Rapid AIDS mortality surveillance in South Africa.
Int Conf AIDS 2002 Jul 7-12; 14:(abstract no. C10709)
Laubscher R, Bradshaw D, Dorrington R, Bourne D, Timaeus I Senior Statistician, Biostatistics Unit, MRC, Cape Town, South Africa
Reliable statistics on AIDS deaths are not available, despite extensive efforts to improve vital registration in SA. The most recent cause of death statistics are for the year |1996, before deaths from the current epidemic were significant. Furthermore, vital registration statistics may well be unreliable because, although AIDS deaths have been recorded on death certificates since the 1980's, the true cause of death of someone dying from AIDS is often misreported. We have made use of a routine administrative data set, which includes age- and gender-specific mortality data, i.e. without explicit information on cause of death. Indirect demographic techniques have been applied to estimate the extent of under-reporting of deaths during the study period. The coverage of death registration has increased from around 50% of deaths registered in 1990 to over 90% of adult deaths being registered in 2000. Data on adults show a fairly consistent rise in young adult mortality for both men and women. The rise is most noticeable for young women in the age group 25-29 years, where rates are now over three times what they were a decade ago. The recent increases in mortality among men are spread out over an older and wider age range, in keeping with the different age patterns of HIV infection between men and women - a consequence of older men tending to have sex with younger women. Alternative explanations for this increase have been explored and comparison with the projections from AIDS models leads to the conclusion that this increase is largely a result of AIDS. On this basis, it is estimated that 40% of the adult deaths aged 15 to 49 that occurred in the year 2000 were due to AIDS and that about 20% of all adult deaths in that year were due to AIDS. Monitoring the age pattern of all cause mortality has provided important information that can be used in conjunction with models to provide information on the impact of AIDS.
Keywords: AEGIS, Acquired Immunodeficiency Syndrome, HIV Infections, Death Certificates, Demography, Disease Outbreaks, South Africa, Human, Male, Female, Adult, mortality, epidemiology
