Kamala Mallik

Date of Award


Document Type

Open Access Thesis

Degree Name

Master of Public Health (MPH)



Prison and jail inmates constitute a growing population at risk for HIV infection. Inmates are at high risk for HIV infection because of pre-incarceration drug use; 60% of prison and jail inmates reported using drugs regularly at some point in their lives and 27% to 41% of inmates have a history of injection drug use. The high concentration of drug users in correctional facilities offers unparalleled access to a population that is otherwise difficult to reach for public health surveillance and intervention.

The Baltimore City Correctional Facility HIV Serosurvey was conducted by the Baltimore City Health Department between June and November, 1992 in order to estimate the seroprevalence of HIV-1 antibody and characterize the risk factors for HIV infection among entrants to a Baltimore, Maryland correctional facility. Since risk behavior information is not routinely collected at the correctional facility and the substance abuse literature suggests that certain patterns of criminality are associated with narcotics addiction, investigators at the Baltimore City Health Department were interested in researching whether incarceration characteristics such as criminal charge and recidivism could serve as a valid marker for HIV risk. This serosurvey was a cross-sectional unlinked (blinded) study. Excess sera collected from routine syphilis screening were unlinked from all personal identifiers and tested for the presence of HIV-1 antibody. Demographic and incarceration-related data were collected and linked to the sera by means of an anonymous survey identification number. Chi square tests of association and linear logistic regression were used to analyse the relationships between demographics, incarceration characteristics, and HIV risk.

HIV seroprevalence at the correctional facility was estimated to be 10.9%, but was as high as 23% in some sub-populations. Variables significantly associated with increased odds of HIV infection include black race, older age, previous incarceration, multiple incarceration in the last five years, property crime charge, morals charge, and charge for an income-generating crime. Narcotics charge was not associated with HIV infection. The findings of the serosurvey are consistent with the literature, which suggests that drug users commit income-generating crimes such as burglary, robbery, drug dealing, and prostitution to finance their addictions. Although the results of the study were informative, one cannot conclude that incarceration characteristics are valid surrogates for HIV risk behavior information, as the scope of the serosurvey was limited by both a poor phlebotomy success rate (71%) and the survey's cross-sectional nature. The findings of this serosurvey, however, are useful for identifying subpopulations of entrants who are at particularly high risk for HIV infection compared to the rest of the entrant population and in need of public health intervention.

Open Access

This Article is Open Access

Included in

Public Health Commons