A multicolored collection of pills is shown.

Placebo Controls

Here and there, Americans are looking for reasons to swallow hydroxychloroquine to prevent COVID-19, often brandishing studies that show positive effects of the drug. Dr. Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases, has so far rejected each of the studies as failing to meet the requirement for a “randomized, placebo-controlled” study.

Such an expectation is not a captious whine but an expectation generated by the principles of scientific research, honored by psychologists as well as immunologists. Is it clear why it’s necessary for evaluating the effectiveness of drugs and some other treatments?

Placebo treatments are often likened to sugar pills, though they include many inert substances and often include the situation in which they are administered. The notion of comparing the effects of an inert pill, the placebo, with the effects of a new drug is an effort to find out whether a new drug accomplishes more than a supposedly empty placebo procedure would. Drugs in safe doses that produce a significantly larger benefit than whatever the placebo does are more impressive than drugs that succeed only by themselves, without comparison to the effects of just taking an inert pill.

The practice of using control groups of subjects who receive placebo but not the supposedly effective drug got its  name from the wailing women of the late middle ages.

Drugs that alter scores in mental and physical testing may or may not exert placebo effects, but when a placebo effect does occur it can be very large, as in the case of some antidepressants and cough syrups. Painkillers, too: The problem with one of the most common such over-the-counter drugs, acetaminophen (Tylenol), is that its treatment or “real” effect (efficacy) is weak and its placebo effect is large. The list goes on.

Though a treatment drug and a placebo might be acting via different mechanisms, a prescription drug that produces the same effect as a placebo is not marketable—except maybe in the case of hydroxychloroquine, depleting the supply intended for antimalarial or immunoprotective use.

Though antidepressants approved by the FDA are available only by prescription, over-the-counter cough syrups are readily available. People sometimes swear by them, but any effects specific to coughing are only placebo effects.

Do not believe that placebos are simply the absence of a systematic treatment, such as an effective drug provides. Placebos can have striking and unexpected effects of their own!

If you want to be sure your drug is actually a medicine, you will need to examine the performance of placebo controls in drug testing*, keeping in mind that the “real” pills often have a placebo effect of their own. Antidepressants are famous for this, though the Begley article may take an extreme position. Drug company influence has insured that the FDA will not require that manufacturers publish the total therapeutic effect of their drugs but only the difference between their pills and a placebo control. That is, the effect that is common to commercial drugs and placebos is commonly not measured. It might be harder to market pills that were known to be mainly placebos, though placebos work even when we know they’re placebos (opening an interesting angle).

Of course placebos work only for some people. Sometimes they’re effective for many, but the effect depends on the disorder, the placebo chosen, the patient, and the context of using the drug. The same is true for effective drugs. For example, some folks feel positively loopy on a moderate dose of codeine, while others (about 7 percent of the population) lack the liver enzyme that converts codeine into morphine and get no relief from pain at all.

* Is it worth adding a reminder that we can find out what effects to expect from medicine by reading the package insert?

I suppose it’s a peculiar habit to read package inserts, but I read them. Helpfully, the FDA has put them all online. The package inserts assume the reader will be familiar with how drugs work, including how they affect the brain. But anyone who can understand the package insert is a good bet to understand the label.

A medical professional in a white coat is shown with a stethoscope.

BIO: Understanding the placebo effect depends on understanding how it is measured (including double-blind technique) and what the mechanisms are, right down to the molecular level, to genes. It should be no surprise that placebo effects can be measured as both mental and physical changes. This has been a long trek with a lot of controversy, especially when it comes to antidepressants.

The placebo effect is probably not a single, unitary process. The brain circuits that account for placebo effects are distributed through the brain. To make things harder, there are “fake” placebo effects like regression to the mean–apparent drug successes not due either to a placebo or to a drug treatment.

Since placebo effects can slip by undetected in behavioral studies even when psychologists use what they think are appropriate controls, the technicalities become an unavoidable issue (research here).

It’s getting hard to tell what’s mental and what’s physical, and that’s good. When we say that a placebo effect is “all in the head”, that means “in the brain“. That is a physical effect, as are changes in the immune system and perhaps also fatigue caused by placebos.

A cough is a physical reflex. Depression is a mood disorder. Both are alleviated by drugs with large placebo effects. An enumeration of coughs will list physical varieties, but there are others of psychological origin, which can include presumed stress or “body” and “mind” in tandem.

Major depression is related to stress as well. Asking whether antidepressant and cough-suppressing drugs treat the body or the mind is a fruitless question.

PSYCHO: The comment by Dr. Fauci that opened this topic can be misunderstood. A placebo-controlled study does not measure the true size of a placebo effect. It only documents the presence of a placebo effect and measures its magnitude relative to the “real” or treatment drug.

To measure this kind of difference, you have to administer a physical treatment agent to one group of subjects and perform all the same procedures except giving the treatment agent to another group, and then measure the difference between them.

It’s common to imagine a placebo effect as an extra effect added on top of the specific treatment effect, but this additive, non-interactive model is almost certainly incorrect.

Some procedures may produce effects that are entirely placebo. It’s hard to be sure, but this is an excellent topic for getting into experimental design beyond snorkel depth.

To measure the size of a placebo effect you must compare it not to the total effect of the treatment drug but to a test trial with no treatment or placebo at all. Such a procedure is called a no-treatment control. This allows the experimenter to assess the size of the placebo effect against behavior with no intervention.

A recent placebo-controlled study of the success of hydroxychloroquine in fighting Covid-19 found that the drug is not useful against the disease. The placebo used was folate (a B-vitamin) tablets, which resembled the hydroxychloroquine pills in appearance.

Placebo effects are not just about drugs, but about personality, context and expectations as well. It’s surprising how powerful our beliefs can be in coping with problems as diverse as back pain and bereavement. Downtown engineers are willing to prescribe mechanical placebos, too. We might even add the Perkins Tractor.

Placebos work mostly through our expectations and past conditioning. Similar effects can take us beyond our normal physical limits. To those who point out that we are reaching the limits that our evolution has set, we may reply that our mental limits are still uncertain.

When we say that a placebo effect is “all in the head”, that means “in the brain“. That is a physical effect, as are changes in the immune system and perhaps also fatigue caused by placebos. In fact, placebos may be a greater help with Parkinson’s disease than with Alzheimer’s. Some sham surgery is as effective as real surgery. It’s getting hard to tell what’s mental and what’s physical.

Placebos work mostly through our expectations and past conditioning. Similar effects can take us beyond our normal physical limits. To those who point out that we are reaching the limits that our evolution has set, we may reply that our mental limits are still uncertain.

Just don’t believe everything you read about this.

SOCIAL: It would not be surprising if you found most of this post to be unsurprising. Placebo effects have become widely known and, in fact, larger and more effective. As they become more pervasive, placebo effects, particularly in our contacts with health professionals, where placebo effects are on their way to becoming a recognized component of health care, and extending to other social interactions as well. 

Leave a Comment

Your email address will not be published. Required fields are marked *