"snake oil" was not listed. Surprising as it is said to offer real relief for joint pain: http://healthmad.com/alternative/health-benefits-of-snake-oi...

The origin of the term "snake oil salesman" was to disparage people who sold fake snake oil or pushed it for diseases it would not treat (http://en.wikipedia.org/wiki/Snake_oil).

Hmm, first thing I notice looking at their data is that they put red yeast rice on top for blood pressure and cholesterol based on two studies, one of which did not assess red yeast rice, and neither of which assessed blood pressure.

The study that did not assess red yeast rice did assess lovastatin, which is assumed to be the active ingredient in red yeast rice, but if you're going to assume that they are equivalents, red yeast rice is definitely not "worth it" because you can buy a month's supply of the active ingredient for $4 at most any pharmacy, less than you would spend on the rice, and get a more precisely controlled dose.

Still a neat site design.

It's a cool visualization, but their cutoff line is labeled "worth it." To really justify that, they need more information. Thankfully the bubbles are broken down by specific health claim (so that some supplements have several bubbles for their effect on different conditions), but to determine "worth it," they need to include:

1. Some sort of way to account for type and frequency of side effects. For example if everyone who takes the supplement gets an intolerably bad taste in their mouth, that's a mild but prevalent side effect. If one person in 100,000 who takes the supplement dies, that's a comparatively rare but very serious side effect. This factors heavily into the "worth it" determination, because risk is a consideration.

2. Cost. If a supplement has a mild beneficial effect but costs $1,000 a month, I'm not sure most people would say it's "worth it."

3. Strength of effect. A statistically significant improvement is not necessarily clinically significant. For example, with a large sample a 1% decrease in LDL cholesterol may be statistically significant, but it is not clinically significant.

Those factors would help a person determine the pros and cons effectively. What is "worth it" varies considerably from person to person.

Anyone know how they are generating the visualization automatically from that google doc? Is there a tool for doing this?

Good god, I wish there was a visualization like this for everything. Especially if you could input your own parameters and get your own personalized "worth it" line.

Good god indeed - there's a germ of an idea for a Google Labs project right there. And if it worked well, it could change decision making of ALL kinds.

Might be worth emailing them. It's generated from a Google Docs spreadsheet, so it's entirely possible.

I was wondering why Green Tea is on there three times until I realized you can click on each circle to see what symptoms its referring to.

If you prefer to see the chart with that already on there, use this version: http://www.informationisbeautiful.net/visualizations/snake-o...

A major problem with this graph is that it mixes data from substance-deficient groups with data from normal groups, despite claiming that it shows benefits when taken by "an adult with a healthy diet."

For instance, the data on creatine found an increase in cognitive ability in vegetarians. Since creatine is found primarily in muscle tissue, it makes sense that the vegetarians (who were most likely to be deficient in this substance) would see improvement. However, when they tested the supplement in normal adults, there was no improvement. Yet creatine is above the "worth it" line.

Contrast that with vitamin C. It's below the "worth it" line, and it probably is for any normal person. However, if you were to test it on sailors who had eaten nothing besides canned meat for the last 6 months, it would probably be at the top.

Sure, we can assume that most of us are not vitamin C deficient, but we can also assume that most of us are also not vegetarians and therefore not creatine deficient.

(Maybe they considered vegetarianism a "healthy diet"? But it should be obvious that it's not, since it's almost impossible to get many of the top nutrients on this chart from your diet. For instance, if you want to get omega-3 fatty acids from a vegetarian diet the only way is consume supplements extracted from algae.)

Hmm, two on there I use:

1. Vitamin E for nerves. Not nervousness, but things like a little random twitching, especially right below an eye. Clears it up in a few days.

2. Melatonin for sleep regulation. Never more than 3 days straight, but it's a really nice reset switch for sleep if I've been hacking late nights recently.

Of course, ymmv..

(edited for formatting)

Two words: "healthy diet." That's all you need. If you have that, you don't need supplements.

I think I understand where you're coming from, and I mostly agree. However, I think it is a mistake to generalize from the idea that most supplementation is misguided, to the idea that all of it is.

For example, consider vitamin D. Humans make this in their skin, using sunlight. Now, what about those of us who live far from the equator and work indoors? For some time, the standard answer to this has been fortified milk. Okay, but most people in the world lack the mutation that lets most northern Europeans, and some Africans, digest milk sugar as adults. With the limited options available, vitamin D supplements start looking like a worthwhile option.

And then there are the non-nutrient supplements. A healthy diet should get you most of your vitamins, etc., but it won't get you the antidepressant found in St. John's wort. And it might not get you the probiotics that can combat H. Pylori infections, or the stuff in cinammon that seems to be effective against some kinds of diabetes.

This is about snake oil, but there's no "negative evidence" line?

I think negative ones aren't even worth including and would confuse it. Also, when they prove really negative like ephedra, don't they usually get banned?

I was thinking more along the lines of "we tested this, and it doesn't help condition X", not "this will harm you". The article's down now, but I recall the lowest level being "no evidence". That implies that no tests have been done.

Ephdra is not banned in the US, it's merely restricted. It's still available over the counter in Canada as well.

It's definitely banned in the US. There are some "safe ephedra" products which are actually some other stimulant - usually bitter orange.