Subscription Center
Aug 4, 2008 | 1
The U.S. Centers for Disease Control and Prevention (CDC) missed more than 16,000 new HIV infections in its last tally of the disease, according to a new analysis.
The CDC reported in the Journal of the American Medical Association yesterday that its 2006 estimate of new HIV infections in the U.S. was 40 percent off: 56,000 contracted the human immunodeficiency virus—the virus that leads to full-blown AIDS—that year, not 40,000.
Here's why they got it wrong: First, previous estimates relied on extrapolating from limited data taken from small studies on HIV incidence in high-risk populations, like men who have sex with men. Also, the earlier estimates did not take advantage of a new HIV test (called STARHS or Serological Testing Algorithm for Recent HIV Seroconversion), which can tell whether a person has been infected within the last six months. Beforehand, researchers only knew that someone had been diagnosed with HIV, not when someone contracted the virus—and many people diagnosed with HIV contracted it several months or even years prior.
See what we're tweeting about
BoraZ SciAm blogger asks: Who is Greatest American Physicist ever (and does he/she stand out at all in world ranks?) http://t.co/tyjETTG317
DNLee5 You housed 5 unmarked animals from 2 different litters (parents) in the same cage. No, that's not a problem at all. >_>
EricROlson 10 Tips on Writing from David Ogilvy http://t.co/zr2TYcJNLH via @brainpicker
Deadline: Jul 30 2013
Reward: $100,000 USD
The Seeker desires a method for producing pseudoephedrine products in such a way that it will be extremely difficult for clandestine che
Deadline: Jul 15 2013
Reward: $5,000 USD
SciBX: Science-Business eXchange, a joint publication from the makers
Powered By: 
YES! Send me a free issue of Scientific American with no obligation to continue the subscription. If I like it, I will be billed for the one-year subscription.
