Emerging Issues in Today's HIV Response: Debate 2 - Behavior Change for HIV Prevention

This Executive Summary document describes a debate co-hosted on June 29 2010 by the World Bank and the United States Agency for International Development (USAID) on the following proposition: behaviour change in generalised epidemics has not reduced new HIV infections and is an unwise use of HIV prevention resources. Using video conferencing and web-based technologies allowed country teams in Africa and other partners from across the globe to participate in real time in the debate, which took place in Washington, DC, US, and is available as a videocast, as well as in summary form, on the World Bank website.

Because, according to the debate host organisations, there is a scarcity of research evidence for behavioural prevention efforts, "questions about these efforts are being asked with increased intensity. Do behavioral prevention efforts work? If they do work, what kinds of interventions work best? How many new infections could behavioral interventions avert?"

The debate was framed by the question of sustainability of behaviour changes and the possible ephemeral nature of these changes over the long term. "The panelists who spoke in favor of the proposition argued that there have been no rigorous evaluations or convincing data to show that behavior change interventions are effective in promoting and sustaining positive behaviors. They contended that success in reducing HIV incidence can come only from broad policy changes and biomedical solutions. They also asserted that behavior change programs have not created exportable models from the success stories claimed in countries such as Uganda, where behavioral interventions are said to have reduced HIV prevalence. Furthermore, declines in HIV prevalence attributed to behavior change programming could be attributed to other causes, such as the natural course of the disease, behavior change without interventions (e.g. attending funerals every week could spark behavior change), or other contextual factors that may only be understood many years after the fact.

The panelists who spoke against the proposition contended that behavioral programs have been effective in diverse locations and in different stages of the epidemic, as shown in the significant decreases in population level HIV incidence in countries such as Australia, Cambodia, Kenya, Malawi, Thailand, Uganda, and the United Kingdom. They highlighted six reasons as to why there should be more behavior change interventions:

1. behavior change is at the crux of the epidemic, especially with those who engage in multiple partnerships;
2. behavior change has worked in a number of other arenas, and it stands to reason it should work in HIV;
3. risky sexual behavior has declined in a number of places;
4. declines in transmission are partly the result of programmatic efforts;
5. behavior change permeates everything one wants to do in prevention, from condoms to adherence, alcohol risk, cross-generational sex, demand for services, and gender-based violence; and
6. public health professionals have a duty to warn populations of a clear and present danger.

The panelists argued that behavior change is central to any intervention, even those that are primarily biomedical." As part of the argument on decreases in certain countries, the following was noted: "In countries where population-level HIV prevention has worked, the panelist listed three elements that have been shown to effectively reduce HIV incidence in different situations: 1) intense, focused, and systematic national communications on AIDS by politicians and in all media; 2) engagement with social networks that work systematically through groups at risk, village meetings, local leaders, and care networks to get prevention on the agenda; and 3) AIDS has to be an official health fact that is diagnosed and notated on death certificates and at funerals to allow community and care networks to form".

For example [examples excerpted by the editor because they include communication-related information - for the rebuttal to these arguments, please see the original document linked below]:

• "In the early stages of the epidemic, where HIV was seen as a threat in the United Kingdom, individual services, clinics, and testing were not slowly scaled up as has been recommended to Africa. Rather, population-level HIV prevention was added directly and intensely. HIV prevention was directly communicated by political leaders and with blanket media coverage on radio and television. HIV interventions worked systematically through social networks, with outreach to groups at risk and in schools, and every household in the country was sent a letter about HIV. Within six months of the onset of the program, sexually transmitted infections (S.T.I.s) declined by around 50 percent. Some argued that the decline would be short-lived, but it took 15 years for the return to pre-intervention levels.
• ...In Thailand, [t]he government again implemented intense national communication: AIDS was on all 70 radio stations and television stations for 90 seconds every hour, every day, every week, every month. Secondly, the government systematically engaged the population through social networks including brothel owners, sex workers, taxi drivers, and the workforce to disseminate messages about condom use. The intervention resulted in an increase in condom use and a 60 percent decline in men frequenting sex workers
• In the later stages of the epidemic, HIV prevention behaviors have also worked in Kenya and Malawi. In the first decade of the epidemic in Kenya, services had been delivered but HIV had not declined. In 1999, the Kenyan parliament requested to know why HIV prevention was not working, and they organized the first ever emergency parliamentary session on AIDS. The parliament declared AIDS a national emergency and required all politicians to return to their communities each weekend with materials about AIDS. The topic of AIDS was brought into the school curriculum, and STI treatment was widened. The late addition of an integrated community response resulted in behavior changes and a reduction in HIV incidence."

The question and answer session addressed whether structural and behavioural interventions are different; whether interventions could be successful without changes in behaviour, cultural, or gender norms; and how to reach most-at-risk populations (M.A.R.P.s) with behavioural interventions. Panelists were also asked to describe an ideal behaviour change intervention. The debate concluded with a call for additional research "to demonstrate the effects of behavioral interventions, challenged views on the role of behavior in interventions and the efficacy of resources for behavior change, and proposed possible models for effective interventions based on situations where HIV prevalence had declined."