If you do an adequate job of looking for confounders you'll likely torpedo the study because it will turn out to be something already known.