A researcher from the University of Missouri School of Medicine has found evidence of higher occurrence of endocrine disrupting chemicals near fracking sites when compared to presence in control areas. The study was published in the journal Endocrinology.
Based on the information in the abstract, the study collected samples from drilling sites, the Colorado River, and a control site in Missouri. The drilling sites had elevated levels of contaminants, and the Colorado River showed a higher level than that of the control site in Missouri. To be fair, the study does not indicate that these endocrine disrupting chemicals had been released directly into the natural system from fracking sites, but merely points to the evidence as indicating the increase in industrial activity in remote sites present a greater likelihood of contamination. The study certainly verifies the presence of endocrine disrupting chemicals in the fracking process (something, you may recall, that Josh Fox mentions highlights in Gasland without making a qualitative link).
I want to emphasize here again: wastewater storage, treatment, and disposal are the most critical pieces that must be in place for fracking to be any kind of success.