[oblomov]

Why should anyone assume credibility for this type of observational study? It’s just based on data with no experimental design.

[FredZarguna]

It's amazing how many people -- including scientists -- accept these kinds of "studies" at face value without asking whether the two groups were even true cohorts. Or that even if they were, that the raw number of deaths was the same is in and of itself meaningful. Was the mean age of death the same in both groups? What about the number of survivors?

Too many people doing "statistical analysis" without the slightest understanding of what statistical science is or does.

[rlmorel]

I actually polluted my browser to read the article:

"..."However, our findings are quite tentative for mortality because the data are very noisy," he added. "We feel that our study raises important questions about the benefits of mammography screening, but it certainly does not answer them."

More research is needed, he said.

The researchers also warn in their paper that their findings may be limited by so-called ecological bias, which can occur when assumptions are made about individuals from data of a large group.

Uh-huh. Noisy data. They don't know key things about the population. What a stinking piece of garbage. I wonder who paid for that study.