Number Needed to Treat with PSA Screening
The number needed to treat (NNT) to prevent one prostate cancer death with PSA screening is approximately 5 men, based on modeling of the highest quality European screening trial data. 1
Most Recent and Highest Quality Evidence
The 18-year follow-up of the Göteborg trial (2015) provides the most mature data, showing that 13 men needed to be diagnosed with prostate cancer to prevent one prostate cancer death. 1 This represents the number needed to diagnose (NND), which is distinct from but related to the NNT.
For the broader ERSPC trial at 13-year follow-up, 27 additional prostate cancers needed to be detected to prevent one prostate cancer death. 1
Understanding the Different "Numbers Needed"
The evidence presents three distinct metrics that are often confused:
Number Needed to Screen (NNS): The Göteborg 18-year data shows 139 men needed to be invited for screening to prevent one prostate cancer death. 1
Number Needed to Diagnose (NND): 13 men needed to be diagnosed with prostate cancer to prevent one death (Göteborg 18-year). 1
Number Needed to Treat (NNT): Modeling of ERSPC data estimates 5 men with screen-detected prostate cancer need treatment to prevent one prostate cancer death. 1
Evolution of NNT Over Time
A critical pitfall is that NNT decreases substantially with longer follow-up. 2 At 9 years of ERSPC follow-up, the original NNT was 48, but modeling suggests this decreases to approximately 18 by year 12 as the mortality benefit accumulates. 2 The most refined modeling estimates settle on NNT of 5 when accounting for compliance and longer follow-up. 1
Critical Context: Mortality Outcomes
PSA screening does NOT reduce all-cause mortality. 3, 4 The ERSPC trial showed identical all-cause mortality in screened versus control groups (RR 1.00,95% CI 0.98-1.02) after 11 years. 3 Even the most recent 15-year CAP trial follow-up showed no all-cause mortality benefit (RR 0.97,95% CI 0.94-1.01). 5
The prostate cancer-specific mortality reduction is small in absolute terms: approximately 0.71 deaths prevented per 1000 men screened over 9-11 years. 6 At 15 years, the CAP trial showed a reduction from 0.78% to 0.69% prostate cancer mortality—an absolute difference of only 0.09%. 5
Treatment Harms That Offset Benefits
For every prostate cancer death prevented, 200-300 men experience permanent urinary incontinence or erectile dysfunction from treatment. 3, 4 Specifically:
- 1 in 5 men (20%) undergoing radical prostatectomy develop long-term urinary incontinence 4
- 2 in 3 men (67%) experience long-term erectile dysfunction 4
- Up to 50% of screen-detected cancers represent overdiagnosis 3
Clinical Algorithm for Counseling
When discussing PSA screening with patients aged 55-69 years:
State clearly: Screening may prevent approximately 1 prostate cancer death per 1000 men screened over 10-13 years 4, 6
Emphasize: This does NOT translate to living longer overall—all-cause mortality is unchanged 3, 4, 5
Quantify treatment burden: Of men diagnosed through screening, approximately 5 will need treatment to prevent 1 prostate cancer death, but 200-300 will experience permanent sexual or urinary dysfunction 3, 4
Explain overdiagnosis: Up to half of screen-detected cancers would never have caused symptoms or death 3
Common Pitfalls to Avoid
Do not conflate NNT with NNS. 1, 2 The number needed to screen (139-1410 depending on follow-up duration) is vastly different from the number needed to treat (5) among those diagnosed. 1, 2
Do not assume younger or healthier men benefit more. 3 Subgroup analyses showed no effect of screening on all-cause mortality regardless of age at screening. 3
Do not ignore the 10-15 year time horizon required for any benefit. 3 Men with life expectancy less than 10-15 years cannot benefit from screening and should not be screened. 3, 4
Do not present only the prostate cancer-specific mortality benefit without discussing that overall survival is unchanged. 3, 4 This violates shared decision-making standards and misrepresents the evidence. 3