The impact of HIV-related mortality on life expectancy: evidence from the Africa Centre Demographic Surveillance Area.

AIDS-related mortality has contributed significantly to the changing demographic patterns and trends in South Africa. Until the introduction of ART nationwide life expectancy had declined to below 50 years for both sexes. Reversal of this trend became noticeable in 2004 when the effect of the national roll-out of the ART program was making an impact. Objective: To measure the impact of AIDS-related mortality on life expectancy using multiple-decrement life tables during the pre-ART and ART periods using empirical surveillance data. Data: Data on cause of death are obtained using Verbal Autopsy Module used by the Africa Centre to establish the cause of death. The module aims to establish retrospectively the cause of death of the all the deceased persons in the surveillance area. The denominators are obtained by calculating person-years of exposure which is the length of time of residency of a group of people who have been observed for varying periods of time. It is the sum total of the length of time each person has been observed from the time they enter the surveillance population through birth or in-migration or leave the population through death or out-migration. Methods: Using longitudinal demographic surveillance data from the Africa Centre Demographic Surveillance System in northern KwaZulu-Natal and applying multiple decrement and associated single-decrement life-table methods the total number of person-years lost or alternatively would have been saved had HIV/AIDS been eliminated from the population at specific time points are calculated given the age- and cause-specific mortality conditions of the period and gains in life expectancy that are implied at those time points. Cause of death data which are essentially needed for these methods are collected through the verbal autopsy and person-years of exposure are accurately recorded through tracking of vital events. Results: At the start of the surveillance in 2000 30% of the deaths were due to HIV/AIDS for both sexes. There has been a gradual increase in HIV/AIDS-related mortality over the period. At the peak of the epidemic potential years of life at birth lost were 16 years for females and 10 years for males. However the turning point seems to be in 2004 and it has been hypothesised that this is the period when ART was becoming widely available in the area. Sustained declines in death rates were only felt after 2006 when the life expectancies at birth started to increase above 55 years for both sexes. The impact of AIDS-related mortality in the population is reflected in the similarities in the mortality schedules for with or without AIDS up to age 50. After age 50 the effect of AIDS-mortality is exceeded by other causes which are mainly attributable to degenerative diseases. The shape of the mortality schedule has the bi-modal distribution of HIV/AIDS in a population where the effects are high at infancy and age group 30-49 years. This pattern is similar to that observed for Southern African demographic surveillance sites during the compilation of model life tables for INDEPTH Network. Also the predictive models of life expectancy show that without HIV the South African population would have reached the post-transitional phase with very low mortality levels. Thus the role of HIV in reversing earlier transitional gains in mortality is shown in this study by the low life expectancies during peak of the epidemic. Finally increased life expectancies attest to the improved outcomes of HIV-infected patients even in rural populations since the introduction of combination antiretroviral therapy (ART) Conclusion: In conclusion the life expectancy which is an important indicator for MDGs had shown great improvement even in rural populations in South Africa since the introduction of ART in 2004.