Evaluating Bias in HIV Estimates from Household Surveys
Author Information
Author(s): Vinod Mishra, Barrere B, Hong R, Khan S
Primary Institution: Macro International Inc
Hypothesis
To evaluate HIV seroprevalence estimates from demographic and health surveys (DHS) and AIDS indicator surveys (AIS) for potential bias due to non-response and exclusion of non-household population groups.
Conclusion
Non-response and the exclusion of non-household population groups tend to have small, insignificant effects on national HIV seroprevalence estimates obtained from household surveys.
Supporting Evidence
- Non-tested men had significantly higher predicted HIV prevalence than those tested in eight of the 14 countries.
- Non-tested women had significantly higher predicted prevalence than those tested in seven of the 14 countries.
- The overall effect of non-response on observed national HIV estimates was small and insignificant in all countries.
- Estimated effects of exclusion of non-household population groups were generally small.
Takeaway
This study looked at how missing people from surveys might change the numbers we see for HIV rates. It found that not including some groups doesn't really change the overall picture much.
Methodology
Data from 14 DHS/AIS surveys with HIV testing conducted from 2003 to 2006 were analyzed using multivariate statistical models.
Potential Biases
Non-response rates were higher among urban, more-educated, and wealthier respondents.
Limitations
The estimates are only adjusted to the extent that the sociodemographic and behavioral characteristics included in the analysis are correlated with the risk of HIV infection.
Participant Demographics
Nationally representative samples of women age 15–49 and men age 15–59 were tested for HIV.
Statistical Information
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website