Tructure [19,20], the model identifies four sexual activity groups ranging in the number of new sexual partners per year [21]. Data about the proportion of individuals in a particular sexual activity group and their number of new partners are not available. Using the Monte Carlo filtering techniques 
[22] we parameterized the different sexual activity groups and only accepted the 1795 simulations that were associated with a prevalence of 7.7 (60.05 ) from 2002 until 2009 in accordance with Macha. Monte Carlo filtering allowed us to test the impact of PrEP over a wide range of sexual activities, as a wide variety of sexual risk group combinations resulted in the appropriate HIV prevalence (Table 1). In summary, the highest sexual activity group had an average of 13 new partners per year and made up on average just 2 of the population, representing a core group of highly order 4 IBP sexually active individuals. This group is instrumental in determining the peak of the epidemic. Only simulations where this group was small and their number of partners were high allowed the epidemic to peak appropriately. The second highest sexual activity group had on average 2 new partners per year and made up a more substantial 18 of the population, representing individuals whom are not in steady or monogamous relationships. This is the group is an important factor in determining where the equilibrium of theMethods Setting and PopulationOur model is based on the rural SPDP Crosslinker web population of Macha, Zambia and using data from the HIV Clinic at Macha Hospital. Macha 
is located in the Southern Province of Zambia, and approximately 80 km away from the nearest town, Choma [8]. The hospital serves as a district-level referral hospital for rural health centers within an 80 km radius, with 90,000 persons that are aged 12 years and over in the Macha Hospital catchment area [8]. The antenatal prevalence between 2002 [9] and 2009 [local data] was stable around 7.7 . Macha Hospital has provided care to over 7500 HIV-infected adults and children since 2005 through the Government of Zambia’s antiretroviral treatment program, with additional support from the President’s Emergency Plan for AIDS Relief (PEPFAR) through the non-governmental organization, AidsRelief [8]. Since the start of the clinic in 2005, treatment is implemented according to WHO guidelines, initially at CD4,Cost-Effectiveness of PrEP, Zambiaepidemic is reached. The only simulations that were accepted into the analysis were the ones in which this group allowed the epidemic to reach an equilibrium prevalence of 7.7 (60.05 ) from 2002?009 in accordance with Macha data. The two lowest groups had ,1 new sexual partner per year, representing individuals in long term relationships or marriages. The final distribution of proportion of sexual activity groups and number of new partners per year are given in Figure S2. Other variables used to calibrate the model included: transmissibility during the acute stage of infection, transmissibility during the AIDS stage of infection, the rate at which individuals moved from acute to chronic infection, rate at which individuals move from the AIDS stage to the AIDS final stage, and the rate of mixing 1313429 between sexual risk groups (epsilon). Full model description including equations can be found in the Text S1. HIV testing. Approximately 10 of individuals aged 12 and older undergo an HIV-test yearly in Macha. In our model, we studied the impact on the HIV-epidemic of test rates that were rang.Tructure [19,20], the model identifies four sexual activity groups ranging in the number of new sexual partners per year [21]. Data about the proportion of individuals in a particular sexual activity group and their number of new partners are not available. Using the Monte Carlo filtering techniques [22] we parameterized the different sexual activity groups and only accepted the 1795 simulations that were associated with a prevalence of 7.7 (60.05 ) from 2002 until 2009 in accordance with Macha. Monte Carlo filtering allowed us to test the impact of PrEP over a wide range of sexual activities, as a wide variety of sexual risk group combinations resulted in the appropriate HIV prevalence (Table 1). In summary, the highest sexual activity group had an average of 13 new partners per year and made up on average just 2 of the population, representing a core group of highly sexually active individuals. This group is instrumental in determining the peak of the epidemic. Only simulations where this group was small and their number of partners were high allowed the epidemic to peak appropriately. The second highest sexual activity group had on average 2 new partners per year and made up a more substantial 18 of the population, representing individuals whom are not in steady or monogamous relationships. This is the group is an important factor in determining where the equilibrium of theMethods Setting and PopulationOur model is based on the rural population of Macha, Zambia and using data from the HIV Clinic at Macha Hospital. Macha is located in the Southern Province of Zambia, and approximately 80 km away from the nearest town, Choma [8]. The hospital serves as a district-level referral hospital for rural health centers within an 80 km radius, with 90,000 persons that are aged 12 years and over in the Macha Hospital catchment area [8]. The antenatal prevalence between 2002 [9] and 2009 [local data] was stable around 7.7 . Macha Hospital has provided care to over 7500 HIV-infected adults and children since 2005 through the Government of Zambia’s antiretroviral treatment program, with additional support from the President’s Emergency Plan for AIDS Relief (PEPFAR) through the non-governmental organization, AidsRelief [8]. Since the start of the clinic in 2005, treatment is implemented according to WHO guidelines, initially at CD4,Cost-Effectiveness of PrEP, Zambiaepidemic is reached. The only simulations that were accepted into the analysis were the ones in which this group allowed the epidemic to reach an equilibrium prevalence of 7.7 (60.05 ) from 2002?009 in accordance with Macha data. The two lowest groups had ,1 new sexual partner per year, representing individuals in long term relationships or marriages. The final distribution of proportion of sexual activity groups and number of new partners per year are given in Figure S2. Other variables used to calibrate the model included: transmissibility during the acute stage of infection, transmissibility during the AIDS stage of infection, the rate at which individuals moved from acute to chronic infection, rate at which individuals move from the AIDS stage to the AIDS final stage, and the rate of mixing 1313429 between sexual risk groups (epsilon). Full model description including equations can be found in the Text S1. HIV testing. Approximately 10 of individuals aged 12 and older undergo an HIV-test yearly in Macha. In our model, we studied the impact on the HIV-epidemic of test rates that were rang.