prev next front |1 |2 |3 |4 |5 |6 |7 |8 |9 |10 |11 |12 |13 |14 |15 |16 |17 |18 |19 |20 |21 |22 |23 |24 |25 |26 |27 |28 |29 |30 |31 |32 |33 |34 |35 |36 |37 |38 |39 |review

First, we examined a large number of characteristics of the man and of the semen sample, such as age or recent fever that have been associated with semen quality. But adjusting for these potential confounders caused little change in the amount that semen quality differed between centers. Therefore, confounding does not appear to be a likely explanation for our findings.