What's the problem with placebo medicine?
A while ago I wrote about acupuncture working as a strongly-proven placebo. Placebos makes some people feel better. Who cares how it happens as long as it happens? What's the problem with placebos? If you can feel good from a medication or treatment that has or hasn't passed double-blind trials what's the difference as long as you feel good? Short answer: because the community of medical ethicists have determined that selling placebos as medicine is unethical. A much longer answer follows.
First, allowing placebos as legitimate treatment for some issues opens the door to using placebos against other issues. Homeopathy provides relief from some symptoms as a strongly-proven placebo, leading homeopaths to offer their treatment as a way to prevent malaria even though there is no published evidence to support the use of homeopathy in the prevention of malaria. I know doctors who have seen patients in late stages of bone cancer because their chiropractors misidentified their serious conditions as "back and joint pain" and discouraged them from seeking effective mainstream treatment. Compare life expectancies between cultures where allopathic, science-based medicine is prevalent and cultures where it's largely absent. Compare how life expectancies increased in societies that adopted allopathic medicine after we started recording these numbers. Not just in isolated anecdotes; the population as a whole. Science works, bitches.
Not all placebos are as harmless as sugar pills, either. "Placebo effect" means that the physical effect is misunderstood or misreported, not that there is no physical effect. It's not just possible but common for mainstream or alternative patients to report positive effects from treatments whose physical effects are not positive. The history of the patent medicines which required the Food and Drug Act of 1906 is filled with these harmful treatments and their proud supporters who insisted that they felt a positive effect when they were actually poisoning themselves. Medicines based on "natural" ingredients are still drugs; they still operate chemically, they can still produce physical adverse effects even while their supporters insist "it works".
The placebo effect isn't necessarily a therapeutic and healing effect, either. Psychosomatic disorders are illnesses caused by placebo; the belief that something is happening to the body in the physical absence of an active agent. The nocebo effect is the placebo effect in reverse; people have been being bitten to death by nonvenomous snakes, patients report negative reactions from inert or beneficial drugs entirely due to a pessimistic belief or expectation that it will produce harmful, unpleasant, or other undesirable consequences. Allowing disciplines, medicine, and therapy that discounts double-blind testing, controlled trials, and other evidence opens the door not just to worthless and wasteful treatments but to illnesses driven by fantasy as surreal as the demonic posession that physicians of antiquity once treated. It can render genuinely beneficial medicines useless to legitimately needy patients. It makes all disease a matter of personal opinion. If you think you're pregnant who's to say that you aren't? Not science, apparently.
Even if you assume a completely inert and physically harmless placebo and a purely beneficial effect, the benefits of the placebo effect are not exclusive to placebos. You don't have to choose "either/or". Legitimate medical practices can and are deliberately enhanced to be placebo-favorable. Pill size, shape, and flavor, bedside manner, expectation, patient participation, practitioner enthusiasm - there are substantial studies and evidence to indicate that it's not just what you treat your patient with but how you treat them, and allopathic practitioners know this. Effective allopathic practitioners perform their treatment so that the placebo effect will be working with them, not against them. If you're going to get the placebo effect either way, why not get actual productive treatment as a bonus? A beneficial placebo effect can be used in legitimate therapy, it just can't be the only effect from a therapy.
Placebos are expensive. The point of the placebo is that it works just as well as something that's totally inert and not doing anything - another placebo. As long as you're buying something that doesn't do anything you might as well buy the cheapest substitute that also won't do anything. Tap water instead of homeopathic preparations. Rocks that you dig out of the nearest hillside instead of "exotic crystals". Use a stack of washers from Home Depot sewn to an old oven mitt instead of a $200 "magnet pad". It gets a little complicated when you consider that placebos work even better when they're a ripoff - you get an even stronger placebo response if you've paid $500 for something that doesn't do anything than if someone gives you a placebo on sale or for free. You might as well just stick with evidence-based medicine where you have to prove that the treatment actually does something.
Perhaps most importantly, doctors and patients need to have an honest and open relationship with each other. It would be one thing if placebo practicioners were open and honest about the treatments they were providing: "I'm going to give you something that will have no effect except causing you to believe that something is working". But this rarely happens, if ever. (If you're aware of placebo-based treatment where a treatment which produces no physical change in double-blind conditions is actually advertised or represented as such please let me know.) Placebos require that the patient be misled about what isn't actually happening, almost by definition. Blockages in qi flow, subluxation, water memory - mechanisms which are demonstrably and provably false. Placebo medicine is based on deception or self-deception.
This is why maintream doctors try to expose placebos even in their own practices when they mistake them for legitimate treatments. They welcome skepticism and peer review, studies and followup studies. They invite and encourage this sort of criticism because the history of science and medicine is filled with past mistakes, and they want future mistakes to be identified and corrected as quickly as possible. Legitimate practitioners stop treatment when they discover that their formerly respected procedures are only effective as placebos. For example: in the 1950s there was a surgical procedure called mamillary artery ligation in which an artery in the heart was blocked to force more blood through a second artery and "flush it out". Patients reported that this procedure made them feel better. Then a study was conducted where patients were given a small incision which was sewn up and they were told that they'd had the procedure done - those patients reported a benefit that was indistinguishable from patients who'd had the real procedure done. With this news, doctors stopped treating arteriosclerosis with mamillary artery ligation - because the procedure was a placebo, and it's unethical to provide placebos as treatment. It's a mistake to provide placebos as legitimate treatment, even by accident. No reputable doctor today would advise their patient to undergo this procedure, even if they knew it would make them think they felt better, because there's a difference between thinking that something is happening and something actually happening.
The placebo effect isn't "real" or "fake", it's illusory. It's unethical to sell someone an illusion if they believe it's real. It's unethical to sell a vegetarian a ham sandwich even if you think that pig is a vegetable and you persuade them that the the ham is an extremely good vegan soy substitute. It's unethical to sell someone antivirus software that always reports "no viruses found" even if this assures them that their computer doesn't have viruses - even if you believe their computer isn't infected and "no viruses" happens to be a correct diagnosis. It's unethical to sell oregano to high school kids and tell them it's weed even if you think it's weed and the combination of suggestion, inexperience, and peer pressure makes them say "I think I feel something". It's unethical for me to use visual effects to mock up some footage depicting something that I think happened and sell it to Channel Seven as legitimate news even if nobody's property or reputation is harmed by the story and the footage is so flawless that everyone believes it's real. And it's unethical to sell a pill, tonic, ointment, or other treatment by saying "it works; just trust us" whether you're Big Pharma or Big Chinese Tea. There is more to medicine than whatever you can put over on your patient. Every practitioner - mainstream or alternative - needs to be able to demonstrate the efficacy of their treatment independent of placebo effects so that their customers don't come back later outraged that their time and money was wasted on ordinary sugar pills, tap water, and pointless exercises.
The motivation to profit through deception or self-deception is a powerful one no matter who is practicing it. Scientists and doctors don't arrogantly assume they're right about everything, they embrace double-blind testing and peer review because they know they might not be right. Everyone should operate under the same strict regulation with open standards of evidence and transparency. Everyone needs to make sure they're not misrepresenting what they think is true. There should be no difference between the way "mainstream medicine" and "alternative medicine" are regulated. The treatment itself is either effective or it isn't. Anyone who doesn't want to figure out the difference - or argues that there is no difference - is inviting trouble on themselves.
Update: Not that it's not worth studying. This paper just won a 2008 Ig Nobel.