I complained earlier this week about Reuter's coverage of a new HIV study in Zimbabwe. IRIN today published what the story SHOULD look like. In fact, they make a pretty solid study in the difference between bad reporting on HIV and good reporting on HIV.
Content-wise, the IRIN story reveals that things really do look very encouraging in Zimbabwe. It appears that reduction in sexual partners has driven the decrease in infection rates. It's also, as the article points out, a very interesting case, because this success contrasts with lack of or limited success of well-funded prevention campaigns in neighbouring South Africa and Botswana (Full disclosure: I worked on one of these campaigns for several years). It would be fascinating to see some speculation by people with expertise in this field about why. Off the top of my (non-expert) head, I'm struck by the fact that the behaviour change happened early - late 90s according to this study, that mortality was probably higher in Zim than in neighbouring countries and that economic meltdown seems to have played a role.
More research definitely a needed. Also, if the change came from the communities, from the people, rather than from some carefully though-out, evidence-based BCC campaign, this sounds like a strong example of positive deviance that should be supported with research and ongoing attempts to understand how they did it.