Another day, another study that ignores tons of available data and decides to take some indirect loosely related measures on a heavily biased population, using variables that haven't been proved to be correlated to the ones they're supposed to be measuring instead.