Their environmental contaminant hypothesis is very interesting and seems like it could be further confirmed or disproved with more data.<p>Can we link childhood obesity to areas with particularly high rates of some kind of industrial contamination? (Poverty as a confounding effect may make this hard)<p>If the watershed is part of the mechanism, can we confirm lower obesity rates in areas without much industrial pollution that have similar diet & exercise to obese areas? (Independent of the already-mentioned altitude effects)