Is a PSA test worth it? The trials show a small benefit and real harms

Whether a PSA test is worth having has no single answer — it depends on a man's age and how he weighs a small chance of avoiding a prostate-cancer death against a real chance of being diagnosed and treated for a cancer that would never have harmed him [s1][s3]. The largest European trial found screening does cut prostate-cancer deaths; a large UK trial of a single test found no benefit; and US guidance now frames the decision as a personal one, best made with a doctor [s1][s2][s3].

The trial that found a benefit

The European Randomised Study of Screening for Prostate Cancer (ERSPC) followed men in a core age group of 55 to 69 offered regular PSA testing versus no screening [s1]. At 13 years, the rate of death from prostate cancer was 21% lower in the screened group (rate ratio 0.79) [s1]. But the same trial exposes the cost. To prevent one prostate-cancer death, 781 men had to be invited for screening and 27 extra cancers detected — many of which would never have caused symptoms [s1]. The rate ratio of prostate-cancer diagnoses was 1.57, meaning screening found far more cancer than no screening did, much of it overdiagnosis [s1].

The trial that found none

A single test tells a different story. The UK's Cluster Randomised Trial of PSA Testing (CAP) invited 419,582 men aged 50 to 69 to a one-off PSA test or to standard care [s2]. After a median of 10 years, prostate-cancer deaths were essentially identical: 0.30 per 1,000 person-years in the invited group versus 0.31 in the control group, with no significant difference [s2]. The single screen did, however, pick up more low-risk cancer — the diagnosis rate rose to 4.3% from 3.6%, driven by slow-growing tumours unlikely to shorten life [s2]. All-cause mortality was likewise no different between the groups, and an analysis restricted to the men who actually attended still found no survival advantage [s2]. The lesson is that one PSA test does not save lives at a population level; any benefit in ERSPC came from repeated testing over years, and it came bundled with overdiagnosis.

What the harms actually are

Overdiagnosis matters because prostate-cancer treatment is not harmless. The US Preventive Services Task Force notes that about 1 in 5 men who have their prostate removed develop long-term urinary incontinence, and 2 in 3 will have long-term erectile dysfunction [s3]. Set against that, screening men aged 55 to 69 may prevent about 1.3 prostate-cancer deaths and about 3 cases of metastatic disease per 1,000 men screened over roughly 13 years [s3]. Those are the two magnitudes a man is really trading between.

What the guidance now says

The Task Force translates this into age-specific advice [s3]. For men aged 55 to 69, PSA screening is a C recommendation: the decision should be an individual one, made after a discussion of benefits and harms, and clinicians should not screen men who do not want it [s3]. For men aged 70 and older, it is a D recommendation — screening is advised against, because the harms clearly outweigh the benefits at that age [s3]. Men at higher baseline risk — African American men and those with a father or brother diagnosed with prostate cancer — have more at stake on both sides of the ledger and are exactly the men for whom the conversation matters most [s3].

What it means for a reader

A PSA test is neither the life-saver of screening campaigns nor the useless trap of its critics. For a man in his late 50s or 60s who, told the trade-off, would rather accept the risk of overtreatment to lower his chance of a prostate-cancer death, testing is reasonable [s1][s3]. For a man who would be more troubled by a biopsy, an operation and its sexual and urinary side effects than reassured by the small mortality benefit, declining is equally reasonable [s3]. What the evidence rules out is the middle position sold by pop-up screening drives: that a single test is a simple, harm-free good [s2]. It is a decision, not a formality — and after 70, the guidance is clear that the balance has tipped against it [s3].