Monitoring the social and other impacts of HIV and AIDS in South Africa: indicators reflecting mortality

Abstract

The purpose of this study was to monitor the social and other impacts of HIV and AIDS in South Africa by reviewing selected indicators reflecting mortality.

The methods used for reviewing, the selected indicators included a systematic search for the most recent data, reviewing the indicators using the most recent data and updating and defining indicators through validation interviews, where necessary, with key informants from selected government departments and parastatal organisations.

HIV and AIDS have had a serious impact on mortality and other demographic variables in South Africa. By using the crude death rate of the age group 15-49 it is possible to identify changes in mortality levels in the population. The crude death rate can be used to reflect the mortality implications of HIV/AIDS. However, the relative ease of obtaining information to review and update this indicator has helped to understand its usefulness. The death rate in the economically active age group of 15-49 was at 46.1% or 230,300 people per annum in 2002. The actuaries estimated the under-five mortality rate at 100 per 1,000 live births in 2002 while the SADHS 2003 estimated it at 58 per 1,000 live births. U5MR is an important indicator for monitoring both the demographic and health impacts of HIV/AIDS.

Falling life expectancy has been one of the most visible impacts of HIV/AIDS on the human development. Less visible has been the feminization of the disease and the consequences for gender equity. Correctly calculated, this indicator provides the most accurate index to indicate the demographic impact of the HIV/AIDS epidemic. However, data issues and the need to produce life tables annually is a disadvantage. The SADHS (2003) and Dorrington et al. (2004) both estimated the life expectancy at birth to be 50.7 years. There are slight disparities in the IMR - with Stats SA 2006 and Dorrington et al. 2004 estimating it at 56 per 1,000 live births, Bradshaw and Nannan (2004), had a higher estimate of 59 per 1,000 live births while the SADHS 2003 had a far lower estimate of 43 per 1,000 live births.