There is an interesting effect in mood related studies where every group usually improves on average, even if they receive only placebo. The real measure of any depression study is therefore the improvement above and beyond the control group.
The effect is so predictable that it has become an easy way to produce studies showing positive effects from something: All you have to do is track some group of people and then give ALL of them your supplement, and their scores will improve. It makes great headlines because most people don’t know you could have given them sugar pills and the scores would also improve.
EDIT: As another commenter pointed out, this person has an unusual long list of financial affiliations with companies that produce drink products, too. This study should probably be ignored
