Investigating Risk Factors Affecting the United States Male and Female Syphilis Rates Using Artificial Neural Network

Syphilis remains a public health concern here in the United States and demands attention in order to reduce the number of new cases developing each year, as well as to establish operational prevention efforts. Although there was not a remarkable change in female syphilis incidence rate, there was a 37% increase in male syphilis incidence rates among U.S. men from 2000 to 2007 (CDC Wonder, 2008). The purpose of this study is to investigate risk factors that contribute to syphilis incidence rates among both sexes in the United States. U.S. male and female syphilis incidence data from 1984 to 2007 was collected and analyzed using a longitudinal database from the Centers for Disease Control and Prevention. Variables such as state median income, state unemployment rate, state poverty rate, as well as other risk factors were entered into the Artificial Neural Network (ANN) model. In the ANN model, prediction accuracy can be achieved by using the feed-forward of inputs, back-propagation of errors, and the adjustment of connection weights. Our research findings show that male and female syphilis rates in the U.S. are affected by state alcohol consumption and percent of the population with less than nine years education, in ranking order. This finding is vital as it emphasizes a population in which other socioeconomic risk factors are present. Our study suggests that prevention efforts should be directed towards both sexes in the U.S. in order to reduce the rate of syphilis infections.