Understanding the magnitude and spread of HIV/AIDS epidemic in sub-Saharan Africa: evidence from the demographic and health surveys and AIDS indicator surveys

Vinod Mishra, Macro International Inc.
Anne R Cross, Macro International Inc.
Bernard Barrere, Macro International Inc.
Rathavuth Hong, Macro International Inc.
Martin T Vaessen, Macro International Inc.

We describe how HIV seroprevalence data collected in national population-based surveys, such as the Demographic and Health Surveys and AIDS Indicators Surveys, have helped improve understanding of size and spread of the epidemic in sub-Saharan Africa. We describe the methods used to collect nationally-representative data on HIV seroprevalence, compare survey-based estimates with sentinel surveillance-based estimates, and evaluate survey-based estimates for potential bias due to non-response and due to exclusion of non-household populations. We find that in most countries survey-based estimates of HIV prevalence are much lower than sentinel surveillance-based estimates. Analysis of non-response bias indicates that although non-tested males and females tend to have slightly higher predicted HIV prevalence than those tested, overall effects of non-response on the observed national HIV prevalence estimates are insignificant. Our analysis also shows that exclusion of non-household population groups in the surveys is likely to have a minimal effect on the observed HIV estimates.

  See paper

Presented in Session 47: National HIV prevalence estimates: advantages and limitations of different estimation methods