Uptake and impact of HIV combination (HIVCOMB) interventions on HIV incidence among fishing communities in Uganda

HIV-1 remains a major global health problem, with 2.5 million new infections in 2011, of which 70% were in sub-Saharan Africa. In countries with mature generalised epidemics, including Uganda, there are groups that are more disproportionately affected (Most-At-Risk Populations [MARPs]) such as fishing communities, with high HIV infection, and extensive sexual networks bridging into the general population. Proven HIV prevention strategies are often not readily available to MARPs. Recent data from fishing communities in Uganda indicate that HIV infection rates in these communities are up to 5 times higher than the national average. In Uganda, approximately 10% of the population are engaged in fishing-related activities, but this group could be contributing a substantial proportion of HIV infections.

Proven HIV interventions are only partially effective, and need to be combined for effective HIV control. Combination prevention programmes are rights-based, evidence-informed and community-owned programmes that incorporate a mix of effective behavioural, biomedical and structural interventions, prioritised to meet the HIV prevention needs of individuals and communities, to have the greatest sustained impact on reducing new infections. We propose a trial on the effectiveness of implementing combination prevention in fishing communities in Uganda, to generate data that will complement that from other ongoing trials taking place elsewhere in Africa in general populations. The social context of MARPs is different from general populations and designing and evaluating an appropriate combination HIV prevention package in these core groups is critical to reduce HIV rates not only in these populations but also in the wider general population.

The trial will use a stepped-wedge cluster randomised design, in which each cluster (community) will receive the intervention in a phased manner during follow-up. This ensures that all communities will receive the intervention during the trial, whilst minimising logistical difficulties in implementing the intervention in several clusters simultaneously. Twelve fishing communities will be randomised into 4 equally-sized groups, and the time at which each group first receives the intervention is randomised. The study will take 60 months, with the first 3 communities receiving the intervention from month 18 onwards, and the last 3 communities receiving the intervention from month 42 onwards. During the baseline phase before the intervention is introduced, a census will be undertaken in each community from which a simple random sample of 240 adults (aged 18 years or old) will be recruited in each community and their HIV status determined. The incidence of HIV among participants who were HIV seronegative during the baseline survey will be determined during follow-up surveys. This is the primary outcome. Secondary outcomes include the impact on community viral load, behaviour changes, the level of uptake of the intervention, and the cost and cost-effectiveness of the intervention.

Given the burden of HIV, the "standard of care" package consisting of condom distribution, HIV risk counselling and testing (through routine government health units), and selected structural interventions targeting stigma and discrimination, harmful cultural norms and practices will be provided to all communities throughout the trial. We will select the most effective interventions, which, applied together, are most likely to have synergy and impact highly on HIV transmission. We will consider only "real life" evidence-based interventions that are affordable and scale-able within African prevention, care and support programmes.

The trial results will increase understanding of the implementation, uptake and impact of combination HIV prevention on HIV incidence among MARPS, and will aid policy makers in deciding on practical measures to scale up these interventions in similar populations in developing countries.

This is a five-year trial that aims estimate the effectiveness of a combined package of HIV interventions (biomedical, behavioural and structural) on HIV incidence in adults among most-at-risk populations in 12 fishing communities in Uganda. The trial will use a stepped-wedge community-randomised trial (CRT) design in which the unit of randomisation will be a community. Communities will be randomised into 4 equally-sized groups, and the time at which each group first receives the intervention is randomised. Primary outcome measure will be HIV incidence estimated from a random sample of 2880 adults tested for HIV status at baseline and over 6 periods of 8 months each. Secondary outcomes will be: i) Mean and total community viral load (CVL), ii) reported behaviour change iii) process indicators iv) cost per HIV infection, death and DALY averted, v) predicted HIV incidence and impact of single and combined intervention and cost effectiveness obtained from modelling of empirical data from the trial. HIV incidence will be estimated among HIV-negative participants at baseline and determined at follow-up serological surveys. The primary analysis will be at cluster level. For each period we will estimate the HIV incidence rate for each community. CVL and reported behaviour will be analysed using analogous cluster-level methods. The uptake of each intervention will be estimated restricting the analysis to communities in the periods in which they receive the interventions. Qualitative research and ethnographic techniques will be used to assess structural factors that facilitate HIV transmission. Cost per HIV infection, death and DALY averted will be estimated by combining the benefits of HIV prevention with the benefits from uptake of HIV treatment and care. We will use a mathematical model to assess the effect of a combination of interventions and hence determine the benefit of combined over singly applied interventions. This trial complements other ongoing trials in Africa.

Based on the far reaching impact of HIV combination prevention, it is anticipated that there will be a wide range of beneficiaries of the project including decision-makers in governments at national and sub-national levels, international bodies (UNAIDS, WHO) concerned with arresting the HIV-epidemic, bi-lateral research bodies (including DFID), national and international NGOs, academic research institutions, and independent organisations. The research team and academics will also benefit through shared skills and resources. In addition there is direct benefit to study participants the area who will benefit from the combination of prevention offered by the project.

Policy makers, advocates, and researchers from a range of backgrounds will benefit from our work through the development of an approach to combination prevention in most at risk populations. The research will facilitate multidisciplinary work allowing for more integrated innovative research. We will facilitate this by making published papers, policy briefs, harmonised data and resources available to the general public via a website. Data and resources shared through this project will be covered by a Data Exchange Agreement (DEA) and data sharing policies, protecting the copyright, ownership and data protection issues surrounding that data. Research findings will be presented at national and international conferences, which are open to a wide range of governmental agencies, NGOs, and academics. We will bring together key stakeholders in Uganda including representatives from the Ministry of Health, local leadership, NGOs, and academic institutes to set out the aims of the project, ensure that the added value of this work is explained, explore ideas for how the research might be used to serve users better, discuss progress and targets, and disseminate information and findings. Policy and practice implications will be presented at regional and national levels through tailored meetings and focused messages at several points in time, including face-to-face meetings with the policy contacts of the partner organisations. Findings will also be shared with bi-lateral and international bodies through meetings with new and existing contacts. There will be an end of project workshop to share findings with policy makers and other interested parties. Policy briefs will summarise in non-technical language the findings of the project and their implications for policy, and sent to relevant national and international policy makers.

The trial will also provide empirical data for mathematical modelling of the potential impact of combination HIV prevention interventions on the HIV epidemic in eastern Africa, including in MARPs. The cost-effectiveness analysis will provide valuable detailed information on the costs of the intervention compared to `standard of care'.

One of the project's key strengths is the research partners and established links with the policy and practice community in East Africa and internationally; this is a foundation from which the team will build to ensure findings are disseminated as widely as possible.