Updated Cochrane review: PSA screening cuts prostate-cancer death, but not clearly overall mortality

Design

Updated Cochrane systematic review and meta-analysis of randomised trials; GRADE-rated

Population

Six randomised trials, 789,086 men, follow-up 3.2–23 years

Primary outcome

Prostate-cancer-specific and overall mortality

Effect

Prostate-cancer mortality risk ratio 0.87 (0.80–0.95); overall mortality 0.99 (0.97–1.00)

The refreshed Cochrane review of prostate cancer screening pooled six randomised trials and 789,086 men, with follow-up to 23 years. It is the evidence to anchor the screening conversation.

Screening likely reduces prostate-cancer-specific mortality — a risk ratio of about 0.87, roughly two fewer prostate-cancer deaths per 1,000 men screened. But overall mortality barely moved (risk ratio 0.99, confidence interval including no effect), and the interpretation is highly sensitive to what counts as a meaningful difference. Harms, measured here as intervention-related mortality, showed little difference. A newer strategy combining PSA with a kallikrein panel and MRI (ProScreen) changed detection but has no mature mortality data.

The practical message is not for or against screening but how to frame it: an informed, preference-sensitive decision. A man should hear that PSA screening modestly lowers the chance of dying from prostate cancer, has not been shown to help him live longer overall, and carries the downstream harms of biopsy and overdiagnosis — then decide with that in front of him.

• Screening likely reduced prostate-cancer-specific mortality (risk ratio 0.87, 95% CI 0.80–0.95) — about 2 fewer deaths per 1,000.
• Overall mortality was barely changed (risk ratio 0.99, 95% CI 0.97–1.00), with the interval including no effect.
• Interpretation is highly sensitive to the minimal clinically important difference chosen.
• Frame PSA screening as a shared, preference-sensitive decision, naming both the modest benefit and the harms of overdiagnosis.

Why it matters

It gives the hard numbers a man needs to weigh the modest cancer-specific benefit against the harms before choosing to be screened.

Don't overread it

Much of the certainty rests on a sensitivity analysis of one trial; this synthesis informs the screening conversation rather than mandating or forbidding screening.

The statistics, in plain English

A risk ratio of 0.87 for prostate-cancer death is a real but small absolute benefit (about 2 per 1,000); the overall-mortality risk ratio of 0.99 with a confidence interval touching 1.0 means no reliable effect on dying from any cause. Which matters more is a value judgement for the patient.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.