Figure 3.

Distribution of epidemiological observations published before and after the implementation of PSA testing in 1995. We found 12 manuscripts published between 1980 and 1995 that deny any link or report a weak association between alcohol consumption habit and risk of PCa. Within these years, only five publications reported the positive connection. In the observations published from 1995 to the present, the ratio of publications showing no link:positive link was shifted to 18:65. There is a significant difference between the publication years of articles with positive correlation and those with no correlation (Mann–Whitney test, * p < 0.0001; median ± SD, the median for no correlation is 1998, median for positive correlation is 2014).