The Effect of Social Mixing Patterns on the Spread of AIDS

Mathematical models of the transmission of the AIDS virus can help us better understand the spread of the AIDS epidemic and prepare for the future. Model explorations can indicate which factors the epidemic is most sensitive to and provide guidance in designing interventions, educational programs and social behavior studies. We explore the sensitivity of a transmission model to different social mixing patterns. This model continuously distributes a homosexual community according to sexual partner change rates and can account for infectivity and conversion times that vary with time since infection. An acceptance function determines which partners are acceptable to an individual and defines the mixing between groups with different partner change rates. We find that if people only select partners with very similar behavior the epidemic grows much slower than if they are not as discriminating. Therefore, understanding social mixing patterns may be one of the most urgent tasks if we are to anticipate the future. We also find that the epidemic is sensitive to variable infectivity and conversion times.