Sexual behaviours of people living with HIV : implications for prevention with positives

In 2009, Kenya had an estimated 1.5 million people living with HIV and in the same year an estimated 110,000 people were newly infected with HIV. HIV continues to spread in the population despite the existence of prevention programmes. HIV can only be spread from a HIV-positive person to HIV-negative person or in some cases even to HIV-positive person with a different viral strain. Thus PLHIV are a key target group for prevention programmes. This thesis is based on four research studies conducted with PLHIV in Mombasa, Kenya between 2004 and 2008. The first of these studies (articles 1-3) were conducted with PLHIV when ART was first rolled out in Kenya with USAID support. These studies were among the first to examine sexual risk behaviours in the context of ART. Findings from these studies documenting higher risk behaviours among PLHIV co-trimoxazole prophylaxis for opportunistic infections led to the fourth study exploring sexual risk behaviours among PLHIV in the community who were not receiving any treatment. Evidence from that study has led to a fifth study presently underway in Mombasa that evaluates a community based HIV prevention intervention delivered by CHWs for PLHIV in the community not receiving treatment. The studies presented in this thesis provide an overview of sexual behaviours of three different subgroups of PLHIV: (1) PLHIV who a receiving ART and therefore are in regular contact with health workers and prevention services; (2) PLHIV who are not eligible for ART but are visiting health services for follow up services and receiving cotrimoxazole prophylaxis for opportunistic infections and therefore continue to have contact with health workers, and (3) PLHIV in the community who are not accessing HIV care services and therefore do not have contact with health workers or prevention services. Key findings: Sexual risk behaviours defined as multiple sexual partnerships and unprotected sex with partners are prevalent to variable degree among the various PLHIV subgroups; PLHIV not receiving any treatment exhibit the highest level of risky behaviours. A major concern to prevention programmes is the reporting of unprotected sex with sero-discordant or unknown status sexual partners. A large section of PLHIV have partners of unknown HIV status or untested partners; 75 percent of the partners of PLHIV not accessing HIV care services were of unknown HIV status. Non-disclosure of HIV status to partners, perceived or internalized stigma and depression were significantly associated with unprotected sex. We did not find any evidence of an increase in sexual risk behaviours after 12 months of ART; in fact we noted a significant decrease in the proportion of participants reporting unprotected sex with HIV-negative or untested partners after 12 months on ART. Qualitative data revealed that PLHIV attempt to control risky behaviours after testing positive or starting ART, but struggle with disclosure, fear of abandonment, partner reaction or refusal to test, and misconceptions and religious beliefs pertaining to safe sex, condom use and gender norms. Further research is needed to better understand the factors that influence sexual risk behaviours. There is urgent need for intervention research to identify effective HIV prevention strategies or interventions that are acceptable in this community