Depends on lots of things. Are they reporting the true relative risk or has the journalist got the wrong end of the stick? And was it a long-term interventional study (unlikely) with a pre-defined endpoint or was this simply among the best results from a data-dredge, thus worth nothing more than "interesting result that might or might not be worth some follow-up".
Of course in the event of correlation it could be simply something else, such as men who drink 6 or more cups of coffee get rather more of life's greatest pleasure as well, which apparently (according to equally credible studies) reduces the risk of prostate cancer as well. And is that because forming the two backed beast genuinely reduces prostate cancer risk or because those who engage more in hanky-panky are younger and fitter on average than those who do not?