AIDS mortality

Spectrum estimates of AIDS-related deaths are used as the source for this indicator for global monitoring. The number of people dying from AIDS-related causes can be obtained from a variety of sources, including vital registrations, ideally with adjustment for underreporting and misclassification of causes of death; as part of a facility- or population-based survey, including verbal autopsy; or through a mathematical epidemic model such as Spectrum. Epidemic models use demographic data, numbers of people receiving antiretroviral therapy over time, and assumptions around clinical progression and survival patterns to estimate the number of people living with HIV and dying from AIDS-related causes. Depending on the epidemic setting and surveillance system in the country, the Spectrum model can fit the epidemic historic trend to either prevalence data (from serosurveys, sentinel surveillance and/or routine programmatic HIV testing) or—in countries with strong HIV case-based surveillance and vital registration systems—HIV diagnoses and reported AIDS-related deaths. In the latter case, UNAIDS recommends adjusting the input number of AIDS-related deaths for underreporting and misclassification of causes of deaths for years applicable (e.g. early years) before fitting the Spectrum CSAVR incidence option to them. Spectrum-estimated deaths but not raw death counts are accepted as national AIDS-related death data.