The deterministic mathematical model used, which is compartmental in structure, describes dynamically the course of both chlamydial and HIV infections in an active heterosexual population stratified by sex and sexual activity. The different stages of HIV and chlamydial infections are represented by six compartments. In the model, an individual can be susceptible to HIV, be infected with HIV but asymptomatic, or have full-blown AIDS. Each of these groups could be infected with an STD or not. Individuals can pass from one disease state to another at different rates, depending on the demographic and behavioral characteristics of the population as well as the natural history of the STD and HIV infections. The details of the model, including information on the system of nonlinear differential equations describing changes in the size of the population with time for the different disease states, are described by Boily and others (in press) in an upcoming paper.
The numerical studies of the model are based upon an initial population size of 171,481,800 individuals corresponding to the general, sexually active, heterosexual population of the United States in 1995 (Leigh et al., 1993; CIA, 1995; U.S. Census Bureau, 1996). The population growth rate is assumed to be 1.1 percent in absence of HIV infection, with a 1.01:1.00 female-to-male ratio (CIA, 1995; U.S. Census Bureau, 1996). It is assumed that an individual remains sexually active for a period of 55 years from age 15 to 70 (Anderson and Dahlberg, 1992; Leigh et al., 1993; Seidman and Rieder, 1994). Each gender is divided into six sexual activity classes to represent people with different rates of partner acquisition.
The most important assumptions when evaluating the potential impact of STDs on heterosexual HIV transmission center around HIV transmission probabilities in the absence of STDs, the sexual network and the distribution of sexual activity in the general population, the prevalence of STDs, and the nature and the magnitude of the interrelationships between STDs and HIV infection. These parameters determine whether HIV, in the presence or absence of STDs, can establish in the population and what the rate of spread of HIV in different risk groups will be. In the absence of the enhancing STD, the male-to-female per partner transmission probability for HIV is assumed to be two times that of female-to-male transmission (European Study Group on Heterosexual Transmission of HIV, 1992; Garnett and Anderson, 1993b; de Vincenzi and European Study Group on Heterosexual Transmission of HIV, 1994; Mastro et al., 1994;). In addition, HIV transmission probabilities are reduced when partnerships are formed with individuals from the high-activity classes (Jewell and Shiboski, 1990; Brookmeyer and Gail, 1994; Downs and de Vincenzi, 1996) to reflect the fact that very active individuals perform fewer acts per partnership than those with fewer partners (Garnett and Anderson, 1995; Boily and Anderson, 1996; Boily et al., in press). Lastly, different mechanisms have been postulated about