South Africa’s life expectancy has plummeted due to HIV/AIDS, but the UN has just released a report claiming that HIV infections have dropped by 17% in sub-Saharan Africa.

Yes, I know they are not the same places, nor are they the same measures, but they are saying diametrically opposed things. Things are getting better, or things are getting worse.

The UN report (and these are my cursory observations, I haven’t had a chance to read through the entire report yet) claims that in addition to the natural trajectory of the epidemic, programs are making a difference. But it also highlights the fact that much better work needs to be done in order to figure out what is working, where and why.

There’s also the question of measurement. Anyone familiar with Pisani’s ‘Wisdom of Whores’ is familiar with the number cooking that occurs in the HIV/AIDS world. It’s not necessarily a bad thing – we need to extrapolate in order to assess – but it means we often make mistakes. Case in point was the UN’s move in 2007 to revise downwards the number of global AIDS cases by 6 million. Poof! Success through statistics!

I’m not judging – nothing gets done with out estimates – and kudos for the revision. All I’m saying is that a 17% reduction of new infections may very well be the result of better measuring – and this is a very good thing, but for very different reasons.

Measurement is, I think the key problem to a lot of the HIV puzzle. Life-expectancy can go down in South Africa due to deaths from AIDS while new infections go down. They are different sides to the problem, both important, but necessarily and conceptually distinct. Getting someone tested, getting them to use a condom or to stop engaging in multiple partnerships is a much different problem than getting ARVs to people and making sure they take them consistently.

End of the day – this blog post should have been written after I read the report, but I think that sometimes there’s value to first impressions – especially if they coincide with a free 20 minutes…