[kazinator]

If we decrease premature deaths from any cause, particularly a common cause of premature deaths, we increase life expectancy. Or just what do the authors think life expectancy means? It's an average. Every premature death drags down average life expectancy.

If we say that cancer screening has no effect of life expectancy, that's exactly the same as saying that it doesn't prevent premature deaths.

If cancer screening effectively prevents premature deaths, but the effect on the population's life expectancy is small, that effect is the wrong thing to be focusing on, potentially resulting in a harmful takeaway message.

[LorenPechtel]

No, because we might be replacing one cause with another.

That's what they've found with PSA--it kills (via treating things that wouldn't actually have killed the patient) as many as it saves.

[kazinator]

> That's what they've found with PSA--it kills (via treating things ...)

I'm not sure what PSA you're referring to. The Prostate-Specific Antigen PSA is something secreted by the body, and not a treatment, so it can't be that one.

Are you referring to overdiagnosis (via PSA or any other screening), resulting in overtreatment?

[LorenPechtel]

PSA is also commonly used to refer to the test that measures the PSA level.

And, yes, it's treatment of things that weren't actually going to kill the patient. It's been known for quite a while that doing the test for screening produces no increase in life expectancy.

What this study is saying is that the same problem seems to apply to most other cancer screenings.

[vc8f6vVV]

You may be decreasing premature deaths from cancer but increasing premature deaths from other causes related to screening, thus you don't increase life expectancy. Read the original article in JAMA.