Map of life expectancy at birth from Global Education Project.

Friday, April 01, 2011

I told you so . . .

In a population based randomized controlled trial, "After 20 years of follow-up the rate of death from prostate cancer did not differ significantly between men in the screening group and those in the control group."

If you read my previous post, you know that "significantly" doesn't mean "significantly." The 95% confidence intervals for the screened and unscreened groups overlapped, but actually the most likely interpretation is that screening did result in a slightly lower risk of ultimately dying from prostate cancer, (although of course you'll still die of something!) "To prevent one death from cancer, 1,410 men would need to be screened and 48 treated." That means 47 men who got no benefit but who underwent surgery and/or radiation therapy and risked incontinence, erectile dysfunction, and bowel symptoms. And maybe even that single beneficiary who got a slightly longer lifespan is fictitious.

No thanks.

No comments: