Alex Lickerman M.D.

Happiness in this World

The Problem With Alternative Medicine

Why "natural" isn't necessarily better

Posted Sep 12, 2010


In general, alternative medicine is used to describe practices outside the bounds of conventional medicine. It's also often connoted to include practices that haven't been shown to be effective. Mainstream scientists often criticize alternative medicine as charlatanism, arguing that anything alternative that's been proven to work is in fact...mainstream medicine. Advocates of alternative medicine, in contrast, typically point to their personal experiences as proof of the effectiveness of many such "unproven" interventions.

Many people feel more comfortable using natural compounds rather than man-made ones because of a belief that what's natural is, by definition, healthier and safer—but this is a dangerous assumption. Numerous natural compounds are, in fact, poisonous (e.g., cyanide). And natural compounds that have health benefits often have a narrow therapeutic index (meaning the amount that brings benefit is only slightly less than the amount that causes toxicity, making it alarmingly easy for their use to cause harm), like digitalis, a drug derived from the foxglove plant that's been used to treat heart failure since the late 1700s.

Many patients tell me they "don't like to take pills" but paradoxically think nothing of ingesting "natural" herbs or plants. This distinction has always struck me as absurd. Both man-made and natural compounds have effects on biological systems. Nothing about natural compounds makes them more or less effective or safe. How confident we can be that a given compound is both effective and safe has nothing to do with where it comes from but rather with how rigorously its effectiveness and safety have been studied.

This, then, is the main problem I have with alternative medicine, that in general its interventions haven't been subject to rigorous study. I'm open to believing anything, no matter how far fetched it may seem (after all, who would have believed that television would be possible five hundred years ago?). I don't even need to know an intervention's mechanism of action. I just need to know that in well-designed studies it's been proven effective.

And safe. Prima non nocere: first, do no harm—medicine's most important credo. When you're the one with the authority to recommend treatments, you take very seriously the possibility that what you recommend may cause harm. I'm even open to the use of interventions that haven't been proven ineffective as long as there's reason to believe they're safe.

But safety is a tricky thing. First, it never exists as an absolute. That is, an intervention can be safe for some people but deadly to others (e.g., penicillin is safe for almost everyone who takes it—except for those who are severely allergic). Further, "safe" is often a value judgment that varies from person to person and for one person from situation to situation. We may think it obvious that anything with a risk of death isn't safe—but then again, what if that risk of death is less than 1% and we're talking about performing a cardiac catheterization in a patient who's just had a heart attack?

Alternative medicine enthusiasts often point to centuries of use to justify their belief in the safety of many alternative medicine practices (if acupuncture kills or maims, why hasn't it been reported?), but without carefully designed studies, how do we know, for example, that black cohosh doesn't increase the risk of auto-immune diseases like lupus? Certainly, numerous examples exist in which man-made compounds were subject to rigorous study in clinical trials, labeled acceptably safe, and then later discovered not to be (remember phen-fen?). But this is the exception rather than the rule.


Interestingly, as more and more studies of alternative therapies are being conducted, we're discovering that many of them do work—but not necessarily for the things people believe. Acupuncture, for example, has been suggested to be effective in reducing the nausea associated with chemotherapy. But trials also show for back pain there's a good chance it's no better than placebo.

On the other hand, even when clinical trials show an intervention works, it still may not. Some kinds of experimental designs inherently produce less convincing results than others. And even double-blind, prospective, randomized, placebo-controlled trials—the gold-standard of clinical trial design—often turn out to contain flaws, even those published in well-respected scientific journals. Often these flaws are small enough not to impact the validity of results-but sometimes they do. This is why treatments must often be studied multiple times in multiple ways before they gain acceptance in the scientific community.

People turn to alternative therapies for many reasons. Perhaps because of a bad experience with traditional medicine. Perhaps because personal experience strongly suggests they work (when my wife was pregnant with our son and feeling intense nausea, I tried applying pressure to her wrist at a standard "acupressure" site and was astounded to hear her say, repeatedly, the nausea vanished when I pressed and returned when I released). Perhaps because traditional medicine has no more to offer. Especially when the diagnosis is terminal, what's to lose?

If you're going to go the alternative route, do as much research as you can. Look for treatments that have been shown in some kind of study or studies not to cause significant harm. Look for treatments that at least have some anecdotal reports of success for the specific problem you have. (And I don't mean reports offered by people selling alternative interventions. Though many, if not most, alternative medicine practitioners are genuinely interested in helping others and believe in the effectiveness of the therapies they offer, snake-oil salesmen abound. Beware overconfident statements about efficacy. Remain suspicious about any claims that a given intervention treats or cures a multitude of disparate conditions.)

Traditional treatments aren't inherently superior to alternative ones—they're just better studied. Which doesn't mean that they're effective or safe beyond any doubt—just that they're more likely to be. The praises we should be singing aren't to any one specific intervention, traditional or alternative, but to the scientific method itself, a way of making valid observations about cause and effect in the phenomenal world. It's both a lot harder to do well than many of its proponents understand, yet simultaneously a far more valid means of assessing efficacy than anecdotal reports and length of use could ever be.

If you enjoyed this post, please feel free to explore Dr. Lickerman's home page, Happiness in this World.