Weak Drugs or Powerful Placebos II: Don't Shoot the Messenger

In the previous post based on Steve Silberman's excellent Wired article (http://www.wired.com/medtech/drugs/magazine/17-09/ff_placebo_effect?currentPage=all) I took the perspective more favorable to the plight of the drug industry, worried that even good new drugs will fail to pass muster because of an overly inflated placebo response in double-blind trials. Now let's balance out the ledger by looking at the case for the drugs being weak, and that beating up on the placebo as the culprit is simply shooting the messenger.

First of all let's recall the complaints about the industry from critics such as Marcia Angell (The Truth About the Drug Companies). Angell and others argue that the companies get away with a lot, and are allowed to peddle a ton of me-too drugs, because a placebo-controlled trial sets such a low bar. All they have to do is prove their drugs are better than sugar pills. Ideally, for medical practice to move forward, companies would have to show that their new drugs beat out not the placebo, but the standard treatment already on the market (head to head trials)--the sort of comparative effectiveness research the drug industry and its minions are doing their level best to scuttle. So on that basis it seems weird to hear the drug companies complaining that demanding that they prove that their new drugs are better than placebos sets the bar too high, due to claims that the ol' sugar pill is suddenly on steroids.

There are several other passages in Silberman's article that suggest that the real problems may not be what the drug company wishes to think. An example:

"Big Pharma faces additional problems in beating placebo when it comes to psychiatric drugs. One is to accurately define the nature of mental illness. The litmus test of drug efficacy in antidepressant trials is a questionnaire called the Hamilton Depression Rating Scale. The HAM-D was created nearly 50 years ago based on a study of major depressive disorder in patients confined to asylums. Few trial volunteers now suffer from that level of illness. In fact, many experts are starting to wonder if what drug companies now call depression is even the same disease that the HAM-D was designed to diagnose."

Here's a different way to view the same evidence. Most psychiatrists were very impressed with the first generation of drugs (tranquillizers, antidepressants) that appeared in the 1950s. These drugs clinically appeared to make a huge, indeed unprecedented difference in the condition of seriously ill patients, many of whom in those days were hospitalized for their serious mental disorders. The next generation of meds (SSRI antidepressants of the Prozac family; atypical antipsychotics such as Risperdal) simply did not seem to have anywhere near as much oomph as their older counterparts. Had they been tried against patients with the same severity of illness as the older drugs had been used to treat in the 50s, they would probably have flopped. But two things had happened in the meanwhile. First, we stopped throwing the mentally ill into hospitals and now treated a lot of disorders as outpatients. (That was both good and bad and too long a story to tell here.) Even more important, the drug companies realized that so long as they sold antidepressants only for people as seriously depressed as the ones treated in the 1950s--the ones, incidentally, the HAM-D was designed to diagnose--they would be stuck with low sales volumes. So they had to find ways to convince physicians, especially primary care physicians, that somebody who felt down for a few weeks really had depression and really needed to be treated with medications. Only then would sales reach their full potential. So no wonder HAM-D doesn't seem any longer to demarcate the population of "depressed" patients--we have decided that the entire northern hemisphere is depressed and needs meds. And since most of the northern hemisphere has such a mild case of depression that it would get better if you gave out sugar pills, it's not at all surprising that the newer antidepressants don't work any better than sugar pills. In short, don't blame a revved-up placebo effect for the industry's bald-faced marketing ploy of disease mongering.

Another example: "As a psychiatrist, [Merck VP Dr. William] Potter knew that some patients really do seem to get healthier for reasons that have more to do with a doctor's empathy than with the contents of a pill. But it baffled him that drugs he'd been prescribing for years seemed to be struggling to prove their effectiveness." Memo to Dr. Potter: Been there, done that. Think of the physicians who began practice in the 1920s and lived into the 1950s. When the era of the double-blind randomized trial came along after WWII, hardly any of the drugs that were routinely prescribed back in the 1920s were able to pass muster as being superior to placebo. That did not stop the docs of the 20s from prescribing them by the bushel, or from believing in their heart of hearts that the drugs worked just great. Or take more recent double-blind controlled trials of surgical interventions like knee arthroscopy for arthritis and injection of cement for vertebral compression fractures--surgeons and patients alike swear these are extremely valuable procedures, but when you compare them to the sham procedure, they don't work any better.

In summary, one take on the apparent increase in the power of the placebo is that this is a real problem and needs to be addressed lest potentially useful new drugs be discarded. And as it's been historically so difficult to get funding to study the placebo effect in any systematic way, people like me are not likely to complain if the NIH and the drug industry are actually teaming up to pay for this new research into the scope and mechanisms of the placebo effect. (As galling as it may be to read that my own heroes of placebo research, such as Ted Kaptchuk of Harvard and Fabrizio Benedetti of Turin, are now on the drug companies' payrolls.) But another take that seems to be just as plausible is that we have simply run out of really effective and safe new drugs, and the industry stupidly bought into an industrial, assembly-line model of drug discovery and imagined they could keep blockbuster drugs rolling out of their "pipeline" into eternity. They forgot that the reason the post-WWII era was a time of huge advances in pharmacotherapy was because of an older research model that was ascendent between 1900 and 1950, and that made huge breakthroughs in our basic understanding of the mechanisms of disease--without which breakthroughs the blockbuster drugs would never have happened. Today, lacking any comparable commitment to the basic research that unlocks really novel understandings of disease, we should hardly be surprised that the drugs dribbling off the pipeline are mostly me-too drugs. The industry thought it could do its research on the cheap, discovering new drug molecules by an assembly-line process, while at the same time vastly expanding its market by calling a whole lot of people who used to be normal "diseased." That's the business model that was followed for the past couple of decades and its chickens are now coming home to roost.

Meanwhile, how likely are we to learn a lot about the underlying mechanisms of the placebo response--a hugely important question in medical practice? It's hugely important because of a line of research by Benedetti and colleagues in Turin, that Silberman alludes to only briefly at the end of his article. The Turin group has recently focused a lot of attention on studying the placebo effect via a new method--open vs. hidden administration of drugs. Example: patients after surgery keep regular pain-level diaries. All get IV injections of narcotic painkillers at intervals. Half of them get "open" injections--a doctor or nurse walks in, says "here's your pain shot" and visibly injects something into the IV tubing. Half get a "hidden" administration-- a computer-timed pump behind a screen injects the medication into the IV line without the patient's knowledge.

In virtually all the trials done in this manner so far, the results are that the open administration of the drug produces about twice as great an effect as the hidden adminstration. Maybe you had better be sitting down when you seriously ponder that concept--that perhaps half of the power of all the drugs that we currently consume is due to the placebo effect, our positive expectations of relief based on knowing that we are getting the drug, and not on the purely chemical properties of the drug. Of course we don't know that this is true for all drugs because only a few have been subjected to the open vs. hidden method. But hopefully you can now see why I say that it is huge that we better understand what is going on here.

So I ask whether the right way to find out what is going on in this hugely important area of medicine is what Silberman characterized as follows (repeated from previous post): "Under the auspices of the NIH, Potter and his colleagues are acquiring decades of trial data—including blood and DNA samples—to determine which variables are responsible for the apparent rise in the placebo effect. Merck, Lilly, Pfizer, AstraZeneca, GlaxoSmithKline, Sanofi-Aventis, Johnson & Johnson, and other major firms are funding the study, and the process of scrubbing volunteers' names and other personal information from the database is about to begin.
In typically secretive industry fashion, the existence of the project itself is being kept under wraps. NIH staffers are willing to talk about it only anonymously, concerned about offending the companies paying for it."

If there's this much secrecy involved, how can we be sure that the object of the research is really to understand better how the placebo response works--as opposed to simply figuring out better ways to cook the books when you do a clinical trial to make the drug appear to be better than placebo, no matter how good the placebo turns out to be? (Such as present-day trials designed with placebo run-in periods aimed at eliminating from the study all subjects who seem to be excellent placebo responders?)

If you say, but of course we can trust the drug companies to do what's scientifically correct, I remind you of another passage in Silberman's article: "Thinking that something crucial may have been overlooked, Potter tapped an IT geek named David DeBrota to help him comb through the Lilly database of published and unpublished trials—including those that the company had kept secret because of high placebo response." Omigosh! A big drug company suppressing data because the results didn't come out the way the marketers wanted? What a shock to all readers of this blog...

Colloca L, Lopiano L, Lanotte L, Benedetti F. Overt versus covert treatment for pain, anxiety, and Parkinson’s disease. Lancet Neurol 3:674-84, 2004.