The prevalence for human immunodeficiency virus (HIV) in South Africa is 18% among 15-49 year old adults and 30% among female antenatal clinic attendees (UNAIDS, 2007), indicating continuing need for effective HIV prevention. Further, recent studies in sub-Saharan Africa found 60-94% of new HIV infections are occurring within marriage or co-habiting heterosexual partnerships (Dunkle et al., 2008). These findings signal the need for HIV prevention interventions that target couples in South Africa. This study is a randomized controlled trial of a behavioral intervention to increase HIV testing among couples living in Vulindlela, South Africa.

The proposed intervention consists of six sessions (one mixed gender group, one single gender group, and four couples' counseling sessions). Using a randomized controlled trial (RCT) design with 350 heterosexual couples, we will test the hypothesis that compared with a one-time mixed-gender group session, the proposed intervention will improve communication, intimacy and trust necessary for mutual decision-making about behaviours related to sexual risk behaviour and testing for HIV. Improving couples' ability and motivation to participate in Couples HIV Testing and Counseling (CHTC) for HIV will in turn lead to reductions in sexual risk behaviour. Both of these outcomes are necessary and effective strategies to reduce the risk of HIV transmission within primary partnerships.

This intervention is informed by several qualitative studies conducted in Vulindlela and Soweto, South Africa via a K08 award from NIH, as well as other funding sources. These preliminary studies provided insight into the challenges couples face in participating in CHCT, as well as the skills they need in order to address these barriers. Our experience conducting both qualitative and quantitative studies with comparable populations (i.e. South African couples) has also informed the recruitment and retention methods in this intervention. The proposed study takes advantage of the infrastructure and collaborative relationships that the PI has developed that have enabled her to implement and conduct research within these communities.

The specific aims of the project are to test the efficacy of a theory-based and culturally appropriate couples-based intervention on the following outcomes:

1. Rates of testing for HIV,
2. Sexual risk behaviour for HIV (with primary and any concurrent partners).

In addition we will evaluate the extent to which hypothesized mediating factors (e.g., relationship dynamics) explain the major outcomes and the extent to which the intervention affects these factors.

Ultimately, our goal is to facilitate the outcome that members of partnerships learn their own and their partner's HIV status. This is a crucial step for effective behavioural risk reduction, yet it is a relatively uncommon occurrence for partners in Vulindlela, South Africa. Specifically, mutual disclosure of HIV status accomplishes two important goals.

First, this knowledge can facilitate risk-reduction behaviour within partnerships via effecting positive changes (e.g., condom use) in sexual behaviour with primary and any concurrent partners. Second, knowledge of HIV status can increase access to treatment and care for HIV-positive individuals, as well as reinforce behavioural choices (e.g., limiting concurrent partners) to stay HIV-negative. As couples are particularly vulnerable for HIV infection in this context, increasing testing for HIV and reducing likelihood of behavioural transmission of HIV within partnerships would be a high-impact outcome with the potential to significantly reduce the impact of HIV in an area already severely affected by the pandemic.

Inclusion Criteria:

• both partners 18 or older
• in a relationship with opposite sex partner (defined as: "are you currently in a relationship with a person of the opposite sex to whom you feel committed above anyone else and with whom you have had sexual relations?")
• Both partners report the relationship length has lasted at least 6 months
• Both partners willing and able to provide informed consent

Exclusion Criteria:

• Both partners report participating in couples-based testing for HIV with their current partner (Defined as: Both partners tested for HIV at the same time and disclosed results).
• Either partner reports current marital status as polygamous
• Either partner reports domestic violence (as actor or recipient) within the past 6 months with the current partner.
• Both partners have ever disclosed HIV test results to each other via individual or couples testing