NBC opened the Nightly News last night with a story about how screening heavy smokers for lung cancer can reduce lung-cancer deaths. Specifically, when heavy smokers and people who used to be heavy smokers were screened using spiral (helical) CT scans, the death rate from lung cancer was lowered by 20%. (I prefer the old name CAT scan, because it lends itself to so many geeky jokes, such as that when you generalize the principle of the CAT scan, you get the PET scan. And the whole idea of a spiral scan makes one think about there being more than one way to scan a ...) CBS and ABC also gave prominent coverage the the story.
No doubt, this is good news. But every journalist should read (among others) my first substantive post on this blog, in which I mentioned the importance of long division in policy analysis. There were 53,000 people in the study. Over eight years, 442 people in the group getting chest X-rays died of lung cancer, versus 354 in the group getting spiral CT scans.
Let's assume equal numbers in each group. Then if you're a heavy smoker, your chance of dying of lung cancer over eight years is 1.67% with the chest X-ray, and 1.34% with the CT scan. By getting the CT scan you reduce your chance of dying of lung cancer over the next eight years by 0.33%. When you consider that the CT scan has a higher dose of X-rays, and that information about deaths from other causes won't be published for several months, it would seem wise to hold off on the stampede for now. At least the health care debate is over, sort of, so we'll probably be spared Republicans saying that Democrats want to ration CT scans.
By the way, does anyone else find these numbers (remember, heavy smokers and former heavy smokers) surprisingly low? I guess I was so heavily propagandized in elementary school that I thought your odds were about like playing Russian roulette. Of course this still represents hundreds of thousands of people, and doesn't include other cancers, heart disease, emphysema, and so on.
No comments:
Post a Comment