You should take a course in research design and biostatistics... Just because you can find a statistical correlation (which isn't necessarily that difficult to do) doesn't imply that it is a scientifically meaningful correlation, or that it is necessarily statistically significant. Furthermore, correlation doesn't necessarily imply causation. There are lots of things that need to happen in order to show causation, including, but not limited to, the need to control for other factors and to show dose-dependency. It's a good idea to build a good understanding of research principles. As future physicians, we are obligated to teach our patients what is good science and what is bunk, because they are going to come to you with all kinds of crap from the media.
This being the case, I say to your proposal, "so what, if you can find a correlation?" It doesn't mean much to educated populations, unless you can, through a well-designed study, show scientific significance and determine a meaningful relationship.
It's easy to take a bunch of data and make it mean something, to conveniently fit your hypothesis (to meet your ends), but if you want to show something scientifically, you are going to have to be more elaborate than that. And thank goodness for that.