- Design
- systematic review and fixed-effect meta-analysis of randomised controlled trials, GRADE-assessed
- Population
- 721,607 participants aged 45 to 80 across 5 trials, with 11 to 23 years of follow-up
- Primary outcome
- prostate cancer-specific mortality at longest available follow-up
- Effect
- reduced prostate cancer-specific mortality at high certainty (p<0.001); the pooled incidence rate ratio is not given in the abstract
Five randomised trials covering 721,607 participants aged 45 to 80, with follow-up of 11 to 23 years, were pooled with risk of bias assessed by ROBUST-RCT and certainty graded by GRADE. The primary outcome was prostate cancer-specific mortality at the longest available follow-up.
PSA-based screening reduced prostate cancer-specific mortality, with high certainty of evidence (p<0.001). The abstract does not give the pooled incidence rate ratio, so the size of the reduction is not quotable from it — what is reported is the direction, the statistical significance and the certainty grade. Secondary analyses suggested the relative reduction grew with longer follow-up, and the authors note that differing screening protocols and contamination of control arms would have pushed the estimate towards no effect, meaning the true benefit is likely larger than measured.
This contrasts with earlier syntheses, which is the point of the paper. For twenty years the honest position has been that the mortality benefit of PSA screening was uncertain and the harms were not. On extended follow-up the benefit is no longer the uncertain half.
The authors then do something unusual and worth respecting: they refuse to draw the policy conclusion. A mortality benefit does not establish that population screening is justified, nor that a well-informed man would choose it. Overdiagnosis, biopsy morbidity and the treatment of cancers that would never have surfaced are unchanged by this analysis. For the Indian clinician, where opportunistic PSA testing is common and structured shared decision-making is rare, the practical implication is about the quality of the conversation rather than about ordering more tests.
- Update the conversation: the mortality benefit is now high-certainty evidence, the harms are unchanged
- Frame the discussion around overdiagnosis and biopsy morbidity, which this analysis does not address
- Do not convert this into unselected population screening; the authors explicitly decline that step
- Take life expectancy seriously — benefit accrued over 11 to 23 years, which excludes many men who get tested
- Where you do test, have the pathway for a raised result agreed in advance: repeat, magnetic resonance imaging, then targeted biopsy
Why it matters
The uncertain half of the PSA debate has moved, which changes what an honest consultation has to cover.
Don't overread it
A demonstrated mortality benefit is not an endorsement of population screening — overdiagnosis and treatment harms are untouched by this analysis.
The statistics, in plain English
High certainty on GRADE is a strong statement: it means the reviewers judged further research unlikely to change the direction of the estimate. Note what is absent — the abstract reports significance without the pooled rate ratio, so the magnitude cannot be quoted, and magnitude is what a shared decision needs. Control-arm contamination, where men in the no-screening group get tested anyway, biases a screening trial towards finding no difference, so it strengthens rather than weakens a positive result. The growing relative reduction with longer follow-up is consistent with a slow-growing cancer, and it also means short-horizon patients gain least.
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.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for urology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free