Six RCTs with 387,286 participants (range 9,026 to 162,243) were included in the review. Median follow-up time ranged from four to 15 years. Methodological limitations were common; only three trials reported blinding of outcome assessors and only one reported adequate allocation concealment. Possible reporting bias was also identified as an issue. Quality of evidence for individual outcomes, assessed using the GRADE (Grades of Recommendation Assessment, Development and Evaluation) system, ranged from moderate to very low.
Screening was associated with an increased probability of receiving a diagnosis of prostate cancer (RR 1.46, 95% CI 1.21 to 1.77; five RCTs; n=340,800 men; statistical heterogeneity was significant for this outcome) and of being diagnosed at the earliest stage (stage I RR 1.95, 95% CI 1.22 to 3.13; four RCTs; n=332,743 men). Screening did not significantly affect death from prostate cancer (RR 0.88, 95% CI 0.71 to 1.09; five RCTs; n=302,500 men) or all-cause mortality (RR 0.99, 95% CI 0.97 to 1.01; four RCTs; n=256,019 men).
There were no data on effects of screening on quality of life and limited data on potential harms of screening.
Results for publication bias were not reported separately because the method was considered unreliable for reviews with fewer than 10 trials.
Results of other analyses were reported in the paper.