Base Case Inputs and Assumptions.

*Low and high values of the base case are provided for inputs that vary based on the country case scenario. Each case scenario is characterized by low or high values on three factors: the prevalence of a reactive syphilis serological test in pregnant women, the percentage of women tested and treated for syphilis in ANC at the current level of services, and the relative cost of health services (including the cost of PMTCT of syphilis AOs). Accordingly, the percentages of women attending ANC and tested and treated for syphilis under the expanded program vary by case scenario, as do the costs of syphilis AOs and of screening and treating syphilis in the mother. All other base case inputs are constant across the eight case scenarios.
**Two sensitivity analysis ranges (high and low) are provided for inputs that vary based on the country case scenario. For all other inputs explored in SA, a single range is given
†For each MTCT of syphilis AO we estimate the cost by subtracting the cost of the healthy childbirth from the cost of the AO; this might be overestimate the benefits of preventing MTCT of syphilis as it assumes that all infants would otherwise be born healthy.
? Negative costs imply savings. Based on published data from South Africa, the cost of delivery of a stillborn infant was assumed to approximate the cost of delivery of a healthy infant (i.e., $58), and the cost of a spontaneously aborted pregnancy was ∼$57). We estimated the cost of a stillbirth/2nd/3rd trimester fetal loss as the cost of a spontaneously aborted pregnancy minus the cost of a normal delivery (what the cost would have been in the absence of the AO), i.e., −$1. In settings where the cost of health care services is high, the estimate is −$1, and in those where the cost of services is low, the estimate is $0 (because of the adjustment for the cost per hospital day).