Mapping HIV prevalence in Nigeria using small area estimates to develop a targeted HIV intervention strategy

Area of Science:

• Epidemiology
• Geographic Information Systems (GIS)
• Public Health

Background:

• Nigeria's HIV/AIDS policies rely on national and state-level data, potentially missing crucial sub-state variations.
• Targeted HIV prevention and treatment strategies require geographically specific data for effective implementation.

Purpose of the Study:

• To calculate sub-state estimates for HIV prevalence, antiretroviral therapy (ART) coverage, and viral load suppression (VLS) in Nigeria.
• To identify geographic disparities in the HIV continuum of care at a granular level.

Main Methods:

• Geospatial analysis using data from the Nigeria HIV/AIDS Indicator and Impact Survey (NAIIS) (2018).
• Ensemble modeling (stacked generalization) and geostatistical models were employed to estimate indicators.
• Raster estimates were produced on a 5x5 km grid, with sub-state/LGA and state-level estimates.

Main Results:

• Significant variations in HIV prevalence, ART coverage, and VLS were observed within and between states at the local government area (LGA) level.
• LGA-level HIV prevalence ranged from 0.2% to 8.5%, with substantial differences in ART coverage (mean absolute difference of 16.7 percentage points between LGAs within states).
• Disparities in ART coverage correlated with VLS differences, indicating state-level targeting may be insufficient.

Conclusions:

• Geospatial analysis effectively highlights sub-state variations in the HIV continuum of care.
• Local-level estimates are essential for identifying areas needing targeted interventions and optimizing resource allocation for HIV epidemic control.
• This granular data empowers policymakers and implementers to refine strategies for greater impact.