Screening for syphilis and HIV in North Carolina jails Public Deposited

Downloadable Content

Download PDF
Last Modified
  • March 21, 2019
  • Sampson, Lynne Arlene
    • Affiliation: Gillings School of Global Public Health, Department of Epidemiology
  • Sexually transmitted diseases are prevalent among incarcerated populations. Screening for STDs in county jails serves as a form of community screening, often reaching individuals with poor access to other health care services. The goal of this dissertation was to develop and test screening algorithms to improve the effectiveness of jail screening for syphilis and human immunodeficiency virus (HIV) infection and to examine the costs of adding syphilis screening to existing HIV programs. The studies included men and women screened for syphilis (and some also for HIV) in seven North Carolina jails in 2002-2005. A screening algorithm derived from predictive modeling of new syphilis cases can improve screening efficiency for male inmates. Age, race/ethnicity, and reporting an STD diagnosis in the last six months were all associated with new syphilis infections. When resulting risk scores were applied to hypothetically testing ~50% of the inmate population, the algorithm was able to detect 83% of the cases. Women were more likely than men to have syphilis (OR 2.5, 95% CI 1.8 - 3.4) but the data did not yield a useful predictive model. The prevailing strategy of screening as many women as possible is recommended. Programmatic and funding changes have resulted in a shift to HIV screening in NC jails and new protocols must be designed with HIV as the primary goal. Screening for syphilis under this new paradigm is effective and low cost and should continue. A predictive model of HIV infection among jail inmates included age, race/ethnicity, gender, history of HIV testing and, for men, men who have sex with men status. Risk scores derived from the model yielded screening algorithm with 83% sensitivity for detection of HIV when applied to testing ~50% of the population. This same algorithm was able to detect 73% of syphilis cases. Using the algorithm for targeted screening decreased the cost per HIV case detected from ~$2,200 to ~$1,300. The cost of adding syphilis to the existing HIV jail screening program was low (less than $300 per case detected) and is recommended in areas with incident syphilis.
Date of publication
Resource type
Rights statement
  • In Copyright
  • Miller, William
Degree granting institution
  • University of North Carolina at Chapel Hill
  • Open access

This work has no parents.