In May, the Government Accountability Office of the United States Congress released a report on herbal supplements, alleging that they are often marketed deceptively, and contaminated with the heavy metals arsenic, cadmium, lead and mercury[1].  The report was authored by Stephen D. Kutz, the GAO’s Managing Director of Forensic Audits and Special Investigations.

Of course the report got a lot of press.  Once again, herbal supplements were made to appear scary, dangerous, and here’s that word again unregulated.

This wasn’t the first time these issues have been raised, and it certainly won’t be the last.  But this particular report, coming as it does from the “investigative arm of Congress,” presented as testimony to the Senate Special Committee on Aging, as a likely precursor to Congressional hearings regarding the future of the supplement industry, demands scrutiny.

Allegations of Deceptive Marketing

Before I address any specific allegations, let me say that every industry has its share of deceptive marketing.  No matter what’s for sale, someone is going to try and deceive you to sell it.  I certainly don’t mean to say our industry is blameless, of course.  Still, I take strong exception to what the GAO reported.

The GAO did not list a single instance of “deceptive marketing” that came from a reputable manufacturer.  All of the claims were made either on unnamed websites, or by retail clerks speaking with undercover GAO agents.  (The agents posed as elderly customers looking for help, and led the clerks through a series of scripted questions).

Of course deceptive marketing should be upsetting, no matter who’s making the claims.  But normally we don’t let our experience on an unscrupulous website or with an uninformed clerk indict an entire industry.

But then we look at the claims that the GAO considers “deceptive.”  I’ll let you judge for yourself:

  • Chamomile may “treat insomnia.”
  • Garlic “could be taken in lieu of a prescription medication.”
  • Ginseng “cures diseases.”

In fact all but one of the GAO’s “deceptive” claims were simply statements that an herb might actually work.  Some were pretty outlandish.  But most were (in my opinion) well within reason.

It is of course illegal to imply in marketing materials that an herb can “diagnose, prevent, treat or cure any disease.” So if the GAO report wanted to talk about “illegal marketing,” I’d understand.  But “deceptive”?  I don’t see it.

(And then to add insult to injury to irony, this same report that considers it “deceptive” to provide information about the health benefits of herbal products notes that “consumers did not consistently receive clear, scientifically supported information concerning products’ health benefits.”)

Finally there are the clerks.  I won’t deny there’s ignorance out there, and if you ask the right questions you’ll expose it.  But I wonder what would have happened if the GAO had sent those same undercover agents in to doctors’ offices and pharmaceutical sales conventions, armed with their scripted questions and their hidden microphones.  How many physicians and drug reps would they have been able to entrap into saying that certain drugs are “safe” (when they’re not), or that there is “no research” on a natural treatment (when there is)?

In my experience, the majority of deceptive claims about natural medicine are made by representatives of the mainstream medical community.  But that’s just my opinion.  Ultimately we can’t know how widespread the problem is on either side since the GAO did not quantify its data.

Allegations of Contamination

Here, the GAO report does provide quantifiable data.  But before we address them, let’s start with some background:

  1. The GAO did not test vitamins, minerals, fatty acids, amino acids, or probiotics.  The only supplements tested were herbal products.
  2. Herbal products are agricultural products.
  3. Agricultural products are generally grown in soil.
  4. There is no such thing as “pure” soil, free of “contaminants.”
  5. Therefore there is no such thing as a “pure” agricultural product, free of “contaminants.”

Organics are a big step in the right direction, of course.  But no matter how you grow your crops, you’re going to find something in them you don’t want, at least if you test them on equipment sensitive enough.  This is as true for garlic pills as it is for garlic pesto.

So it’s a little disingenuous to simply report on the presence of contaminants; we need to look at the actual amount.  Here’s where I could just rely on the GAO report itself, which states that “the levels of contaminants found do not exceed any FDA or EPA regulations.”  But I won’t.

I could also say that most of the companies we deal with put their products through rigorous pre- and post-production testing for heavy metals, pesticide residues, and microbial contamination.  But that’s not enough, either.

Instead, let’s take the GAO report at face value let’s assume these anonymous products are representative of our entire industry.  Starting alphabetically with arsenic, let’s put the GAO’s numbers in perspective.


Of the 40 products tested, 21 showed traces of arsenic.  The average concentration was 0.115 ppm[2].

I wanted to compare this to common foods, or average daily intake.  Unfortunately, the only place I could find consistent data was with rice[3].

A paper published in the April 2007 edition of Environmental Science and Technology examines arsenic levels in rice grown both in Louisiana and California.  Of the 134 samples tested?including both white and brown rice, long-grain and short, organic and conventionally grown?not a single one tested free of arsenic.  The average concentration in California rice was 0.17 ppm.  In Louisiana rice, it was 0.3 ppm.  The lowest arsenic concentration out of the 134 samples was still 0.1 ppm, in a California-grown organic brown rice.

In other words, herbal supplements contain less arsenic than rice.

But that’s only half the story.  A suggested serving of rice is ¼ cup (uncooked), or roughly 50 g.  Bearing that in mind, an average serving of California rice would contain more arsenic than 147 standard capsules[4] of an average herbal supplement; a serving of Louisiana rice would contain more arsenic than 261 capsules.  None of which accounts for the fact that rice absorbs water during cooking, which could add even more arsenic.

I eagerly await the GAO’s report on pilaf.


28 of the 40 herbal supplements tested by the GAO contained detectable cadmium.  The average concentration was 0.083 ppm.

It might be sufficient to compare that to the 15 ppm (181 times as much) that the FDA allows in food colorings.  But the goal here is not to compare supplements with useless junk; it’s to place them in the context of a healthy diet.

And cadmium, like it or not, is part of a healthy diet.  According to the U.S. Agency for Toxic Substances and Disease Registry (ATSDR) of the Department of Health and Human Services “All soils and rocks… contain some cadmium.”  Most of our daily exposure comes from foods, especially shellfish, chocolate, grains and beans, as well as trace amounts from air and water.  Tobacco smoke can also be a very significant source of cadmium.

Let’s leave the smokers out of it.  According to the ATSDR, the average American consumes roughly 30 mcg of cadmium a day from food.  According to the CDC, the average American consumes 2247 calories a day.  An average standard capsule of the herbal supplements tested by the GAO contains 0.041 mcg of cadmium.  So when we put these three figures together, we see that a standard capsule of the herbal supplements tested by the GAO contains as much cadmium as you’d expect to get from about three (3) calories of food.


The GAO reported detectable levels of lead in 37 of the 40 samples tested.  The average sample contained 0.225 ppm.

It is nearly impossible to generalize about exposure to lead in the general population.  While a lot of work has been done to establish blood levels, dietary intake is still a huge question mark.

The FDA has not established daily levels it considers safe or unsafe.  However the Food and Agricultural Organization of the United Nations World Health Organization (FAO/WHO) sets a provisional tolerable weekly intake limit at 25 mcg/kg of body weight, or 1,750 mcg/week for a 154 pound adult.

First, it should be noted that FAO/WHO only sets provisional tolerable weekly intakes limits for “those contaminants unavoidably associated with the consumption of otherwise wholesome and nutritious foods” (emphasis added).

Considering that the average standard capsule tested by the GAO contains 0.113 mcg of lead, you could take 22 of them a day and still get less than 1% of the cadmium considered acceptable by the FAO/WHO.


32 of 40 herbal supplements tested by the GAO contained detectable traces of mercury.  The average concentration was 0.0048 ppm.

(More than half the mercury came from just two products.  Adequate quality control in these two should definitely have brought the average down.  But as a matter of principle, I’ve included them in my calculations).

It’s nearly impossible to generalize about mercury intake, since so much of it correlates with fish intake, which varies wildly from person to person.  So let’s look at this a different way.

At least three of the H1/N1 “swine” flu vaccines produced last winter used a mercury-containing preservative called thimerosal, supplying 25 mcg mercury per injection.  You would need 10,504 standard capsules of the herbal supplements tested by the GAO to equal that amount.

Or think of it this way: the FDA allows .002 ppm mercury in drinking water, or 0.472 mcg in an 8-ounce cup.  So, let’s say you wash down your daily pills with a nice glass of water.  You could be getting more mercury from that one glass than from 198 average standard capsules.


Anyways, that’s the GAO report in a nutshell.  And by the time you read this, this particular round of accusations and allegations will have faded away.

But it’s only a matter of time before the next set of overblown data, the next round of political grandstanding, the next media feeding frenzy, the next condescending “scientific” reproach to the quality of herbal products.

[1] You can read the full report online at

[2] This figure assumes, pessimistically, that the samples with no detectable levels still contained arsenic equal to half the GAO’s threshold of detection, or 0.0375 ppm.  I will be using the same methodology throughout the rest of this paper.

[3] It may be argued that rice, as a “wet crop,” is more likely to accumulate heavy metals than other foods.  Fair enough.  But it’s still the only crop I could find consistent data for, and a crop that’s generally considered safe.

[4] Here, a standard capsule is 500 mg.  It should be noted, however, that two of the herbs repeatedly tested in the GAO report, Ginkgo and Saw Palmetto, are customarily sold in much smaller capsules of concentrated extract (60 mg and 160 mg, respectively).  The GAO did not specify the size or form of the capsules tested.