Paul Hutchinson
Jennifer Wheeler
Eva Silvestre
Dominique Meekers
Philip Anglewicz
John Hembling
Evan Cole
Joe Keating
Publication Date
April 14, 2014

Tulane University School of Public Health and Tropical Medicine

"...reviewers have called for more rigorous outcome evaluation designs in the manner conducted here."

Carried out in 2011 by the Tulane University School of Public Health and Tropical Medicine, this post-project evaluation and cost-effectiveness analysis of the Southern African Regional Social and Behaviour Change Communication Program in 8 countries of southern Africa finds that the effects of the behaviour change communication (BCC) interventions varied greatly depending on the nature of the media mechanism and the outcomes examined. "That effects were more frequently detected among the precursors to behaviour change - rather than actual behaviours - is in line with the majority of behaviour change theories."

As detailed here, the Regional Programme was designed to increase health awareness and facilitate social and behaviour change related to prevention and treatment of HIV and AIDS as well as in the areas of sexual and reproductive health and gender-based violence. Carried out in Malawi, Zambia, Zimbabwe, South Africa, Mozambique, Lesotho, Namibia, and Swaziland, it combined the mass media of the Soul City Institute for Health and Development Communication (IHDC) and its local affiliates with the community-based approaches (e.g., face-to-face interactions and community and social mobilisation) of the Community Media Trust (CMT) and the Southern Africa HIV and AIDS Dissemination Information Services (SAfAIDS). The principal component of each country's strategy had 3 primary objectives: 1) to increase risk perception of multiple and concurrent partners; 2) to alter social norms that are accepting of multiple and concurrent partnerships; and 3) to encourage HIV-risk related behaviour change that can decrease the likelihood of HIV transmission. A regional approach was developed in an effort to ensure consistent, coherent messaging within the context of high inter-regional mobility. The programme was funded by the Department for International Development (DFID), which also commissioned the external evaluation summarised here.

Researchers collected data on HIV/AIDS-related health behaviours, attitudes, risk reduction strategies, and exposure to behaviour change communication programmes using multi-stage cluster sample surveys of randomly selected adults aged 15-49 years in each of the 8 countries. Across the surveyed countries, sample sizes ranged from 3,000 to 12,000 respondents. In each of the countries, multivariate regression analyses tested for the effects of exposure to each of the programme partner's interventions on HIV/AIDS-related outcomes, while controlling for multiple sources of HIV/AIDS communication and non-random exposure.

An overall finding: "In all of the countries surveyed, multivariate regression analyses indicated positive effects of exposure to the programmatic activities of Soul City and its partners on many of the key behaviours (e.g., HIV testing, condom use, sexual partnerships). The effects were most evident for HIV testing and condom use behaviours. Radio and print media appeared to be the most effective in achieving behaviour change. There was little evidence to support the effects of OneLove media [the principal component of each country's strategy - see Related Summaries, below], regardless of type, on a key programmatic goal - reductions in multiple and concurrent partnerships. Across all countries, there appeared to be limited effects of exposure to program media on outcomes related to the exchange of gifts or money for sex, although attitudes towards such exchanges were less favourable among certain groups."

Selected organisation-specific findings [for more details about the communication activities and strategies, please consult the full document and/or see Related Summaries, below]:

  • Regarding Soul City: "Radio and print media for Soul City had consistently measurable effects across indicators of knowledge, attitude, norms and interpersonal communication but to a lesser extent on indicators of behaviour change....Print media was shown to have 'strong' evidence in support of effectiveness, with measurable effects detected in five...countries, while the evidence to support the effects on knowledge indicators from exposure to OneLove television was categorized as 'small.'...[T]here was 'strong' evidence of associations between exposure to OneLove radio programming and improved HIV attitudes [e.g., about sentiments toward people living with HIV/AIDS]. Exposure to OneLove radio media was consistently associated with greater frequency of discussions and other communication about HIV and sexual relations. The evidence in support of OneLove television programming on most indicators of attitudes, knowledge, norms and behaviours was 'small or limited.'"
  • Regarding SAfAIDS: "There is clear evidence that SAfAIDS programs [such as "cultural dialogues" seeking to engage community members and strengthen capacity - e.g., via pamphlets, toolkits, and training packs] have affected HIV norms and attitudes in the areas in which they have operated but have had much more limited impacts upon HIV-related risk behaviours. Overall exposure to SAfAIDS interventions varied from 8.0% of respondents in Mozambique to approximately 86% of respondents in Malawi. HIV risk perceptions tended to be greater among those exposed to SAfAIDS interventions. Further, there are clear indications that SAfAIDS has contributed to improving the skills and resources of communities to respond to the HIV epidemic. Respondents exposed to SAfAIDS interventions were more likely to report that people in the community are joining together to help people with HIV, that people in the community are discussing multiple partnerships and the risk of HIV, and that leaders are speaking out against the risk of HIV from having multiple partners. In nearly all countries, there was limited evidence that individuals who were exposed to SAfAIDS interventions were more or less likely to have experienced physical or sexual violence. The evidence surrounding reporting of violence to authorities, friends or family was also mixed, though increased reporting was apparent in a number of countries."
  • Regarding CMT: "Exposure to CMT interventions varied by country, from 7.8% of respondents who reported exposure to any CMT interventions in Lesotho, where Beat It! [a weekly television programme also involving a DVD series transmitted via mobile audio-vidual kits and acommpanying manuals] was accessible principally through treatment literacy sessions, to 25.1% in Malawi. Exposure to CMT interventions had clear effects on indicators related to HIV knowledge and treatment literacy, as well as on the behavioural indicators related to testing and treatment. Respondents exposed to CMT activities perceived themselves to be at higher risk for HIV than unexposed respondents, a likely reflection of the fact that CMT treatment literacy programs are targeted towards HIV positive individuals. CMT programs also had notable effects on norms and attitudes surrounding HIV risk behaviours, particularly with respect to the acceptability of negotiating condom use among partners."

In short, "the three years of the Regional Programme may have been insufficient to have achieved broad behaviour change, though the research design cannot tell us whether longer periods of exposure would have yielded more statistically significant effects or larger effect sizes."

Click here for the 22-page report in PDF format.
Click here for the 2-page "DFID Management response to the external evaluation of the Southern Africa Regional Social and Behaviour Change Communication Programme."


UK Government website, August 26 2014. Image credit: Mohan Dutta, Culture-Centered Approach blog