Background: Verbal autopsy (VA) is currently the only option for obtaining cause of death information in most populations with a widespread HIV/AIDS epidemic.
Methods: Using a data driven algorithm, we trained a set of criteria for classifying AIDS mortality. We use data from two longitudinal community studies in Tanzania and Zimbabwe, both of which have collected information on the HIV status of the population over a prolonged period of time and maintain a demographic surveillance system that collects information on cause of death through verbal autopsy. We then test the algorithm in different times (two rounds of the Zimbabwe study) and a different places (Tanzania and Zimbabwe).
Results: The trained algorithm, including none signs and symptoms, performed consistently based on sensitivity and specificity on 15 to 44 year old VA data from Zimbabwe Phase 1 (sen 79%; spec 79%), Phase 2 (sen 83%; spec 75%) and Tanzania (sen 75%; spec 74%). The sensitivity dropped markedly for classifying deaths in 45 to 59 year olds.
Conclusions: Verbal autopsy can consistently measure AIDS mortality with a set of nine criteria. Surveillance should focus on deaths that occur in the 15 to 44 year age group where the method performs reliably. Addition of a handful of questions related to opportunistic infections would enable other widely used verbal autopsy tools to apply this validated methodology in areas where HIV testing and hospital records are unavailable or incomplete.