Background: While human immunodeficiency virus (HIV)-related causes of death have been well documented in developed countries, in Africa data are scanty and mainly based on autopsy studies from city hospitals which are highly selective and may not represent causes of HIV-associated deaths in the general population. This study, from a rural population, describes the causes of death in HIV-positive people and their HIV-negative controls.
Methods: A natural history cohort comprising HIV-1 infected participants and HIV-negative controls was established in rural Uganda in 1990. Causes of death were determined by reviewing the premorbid clinical and laboratory findings and from information obtained from relatives. Blindness to the deceased's HIV serostatus was maintained throughout.
Results: In all, 78 deaths occurred over a 6-year period: 63 deaths occurred in the HIV-positive cases (53 prevalent and 10 incident cases) and 15 deaths in the HIV-negative controls. Of the prevalent cases, 56%, and 9% the incident cases enrolled died, compared with 7% of the HIV-negative controls. Of the 55 HIV-positive cases with sufficient data to establish cause of death, 52 (95%) were assessed as having HIV-associated deaths and 48 (87%) died in WHO stage 4 (AIDS). The main causes of death were wasting syndrome (31%), chronic diarrhoea (22%), cryptococcal meningitis (13%) and chest infection (11%).
Conclusions: Our results represent an unbiased selection of deaths in a rural area. The HIV-positive cases have high death rates and die of HIV-related pathologies. The main causes of death reflect the WHO clinical case definition of AIDS. Cryptococcal meningitis is also a common cause of death in this population.
PIP: A natural history cohort (NHC) of HIV-1-infected subjects and HIV-negative controls was established in rural Uganda in 1990. By the end of 1996, 440 participants had enrolled in the cohort: 107 prevalent cases, 108 incident cases, and 225 HIV-negative controls. The authors report the causes of death among HIV-infected cohort members over the 6-year period ending December 1996. Causes of death were determined by reviewing the premorbid clinical and laboratory findings, as well as from information obtained from relatives. All study clinic staff are blind to the HIV serostatus of participants in the NHC. 78 deaths occurred over the 6-year study period: 63 deaths among HIV-positive cases (53 prevalent and 10 incident cases) and 15 deaths among HIV-negative controls. 56% of prevalent cases, 9% of incident cases and 7% of controls died. Of the 55 HIV-positive cases with enough data to establish cause of death, 52 were determined to have HIV-associated deaths, of whom 48 died in World Health Organization stage 4 illness. Main causes of death were wasting syndrome (31%), chronic diarrhea (22%), cryptococcal meningitis (13%), and chest infection (11%).