Application of Two-level Fractional Factorial Design to Determine and Optimize the Effect of Demographic Characteristics on HIV Prevalence using the 2006 South African Annual Antenatal HIV and Syphilis Seroprevalence data

A Two-Level Fractional Factorial design was employed to develop and optimize the combination of demographic characteristics that has the greatest effect on the spread of HIV in the South African population. HIV prevalence (dependent variable) was found to be highly sensitive to changes in the mother’s age (15-55years) and level of education (Grades 0-13) (independent variables), using the 2006 South African Annual Antenatal HIV and Syphilis Seroprevalence data. HIV prevalence was the optimization objective (R = 0.842, Durbin-Watson Index = 3.440%, Coefficient of Variation (CV) = 52.169%). The Lagrangian technique produced no significant difference (P>0.05) between the experimental and predicted HIV prevalence values. General Terms Two level fractional factorial design