HIV Knowledge Among Cisgender Female Sex Workers of Haitian Descent Working at the Border of Haiti and Dominican Republic

• Main Authors: Henna Budhwani, Kristine R. Hearld, Julia Hasbun, John Waters
• Format: Article
• Language: English
• Published: Frontiers Media S.A. 2021-10-01
• Series: Frontiers in Reproductive Health
• Subjects: HIV; sex workers; Haiti; Dominican Republic; HIV knowledge
• Online Access: https://www.frontiersin.org/articles/10.3389/frph.2021.700861/full

In this brief report, we aim to assess levels of HIV mis-information among cisgender Haitian female sex workers engaged in sex work at the Haiti and Dominican Republic border. We conducted bivariate analyses on the 2014 Border Study on Sex Workers comparing responses from female sex workers on the Haiti side of the border to those from their peers on the Dominican Republic side (N = 212). Prevention of HIV acquisition by correct and consistent condom use with each sex act was correctly endorsed by 90.5% of female sex workers in Haiti but only 57.0% of their peers in Dominican Republic (χ2 = 32.28, p < 0.001); 84.1% of respondents in Haiti correctly identified that HIV can be transmitted through a single unprotected sexual act, compared to 52.3% in Dominican Republic (χ2 = 25.2, p < 0.001). Significantly higher percentages of female sex workers in Dominican Republic correctly responded that HIV can be transmitted in pregnancy, compared to respondents in Haiti (96.5 vs. 71.4%; χ2 = 21.42, p < 0.001). Higher percentages of respondents in Dominican Republic correctly answered that HIV can be transmitted through needle sharing, relative to respondents in Haiti (100.0 vs. 89.7%; χ2 = 9.45, p < 0.01). Respondents in Dominican Republic more accurately rejected the possibility of transmission through food or through mosquito bites, compared to respondents in Haiti (95.4 vs. 81.8%, χ2 = 8.51, p < 0.01; 97.7 vs. 86.5%, χ2 = 7.81, p < 0.01, respectively). Findings indicate that if HIV knowledge is examined aggregating responses to individual questions, then elements of misinformation may remain unaddressed. For example, we found significant differences in correct answers ranging from 16.7 to 100.0%.