Background Syphilis is a major public health problem in many regions of China, with increases in congenital syphilis (CS) cases causing particular concern. Precisely determining the extent of CS burden has been challenging given incomplete prenatal screening coverage and limited CS diagnostic capacity. As a result, both net under and over-reporting are plausible. In addition, there are a range of policies that could be expanded as part of the new National Syphilis Control Plan. This paper uses a Markov model to estimate the number of CS cases and other adverse outcomes and prioritise control measures in Guangdong Province, China.

Methods We developed a Markov model that incorporated age-stratified fertility rates, female adult syphilis reported cases, and empirical syphilis transmission rates to estimate the number of adverse pregnancy outcomes associated with prenatal syphilis infection on a yearly basis from 2005 to 2009. Guangdong Province was the focus of this analysis given access to high quality demographic and public health data required for model inputs. Adverse outcomes examined included CS, stillbirth, neonatal death, and low birth weight. Sensitivity analyses were performed to identify variables most influential for achieving WHO and Chinese program benchmarks.

Results Our model estimates 264 CS cases per 100?000 live births in Guangdong during 2009, substantially greater than the 129 reported CS cases per 100?000 live births. This is consistent with a net 49% under-reporting of CS cases in 2009. Higher syphilis prevalence among women in their peak childbearing years (20–30 years old) is closely related to greater numbers of CS cases see Abstract P1-S1.47 Figure 1. Expanding prenatal screening coverage was the single most important mechanism for preventing CS cases; increasing prenatal screening from 50% to 100% coverage would be associated with 98 CS cases averted per 100?000 live births. Alternatively, a comprehensive syphilis control strategy including 100% prenatal screening, 100% treatment completion, and 100% early screening in pregnancy would be associated with 212 CS cases averted per 100?000 live births.