Using verbal autopsy to assess the prevalence of HIV infection among deaths in the ART period in rural Uganda: a prospective cohort study, 2006-2008
2011

Assessing HIV Deaths in Rural Uganda Using Verbal Autopsy

Sample size: 333 publication 10 minutes Evidence: high

Author Information

Author(s): Mayanja Billy N, Baisley Kathy, Nalweyiso Norah, Kibengo Freddie M, Mugisha Joseph O, Van der Paal Lieve, Maher Dermot, Kaleebu Pontiano

Primary Institution: MRC/UVRI Uganda Research Unit on AIDS

Hypothesis

Verbal autopsy can effectively identify the prevalence of HIV infection among adult deaths in rural Uganda before and after the introduction of antiretroviral therapy (ART).

Conclusion

Verbal autopsy is a reliable method for estimating the prevalence of HIV infection among adult deaths in resource-limited settings, showing a decline in HIV-positive deaths after ART introduction.

Supporting Evidence

  • 79.3% of deaths were assessed by verbal autopsy.
  • The HIV-attributable mortality fraction decreased from 47.0% to 25.8% after ART introduction.
  • Verbal autopsy had a specificity of 90.2% and a positive predictive value of 70.6%.

Takeaway

This study shows that asking family members about how someone died can help us understand how many people died from HIV, especially after new treatments became available.

Methodology

A prospective population-based cohort study was conducted where verbal autopsy was used to assess adult deaths and determine HIV serostatus through annual serosurveys.

Potential Biases

The assessment of HIV status may have been influenced by the opinions of relatives, potentially affecting the accuracy of verbal autopsy results.

Limitations

Some deaths could not be assessed due to lack of relatives to interview or refusal to participate, and not all participants had recent HIV test results.

Participant Demographics

The study included adults aged 13 years and older from a general population cohort in rural Uganda.

Statistical Information

P-Value

0.001

Confidence Interval

95% CI: 9.8, 12.1

Statistical Significance

p<0.001

Digital Object Identifier (DOI)

10.1186/1478-7954-9-36

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication