Community-based implementation and effectiveness in a randomized trial of a risk reduction intervention for HIV-serodiscordant couples: study protocol

The HIV/AIDS epidemic continues to disproportionately affect African American communities in the US, particularly those located in urban areas. Despite the fact that HIV is often transmitted from one sexual partner to another, most HIV prevention interventions have focused only on individuals, rather than couples. This five-year study investigates community-based implementation, effectiveness, and sustainability of 'Eban II,' an evidence-based risk reduction intervention for African-American heterosexual, serodiscordant couples. This hybrid implementation/effectiveness implementation study is guided by organizational change theory as conceptualized in the Texas Christian University Program Change Model (PCM), a model of phased organizational change from exposure to adoption, implementation, and sustainability. The primary implementation aims are to assist 10 community-based organizations (CBOs) to implement and sustain Eban II; specifically, to partner with CBOs to expose providers to the intervention; facilitate its adoption, implementation and sustainment; and to evaluate processes and determinants of implementation, effectiveness, fidelity, and sustainment. The primary effectiveness aim is to evaluate the effect of Eban II on participant (n = 200 couples) outcomes, specifically incidents of protected sex and proportion of condom use. We will also determine the cost-effectiveness of implementation, as measured by implementation costs and potential cost savings. A mixed methods evaluation will examine implementation at the agency level; staff members from the CBOs will complete baseline measures of organizational context and climate, while key stakeholders will be interviewed periodically throughout implementation. Effectiveness of Eban II will be assessed using a randomized delayed enrollment (waitlist) control design to evaluate the impact of treatment on outcomes at posttest and three-month follow-up. Multi-level hierarchical modeling with a multi-level nested structure will be used to evaluate the effects of agency- and couples-level characteristics on couples-level outcomes (e.g., condom use). This study will produce important information regarding the value of the Eban II program and a theory-guided implementation process and tools designed for use in implementing Eban II and other evidence-based programs in demographically diverse, resource-constrained treatment settings. NCT00644163.