Country-specific model inputs.

Distribution in the PSAs is given in parentheses in the column heads.aThe weighted average rate of 40.7% was used in cases where the proportion of women screened for syphilis infection in the antenatal care setting was not reported.bThe syphilis prevalence estimate of 1.9% was reported in the 2002 HIV/AIDS surveillance update by the WHO [23]. For this specific country, syphilis prevalence data from the antenatal care setting were not available, neither in one of the WHO reports nor in the scientific literature.cThe 2002 HIV/AIDS surveillance report by the WHO [23] reported a syphilis prevalence of 11% in the general female population; a study by Bertherat et al. [51] reported 8.6%, also in the general female population; and a joint report by the WHO, Joint United Nations Programme on HIV/AIDS, and United Nations Children's Fund [19] in the antenatal care setting reported 1%, although with a cautionary note that the data may not be nationally representative. Therefore, the lower of the two other estimates was used.dThe syphilis prevalence estimate of 7.0% was reported in the 2002 HIV/AIDS surveillance update by the WHO [23]. For this specific country, syphilis prevalence data from the antenatal care setting were not available, neither in one of the WHO reports nor in the scientific literature.eThe syphilis prevalence estimate of 3.0% was reported in the 2007 HIV/AIDS surveillance update by the WHO [24]. For this specific country, syphilis prevalence data from the antenatal care setting were not available, neither in one of the WHO reports nor in the scientific literature.fThe 2002 HIV/AIDS surveillance report by the WHO [23] reported a syphilis prevalence of 20.3% in the antenatal care setting, the 2007 HIV/AIDS surveillance report by the WHO [24] reported 8.3% in the general female population, and a joint report by the WHO, Joint United Nations Programme on HIV/AIDS, and United Nations Children's Fund [19] in the antenatal care setting reported 5.3%, although with a cautionary note that the data may not be nationally representative. Therefore, the lower of the two other estimates was used.