Factors associated with voluntary HIV testing and serostatus among North Carolina state prisoners, 2004-2006 Public Deposited

Downloadable Content

Download PDF
Last Modified
  • March 21, 2019
  • Rosen, David L.
    • Affiliation: Gillings School of Global Public Health, Department of Epidemiology
  • Background. Despite the heavy burden of HIV among prisoners, HIV testing strategies vary widely across US state prison systems. A voluntary HIV testing program in a large southern state prison system was evaluated by estimating HIV testing rates among prisoners, identifying inmate characteristics associated with testing status and infection status, and estimating the number of infected prisoners who remained untested during their incarceration. Methods. Electronic imprisonment and lab records were obtained for all prisoners entering the NC DOC between January 2004 and May 2006. Associations between inmate characteristics and both HIV testing and HIV serostatus were estimated separately using loglinear and logistic regression. The number of undetected cases was estimated using age-sexrace specific HIV prevalences from tested prisoners and from statewide HIV reporting. Results. Eight-five percent of female and 31% of male prisoners were tested for HIV during their incarceration. In four of the six intake prisons for men, <15% of prisoners were tested. Among men, the proportion tested was 10% higher among those reporting heroin use, crack/cocaine use, tuberculosis disease, and any conventional HIV risk behavior (e.g. sharing needles, MSM), but >60% of men reporting a conventional risk behavior remained untested. In bivariate and covariate-adjusted analyses, black men were 30% and 13%, respectively, less likely than whites to be tested. Nearly 3.4% (718/21,419) of tested prisoners were HIV+. Of those HIV+ prisoners, <50% reported a history of any conventional HIV risk behaviors. Infection was most strongly associated with being a MSM (OR=8.0), non-white race (OR=6.2-7.4), and ages 35- 44 years (OR=4.1). The strongest risk factor among women was black race (OR=3.8); ORs < 3.0 were observed for several other risk factors. Sixty-five percent of HIV+ prisoners were HCV-coinfected. Between 23% and 63% of HIV cases remained undetected. Conclusion. HIV testing varied greatly by intake prison, and many male inmates were never tested. The majority of cases denied conventional HIV risk behaviors suggesting limitations of risk-factor based testing. Expanded HIV testing could improve case finding. However, testing expansion must be joined with adequate treatment and follow-up services, and monitoring is needed to ensure testing is not coercive.
Date of publication
Resource type
Rights statement
  • In Copyright
  • Schoenbach, Victor
  • Open access

This work has no parents.