In response to my blog entry yesterday I was sent the following comment:

Dear Dr. Novella,
After reading your blog post on drugs vs. herbs I would like to ask you what your opinion of “adequate scientific evidence” is. I feel less confident with current drug development than throwing a couple of plants that people have been using for thousands of years (how is that for sample size?) in some water than I do trying certain medications.
Allow me to explain. As a scientist I dabbled in a couple of fields before specializing in genomics. One of them was clinical trials for 3 different drugs for ALS. We were the national headquarters so I was privy to data from the whole country. NOT ONE adverse event was attributed to the drugs. I find that hard to believe. I am aware of the implications of ALS as a disease but the symptoms reported were not congruent with that. I also understand the pressure of fellowships, the massive amount of time and money it takes to push a drug through, and the fact that the majority of subjects of clinical trials do not reflect all of the populations the drug is intended for.
I am not forsaking drugs for herbs. I don’t understand why it has to be one or the other. Both sides have instances where public safety is not at the forefront. Herbal companies that treat herbs like drugs (isolating compounds, capsulizing them) can be dangerous and be just as greedy for money. Herbs used in a traditional sense taught by generations can be effective. Just because natives haven’t put their medicine through your process of scientific inquiry does not necessarily mean they are ineffective. The traditional way these herbs are used by definition would not be easily recorded or accurately analyzed by clinical trials.
Plant based pharmaceuticals are rarely in the dosage found in nature. Also, because only certain compounds are isolated (often alkaloids) the chemicals are not working in the same manner as they would in the plant. Plants do have nutritional value, by the way. Ever eat a salad? How about the calcium, iron and other vitamins in Nettles?
Thanks for reading,
Michele

This raises several points commonly raised in defense of herbal remedies, and also requires some clarifications. The most relevant is the implication that herbs are probably safe (and may even be effective) because “people have been using (them) for thousands of years.” I file this under: the plural of anecdote is anecdotes – not data. Even a thousand years of anecdotes is not reliable evidence for safety or efficacy, and there are several lines of argument to support this conclusion.

The first is that there are numerous examples of treatments or remedies that cultures have used for thousands of years under the common belief in efficacy without anyone noticing that the treatment was useless or harmful. Blood letting survived for over two thousand years. Tobacco was used by Native Americans for centuries as a medicinal plant. Urine drinking is a traditional Indian remedy.

Second, most biological activity of plant-derived drugs are subtle. It is reasonable to conclude that pre-scientific cultures, based upon anecdotal experience, would figure out the obvious or immediate effects of many medicinal plants. Many plants are poisons and would make one immediately sick. Others have acute and reliable effects, such as vomiting, diarrhea, or hallucinogenic effects (humans seems particularly adept at discovering hallucinogens). Other effects are apparent and reliable enough to be discovered over time – such as reducing fever, treating a rash, or pain relief. In fact, plants that cause such effects have largely already been studied, their active ingredients identified, purified, and sometimes chemically tweaked, and such substances make up a large part of our modern pharmaceutical repertoire.

But what about more subtle effects? Anecdotal experience, even the more systematic experience within a tradition of natural healing, would not be sufficient to detect statistical benefits or risks of plants and herbs. For example, the recent Vioxx scandal has been used to argue to me that herbs are safer than drugs. But unless a particular herbal drug or preparation were tested in the way that Vioxx was, we would never know if it presents the same risk. It is easy to say that herbs are safe when they have never been studied. None of the millions of people taking Vioxx could perceive that their risk of heart attack was slightly increased.

Further, the placebo effect, regression to the mean, confirmation bias, and a host of other factors conspire to cause the false impression of benefit where none exists. Anecdotal experience is likely to create the false impression that a substance works – it is likely to confirm cultural belief rather than contradict it. That is how blood letting survived for thousands of years.

So the bottom line is that even a very long history of common use does very little to reassure that any remedy is effective and does not rule out harmful effects beyond the most obvious and immediate ones. Michele wrote: “Just because natives haven’t put their medicine through your process of scientific inquiry does not necessarily mean they are ineffective.” True, but irrelevant. What it does mean is that we do not know if they are safe or effective.

The next implication, that the process of drug development is not adequate to insure safety, has some merit but must be put into proper perspective. First, we must avoid the tu quoque logical fallacy of saying that low standards for herbs are OK because the current standards for drugs are low also. My position has always been that we should have high standards for all health products (and again, avoid any false distinctions).

The drug development process in the US certainly has flaws, and even a perfect system must make compromises. For example, critics equally charge that the system is too cumbersome and slows the arrival of life-saving drugs on the market, and that the system is too permissive and allows drugs to come to market too easily. These concerns are at cross-purposes, and so the only option is to strike a balance: require a reasonable degree of evidence for safety and efficacy while making it possible for effective drugs to come to market without excessive delay. Balance means compromise.

Having said that I am all in favor of more thoughtful and effective clinical trials that thoroughly track safety and side effects. I think the system is actually pretty good now, and it is in a constant state of evaluation and adjustment (at least in my experience). Industry regulations also generally work but need constant vigilance. For example, I support recent efforts to register all human clinical trials so that drug companies cannot bury negative outcomes they don’t like. There is also efforts under way to improve the post marketing monitoring of drug safety so that drugs like Vioxx will be picked up more quickly.

But I reject the implication that drugs are slipping through the system without any adequate filter for safety. The ALS trials cited by Michele are likely not representative (a problem with small sample size). I have also been involved in ALS clinical trials. One drug I studied (topiramate) had nausea, weight loss, and kidney stones as side effects – all reported as likely due to the drug. (The trial was ultimately negative.) I have also seen directly many drugs killed in the middle of clinical trials due to adverse events, and other pulled shortly after going to market because of the emergence of side effects or toxicity. There are countless examples of the system working. Most experimental drugs, in fact, do not make it to market because they do not survive the rigorous system of testing required by the FDA.

My point is not to defend the industry or the current state of regulation but rather to point out that most criticism leveled against pharmaceutical research and regulation is designed to make the broad and largely invalid point that current regulation is ineffective, therefore unregulated plants and herbs are just as good as FDA approved drugs. This is not constructive and is misleading. Rather, I accept criticism that is constructive and designed to make the system work better – which should be everyone’s goal.

When evaluating such criticism it is very useful to ask the basic question: what is the goal of this criticism? Is it to suggest that we should throw up our hands and say that anything goes, or is it to point out weaknesses in the system and suggest specific improvements?

Regarding the final point that plants have nutritional value – I never said differently. What I said is that herbal remedies are not taken for their nutritional value and that, as they are taken, they have negligible nutritional content. If you take a pill of St. John’s Wort, for example, you are not getting any significant nutrition. What you are getting are a number of pharmacologically active substances. This has no relevance to plants as food.

What this reflects, however, is the common deliberate blurring of the line between food and drugs – exemplified in the term “neutraceutical.” Of course, food can also contain pharmacologically active substances – caffeine being a common one. But what this movement is all about is getting drugs to be regulated as if they were food. And in fact, in the US this has been successful. Herbal drugs are now regulated largely as if they were food and not drugs. This is a mistake and is against the interests of consumer health.