The BBC reports:
The United Nations has reduced its estimates of how many people are infected with HIV in 2007 from nearly 40m to 33m. Revised figures for India account for much of the decrease, experts say. But the rate of new cases and mortality levels are declining, although figures still show that there are 6,800 new cases each day and over 5,700 deaths.
Chicago’s Emily Oster, an economist, has previously claimed that UN estimates were too high – her own estimates come from working backwards from mortality rates in the cohorts of population you’d expect to be affected. Score one for Oster. (Her estimates are still much lower than the UN’s, as far as I can work out.)
Not long ago I wrote an article arguing that Emily Oster’s work would have been better had she managed to get the epidemiologists involved. I also mentioned that the work still had its merits; evidently so.