From the American Cancer Society article this is derived from:"Finding prostate cancer earlier through PSA testing and advances in treatment helped lower the death rate for prostate cancer by about 4% each year from the mid-1990s until about 2013. But more recently (2013-2018), the death rate is no longer dropping."
And of course there's the history of the USPSTF recommendations for, and against, PSA screening
. From the article: "The USPSTF then updated its 2008 grade for PSA, “I” for insufficient evidence, to “D”, for recommending against PSA, in 2012. Through changes in PSA usage, biopsy, and incidence patterns, the grade shifted to “C” in 2018, for recommending shared decision-making. In the future, risk-stratified screening could be the solution for mitigating the harms of PSA screening without losing its benefits."
2008-2012 = "insufficient evidence"
2012-2018 = "recommend against"
2018-2021 = "shared decision making"
Prior to 2008, PSA screening was a common practice. In 2008, the USPSTF proclaimed "Given the uncertainties and controversy surrounding prostate cancer screening in men younger than age 75 years, a clinician should not order the PSA test without first discussing with the patient the potential but uncertain benefits and the known harms of prostate cancer screening and treatment. Men should be informed of the gaps in the evidence and should be assisted in considering their personal preferences before deciding whether to be tested."
Given the lag time for prostate cancer to develop from an early PSA-only diagnosis (and likely treatable), to waiting for symptoms to develop in late stage prostate cancer (less curable, or not curable), it's not really surprising that the reduced screening recommendations of the USPSTF have led directly, and predictably, to increased death rates from prostate cancer.
Granted, there are also no doubt far fewer men living with life-changing side effects of being treated for prostate cancers that never would have harmed them, detected only through broad PSA screening efforts.
How should we prioritize competitive harms like this? Hard to say.
This is important to me, personally. My specific case was affected by this. In short, I skipped PSA screening from 2008, age 50 (last PSA was under 2.0), until 2012, age 55 (PSA was 4.1, then 5.2 six months later, then dx with G9, stage cT3a). I had the PSA in 2012 as part of a routine physical; being 55 it was "kind of ok" to check it again.
So the avoidance of screening during those four or five years may have let my case develop from contained and "curable" to a much less promising situation. My urologic oncologist certainly thought so.