Article Text

Download PDFPDF

P6.069 Social Mobilization and Peer-Mentoring to Encourage Voluntary Counseling and Testing and Post-Testing Behavior: A Randomized Experiment in Senegal
  1. J Graff Zivin1,
  2. V Orozco2,
  3. H Thirumurthy3,
  4. C Sakho4,
  5. P Diallo5,
  6. M Offer Westort2
  1. 1University of California - San Diego, San Diego, CA, United States
  2. 2World Bank, Washington, DC, United States
  3. 3University of North Carolina - Chapel Hill, Chapel Hill, NC, United States
  4. 4Ministry of Health, Dakar, Senegal
  5. 5National AIDS Commission, Dakar, Senegal


Background To evaluate the effectiveness of two HIV/AIDS sensitization campaigns of the community-based organisations (CBOs) it financed, the National AIDS Commission of Senegal randomly assigned its financial support across the country’s 14 regions, establishing three groups of health districts. Health districts where NAC-funded CBOs carried out i) mass social mobilisation communication campaigns; ii) campaigns using peer-mentoring; and iii) control health districts, where CBOs didn’t receive NAC support.

Methods We exploit the initial random assignment and differences at the time of implementation for the intervention groups (the randomization plan was not followed in the control group so we remove it from the analysis). We use two methods. First, a before-after analysis where we examine trends to estimate the change in outcomes during each of the four quarters following implementation of the intervention, relative to the average of the four quarters before the intervention. Second, we estimate a difference-in-differences regression model to directly compare the effectiveness of each intervention.

Results The before-after analysis shows that immediately after the intervention was completed, peer-mentoring doubled the number of adults who were tested for HIV and the number of adults who picked up their test results. For both outcomes, the number tripled three quarters later. Social mobilisation was less effective over time. For those individuals who test HIV-positive, peer-mentoring induces more HIV-infected individuals to collect their results and engage in post-test counselling. However, the second method finds no significant differences between the two intervention groups.

Conclusion These results suggest that both interventions are successful for the general population and to a lesser for the HIV-infected population. The difference-in-differences analysis suggests that there is little reason to favour one intervention over the other. Analysis at the health district level greatly diminishes our power to detect changes, especially for social mobilisation activities.

  • Community based organization
  • impact evaluation
  • VCT

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.