health system and structural factors with paediatric HIV infections and antiretroviral therapy (ART) need among pregnant women in 30 high-burden countries. METHODS: This longitudinal cross-country ecological study used secondary data from international databases covering 2000-2024. A total of 30 countries were included based on HIV burden and data availability. Data were sourced from UNAIDS (indicators related to the prevention of vertical HIV transmission), WHO/UNICEF (maternal health services) and the World Bank (socioeconomic variables). Missing data were handled using linear interpolation and last observation carried forward (LOCF) to create a balanced panel. Datasets were harmonized by country and year. Linear mixed-effects models were used to account for repeated observations within countries, with year included as a fixed effect and country specified as a random-intercept grouping factor. RESULTS: Adult HIV prevalence showed substantial variability across countries and years. In the linear mixed-effects model, higher adult HIV prevalence was significantly associated with higher levels of log paediatric HIV infections (B = 0.129, p < 0.001) and log ART need among pregnant women (B = 0.086, p < 0.001). Higher ANC coverage was significantly associated with lower levels of both log paediatric HIV infections (B = -0.008, p < 0.001) and log ART need (B = -0.006, p < 0.001). GDP per capita was not significantly associated with log paediatric HIV infections (B = -0.00003, p = 0.127), whereas it was positively associated with log ART need (B = 0.000033, p = 0.004). Paediatric HIV infections decreased significantly over time (B = -0.059, p < 0.001), while year was not significantly associated with ART need (B = -0.00088, p = 0.626). Strong country-level clustering was observed for both outcomes (ICC = 0.921 and 0.975, respectively). CONCLUSIONS: Higher adult HIV prevalence was associated with higher paediatric HIV infections and ART need, whereas higher ANC coverage was associated with lower levels of both outcomes. GDP showed outcome-specific associations, and substantial between-country variability was observed. These findings highlight the importance of epidemic burden and maternal health service coverage in understanding outcomes related to the prevention of vertical HIV transmission.