New research on placebo responses improves our understanding of drug efficacy
Placebo is in the news again, with a feature article in the Wall Street Journal,occasioned by new studies of immune responses in mice and neurological responses in humans. Although the WSJpiece does not mention antidepressants, some of the research it reviews has implications for our understanding of how well they work. The bottom line is the reverse of what you might imagine: Antidepressants are likely to be more effective than our current estimates suggest.
One recent study concerns “conditioned responses.” If you give patients with Parkinson’s disease a placebo, it does nothing for them. However, if you first expose them to an effective medicine and it helps—making their wrist muscles more limber—then you can substitute a placebo, and the patients will enjoy continued benefit. Their nerve cells will show responses that mimic the action of the drug. More frequent prior exposure produces stronger and longer placebo responses on both levels, clinical and neuronal.
The cause and function of conditioned responses remain a mystery. Some Darwinists think that when an animal’s brain knows that help is on the way, it is efficient to get the process of recovery started even before that help arrives.
Prior research suggest that antidepressants, too, produce conditioned responses—again, in the person and in nerve circuits. Wherever placebo responses are conditioned responses, drugs become hard to evaluate. Say that a depressed patient takes an antidepressant and does well. When, in a subsequent depressive episode, she enters a drug trial, she may be sensitized to respond to similar-looking pills, especially in the early going. If antidepressants are especially effective and widely used, placebo responses in testing will run high, and new medications will have difficulty outperforming dummy pills. The better antidepressants work, the more trouble they will have demonstrating their worth in formal trials.
In my new book, Ordinarily Well, I focus on a different problem in antidepressant assessment. Participants in drug trials receive extraordinary levels of social support—the equivalent of low-grade psychotherapy. Again, the result is inflated levels of improvement in participants on placebo. Because antidepressants obviate some of what psychotherapy does—because when psychotherapy is added to medication patients only get a modest bump up in response, and not a full psychotherapy effect—the standard analysis of drug trials will produce an underestimate of the medications’ potency. We’re taking the improvement of patients on medication and subtracting out a benefit (from the low-grade psychotherapy) that they did not enjoy, at least not to its full extent.
This distortion comes under the heading “additivity.” I explain it in slow motion in Ordinarily Well—and somewhat more quickly in a recent book review. In that same review, I discuss prior research on placebo effects in Parkinsonism.
For decades now, researchers in Parkinson’s disease have been aware that their drug efficacy calculations constitute underestimates, because conditioned responses cause placebo effects to run high. The problem is not with the drugs. Conditioned responses are a biological tribute to them. The problem is with the testing.
I am not a fan of the belief that depression is especially responsive to classic placebo effects, that is, improvement grounded in expectations arising from the mere fact of pill taking. My impression is that the whole set-up of drug trials—the support they provide—plays the greater role. But either way, improvement in patients in the placebo arm of drug trials will have its equivalent in brain physiology.
The new research on placebo is beginning to elucidate those biological underpinnings. They may turn out to be varied. Different circumstances will cause the brain to rev up the immune system, release chemicals that restore motion to neurologically impaired patients, or allow the formation of new nerve cells. In that case, there will be not one “placebo effect,” but many, along with many medical conditions for which, sadly, placebo responsiveness plays no substantial role.
But the placebo response need not be a rebuke to pharmacology. Quite the contrary. It can be an indicator that drugs work. Stronger medicines make for “stronger” placebos. What looks like a placebo effect really is a drug effect—only one that, after repeated exposure to medication, can be triggered, for a while, by a dummy pill.
No one believes that drugs used to treat Parkinson’s disease are “placebos with side effects.” That label is no more appropriate when it comes to antidepressants. Instead, often we might understand the placebo response itself to be a side effect—an incidental product of medication’s potent action on nerve cells in the brain.
https://www.psychologytoday.com/blog/in-practice/201607/stronger-drugs-stronger-placebos
No comments:
Post a Comment