The trial that showed PSA screening does stop some men dying of prostate cancer, and showed at the same time how many men have to be tested, and how many treated for a cancer that was never going to harm them, to achieve it.
ERSPC identified 182,000 men aged 50 to 74 through population registries in seven European countries in the early 1990s and randomised them to an offer of prostate-specific antigen screening, on average once every four years, or to a control group that was not offered it. The predefined core age group was 162,243 men aged 55 to 69. Eighty-two percent of the men offered screening accepted at least one round.
At a median of nine years the cumulative incidence of prostate cancer was 8.2 percent in the screened group against 4.8 percent in the control group, and the rate ratio for death from prostate cancer was 0.80. The absolute difference was 0.71 deaths per 1,000 men, which the authors expressed as 1,410 men needing to be screened and 48 additional cancers needing to be treated to prevent one death.
The 16-year report, on 182,160 men with a core group of 162,389, found the same rate ratio with a tighter interval and a larger absolute benefit: the absolute mortality difference grew from 0.14 percent at 13 years to 0.18 percent at 16, the number needed to invite fell from 742 to 570 and the number needed to diagnose from 26 to 18. Those two sets of figures use different denominators, men invited against men screened, which is the commonest way this trial is misquoted.
Numbers are from the trial as recorded here; see the source links in the table below. This is orientation, not medical advice: ask your team how closely the trial population matches you.
Absolute risk difference 0.71 deaths per 1,000 men; 1,410 needed to screen and 48 extra cancers treated to prevent one death. Among men who attended the first round the rate ratio was 0.73 (0.56 to 0.90).
SourceNumber needed to invite 570, number needed to diagnose 18, against 742 and 26 at 13 years; absolute mortality difference 0.18 percent.
Source| Endpoint | Arm | n | Value | HR (95% CI) | p | Source |
|---|---|---|---|---|---|---|
| Death from prostate cancer, core age group 55 to 69, median 9 yearsprimary | Offered PSA screening | - | 0.8 rate ratio | - | 0.04 | link |
| Death from prostate cancer, core age group, 16 years | Offered PSA screening | - | 0.8 rate ratio | - | <0.001 | link |
| Cumulative incidence of prostate cancer, median 9 years | Offered PSA screening | - | 8.2% | - | - | link |
| Control | - | 4.8% |
Shares ProtecT, Localised prostate cancer, intermediate risk, Localised prostate cancer, very low and low risk, Prostate cancer.
Shares ProtecT, Localised prostate cancer, intermediate risk, Localised prostate cancer, very low and low risk, Prostate cancer.
Shares ProtecT, PSA (prostate-specific antigen), Localised prostate cancer, intermediate risk, Localised prostate cancer, very low and low risk.
Shares ProtecT, Localised prostate cancer, intermediate risk, Localised prostate cancer, very low and low risk, Localised prostate cancer, high and very high risk.
Shares ProtecT, Localised prostate cancer, intermediate risk, Localised prostate cancer, very low and low risk, Prostate cancer.
Shares Localised prostate cancer, intermediate risk, Localised prostate cancer, very low and low risk, Localised prostate cancer, high and very high risk, Prostate cancer.
Shares PSA (prostate-specific antigen), Localised prostate cancer, intermediate risk, Localised prostate cancer, very low and low risk, Localised prostate cancer, high and very high risk.
Shares ProtecT, Localised prostate cancer, intermediate risk, Localised prostate cancer, very low and low risk, Localised prostate cancer, high and very high risk.