This article is a rather optimistic spin on these numbers. I'd have said "a large majority of cancer cases--75% in women and 67% in men--were not preventable"
On the other hand, when someone tells me that they've optimized a compiler and it can now compile 3% faster I get really excited and pull their repo, I don't turn around and say "you mean 97% as slow" and keep complaining about how long compilation takes and how nothing can be done about it. Show that there are steps you can take to cut it by 33% more than we currently know how and you'll be a god among programmers.
Sure, not every visitor signs up still, but some group of people found the new copy compelling enough to sign up - with no new, very expensive feature development.
> Lung cancer is by far the leading cause of cancer death among both men and women; about 1 out of 4 cancer deaths are from lung cancer.
http://www.cancer.org/cancer/lungcancer-non-smallcell/detail...
and from the article:
> About 82 percent of women and 78 percent of men who got lung cancer might have prevented it through healthy behaviors.
Back of the envelope: ~20 percent of all cancer deaths are lung cancer deaths that wouldn't have happened without smoking. That leaves just 5-15% of cancer deaths being "potentially preventable" once you've given up smoking. (Here, I'm not distinguishing between cancer cases and cancer deaths; please point to better data if you have it.)
I'd wager less than 5% of cancer deaths are preventable if you neither smoke nor are overweight.
I think this article, and especially its headline, is highly misleading.