Direct link: <a href="http://www.informationisbeautiful.net/play/snake-oil-supplements/" rel="nofollow">http://www.informationisbeautiful.net/play/snake-oil-supplem...</a><p>Interesting, but not very comprehensive. A number of supplements that I've found relevant to recommend to friends, including GABA, 5-HTP and levodopa are absent from the graphic (or did I miss them?).<p>It's also problematic that some of the conditions indicated are very general. There are many possible biochemical causes for depression or insomnia, and a treatment that works for one might have no effect on another.
It is confusing that "Evidence", which is the label of the y axis, is graphically presented in the same way as the values on the y axis ("Strong", "Good", "Promising", "Conflicting").
Don't care much about the data itself but the use of Google Docs as a data source in combination with whatever flash library they're using for the visualization is an interesting approach. Too bad there aren't more details on how they're reading from Google Docs. My guess is it's a Flex graphing library. Anyone recognize it?
It would be more interesting to me to see this with a bar for "none" and "aboslutely not". But there indeed does seem to be some correlation to the number of search results and effectiveness.