Prevalence of and risk factors for Human Immunodeficiency Virus (HIV) infection in entrants and residents of an Ethiopian prison.

Background: Prisoners generally have a higher prevalence of HIV infection compared to the general population from which they come. Whether this higher prevalence reflects a higher HIV prevalence in those entering prisons or intramural transmission of HIV within prisons or both is unclear. Any of these possibilities would increase the prevalence found in resident prisoners above that in the general population. Moreover, comparisons of HIV prevalence in entrants and residents and in men and women in African prisons are not well documented. The purpose of this study was to estimate and compare the prevalence and risk factors for HIV infection amongst both male as well as female and entrant and resident prisoners in a large Ethiopian Federal Prison.

Methods: We studied consenting prisoners cross-sectionally from August 2014 through November 2016. Prison entrants were screened continuously for HIV infection and its associated risk factors and residents were screened in two waves one year apart. HIV was diagnosed at the prison hospital laboratory based on the Ethiopian national HIV rapid antibody testing protocol. An external, internationally-accredited reference laboratory confirmed results. Agreement of results between the laboratories were assessed.

Results: A total of 10,778 participants were screened for HIV. Most participants were young (median age of 26 years, IQR: 21-33), male (84%), single (61%), literate (89%), and urban residents (91%) without prior incarceration (96%). Prevalence of HIV was 3.4% overall. Rates of HIV (p = 0.80) were similar in residents and entrants in wave 1 and in entrants in both waves, but were 1.9-fold higher (5.4% vs 2.8%) in residents than entrants in wave 2 (both p<0.001). At entrance to the prison women were more likely to be HIV+ than men (5.5% in women vs 2.5% in men, p< 0.001). In contrast resident women were less likely to be HIV+, but this difference was not statistically significant (3.2% in women vs 4.3% in men, p = 0.125). Other risk factors associated with HIV infection were increasing age (p<0.001), female gender (p<0.001), marital status (never vs other categories, p = 0.016), smaller number of rooms in their houses pre-imprisonment (p = 0.031), TB diagnosis ever (p<0.001), number of lifetime sex partners (especially having 2-10, p<0.001), and genital ulcer (p = 0.037).

Conclusions: Prevalence of HIV in the residents at this large, central Ethiopian prison was higher than that estimated for the general population and lower than in many other studies from other smaller Ethiopian prisons. A higher prevalence in residents than in entrants were found only in our second wave of screening after one year of continuous screening and treatment, possibly representing increased willingness of residents at increased risk of HIV to participate in the second wave. Thus, this findings did not clearly support intramural transmission of HIV or the effectiveness of screening to reduce prevalence. Finally, the higher HIV prevalence in women than men requires that they be similarly screened and treated for HIV infection.