Reducing the spread of HIV infection in sub-Saharan Africa: some demographic and economic implications
Rowley, J.T.; Anderson, R.M.; Ng, T.W.
Aids 4(1): 47-56
1990
ISSN/ISBN: 0269-9370 PMID: 2317308 Document Number: 367707
The paper examines, through mathematical models, the possible demographic and economic consequences of reducing the rate of heterosexual HIV transmission in a "typical" sub-Saharan African country. The model of HIV transmission was developed by the same group and is extensively described in earlier papers. A key parameter ( beta c) in the model is the average number of persons per year to whom HIV is transmitted by an infected individual. In a "no-control" situation this number is assumed to be about 0.6. Under this assumption, the demographic impact of AIDS is profound, halving the population 100 years after the introduction of HIV.Several control scenarios are considered: reducing beta c by half at different stages in the epidemic; reducing beta c by varying amounts 25 years into the epidemic (roughly now in certain countries). The reduction in beta c can be achieved by the introduction of safer sex practices (reducing the probability of transmission from an infected to an uninfected person) or by reducing the rate of partner change. The effect of these changes in sexual behaviour is investigated for (i) total population, (ii) prevalent AIDS cases, (iii) age structure of the population, (iv) annual cost of treating all AIDS patients, (v) affordable treatment costs per AIDS patient given a fixed health budget.The conclusion is that a reduction in beta c by half or more has a profound effect on all outcome measures, for example, preventing a decline in the total population and halving the %GNP spent on treating AIDS patients. The earlier the introduction of effective control measures in the course of the epidemic, the greater the derived benefits. The paper shows the sensitivity of the mathematical model to assumptions about parameter values. If the "no control" value for beta c is marginally inaccurate quite different quantitative conclusions would have been reached.David Dunn.