INTRODUCTION The relationship between circumcision and prostate cancer has been controversial. A recently published meta-analysis contradicted previous meta-analyses of male circumcision and prostate cancer risk. Our aim was to conduct a de novo meta-analysis and critically evaluate this recent paper published by Van Howe.
We retrieved data from each of the 12 source studies Van Howe used, then performed a random effects meta-analysis of those data. We critically examined the data and other information in Van Howe's study.
Using the same values as Van Howe, we confirmed his finding of a positive association of circumcision with prostate cancer (random effects summary OR = 1.14; 95% CI 0.99, 1.31). However, our independent meta-analysis found a negative association of circumcision with prostate cancer (random effects summary OR= 0.87; 95% CI 0.76, 1.00; p = 0.05). The reason for this critical discrepancy was Van Howe's erroneous transposition of values for circumcised and uncircumcised men in his Table columns, leading to inversion of the result. We further critically evaluated a geographical analysis and cost analysis of circumcision and prostate cancer, as well as claims denying a role for sexually transmitted infections in prostate cancer etiology, finding these too to be misleading.
Van Howe's 2020 meta-analysis was based on erroneous data transposition leading to an inverted outcome. The journal concerned recently corrected his Table. Van Howe's claim of a positive association of circumcision with country-level-age standardized prostate cancer prevalence and his cost analysis were found to be questionable. Our meta-analysis showed that circumcision is associated with lower prostate cancer risk.
The Canadian journal of urology. 2021 Aug [Epub]
Brian J Morris, Jim G Matthews, Noel Pabalan, Stephen Moreton, John N Krieger
School of Medical Sciences, University of Sydney, Sydney, New South Wales, Australia.