Verbal autopsy in the REVIVE trial
Understanding the causes of death among people with advanced HIV disease (AHD) may help inform strategies to improve care and reduce preventable mortality.
In the REVIVE trial, we are using verbal autopsy to better understand the causes of death among participants, particularly where medical information is limited or where a death occurs outside of hospital.
By collecting this information in a structured way, we aim to build a clearer picture of causes of death in AHD.
What is a verbal autopsy?
A verbal autopsy is a structured interview conducted with someone who knew the participant during their final illness, such as a family member, caregiver or another close contact. The interview gathers information on the symptoms, events and care received before death. This information is then reviewed to establish a probable cause of death.
Overview of the verbal autopsy process in the REVIVE trial
How are these deaths adjudicated?
All available information on the participant, including medical records and verbal autopsy data, is reviewed by a trained site investigator, who adjudicates the death. All deaths then undergo central adjudication by an independent adjudication committee to ensure standardisation of cause-of-death ascertainment.
Why is this important in REVIVE?
In many settings in sub-Saharan Africa, cause-of-death information is incomplete or unavailable. This is particularly true when deaths occur outside of health facilities, when medical records are limited or when routine death certification does not fully capture the immediate cause of death.
Verbal autopsy helps us to improve documentation of deaths occurring during the trial. It also helps identify likely causes of death in AHD, understand differing causes of death across countries and assess whether causes of death differ between the azithromycin and placebo arms.
This helps inform future targeted interventions to reduce mortality in this population.
Current progress
Last updated date
2026-08-19
| Country | Sites activated, n | Deaths eligible for cause-of-death ascertainment, n | VA completed or medical records uploaded, n (%)* | VAs or medical records outstanding, n (%) |
|---|---|---|---|---|
| South Africa | 16 | 39 | 31(79%) | 8(21%) |
| Nigeria | 11 | 85 | 73(86%) | 12(14%) |
| Sierra Leone | 5 | 169 | 117(68%) | 52(31%) |
| Cote d’Ivoire | 2 | 7 | 6(86%) | 1(14%) |
| Rwanda | 1 | 4 | 2(50%) | 2(50%) |
| Uganda | 4 | 83 | 3(4%) | 80(96%) |
| Malawi | 1 | 8 | 6(75%) | 2(25%) |
| Ghana | 3 | 44 | 39(89%) | 5(11%) |
| Ethiopia | 4 | 5 | 5(100%) | 0(0%) |
| Zambia | 3 | 0 | - | - |
| Total | 50 | 444 | 282(64%) | 162(36%) |
