Take any observational study with a grain of salt as big as your head.

As the study, published in the BMJ on Tuesday, was observational, conclusions could not be drawn about cause and effect