Whoa. The research is interesting, but the NYT blog post has some red herrings: the idea some women might believe that the mammogram prevents cancer for one.

You can look at the paper. First, they do not have a dataset that allows them precise estimates -- they use some dataset (DevCan) that allows them to estimate the 10 year risk of developing cancer for different age brackets, and they use a different source (National Health Interview Survey of 2003) to estimate the proportion of cancers detected by screening. Then then the authors make the leap:

"The risk of having screen-detected cancer was estimated simply as the product of the risk of developing breast cancer and the proportion of breast cancers found by mammography."

In other words, having mammographies is assumed a completely independent event from the risk of developing breast cancer. Really? I would believe that if someone knows their family has a history of breast cancer, they are more likely to be diligent about mammographies... Anyway, the paper is a good collection of informed guesses, but I would not consider it the definitive word on mammographies.

It was interesting to read the comments to the NYT blog entry and how some people believe they know how to prevent cancer (as in 'good diet, exercise, not smoking'.... it all might help, but cancer hits fit, health-conscious people, too).