Placebo 2 - a myth?

Amidst the great enthusiasm (or great fear) of the placebo effect, it seems that another significant bias was forgotten: how do we know if the improvement in the condition of each of the groups is not due to a natural recovery or a natural temporary improvement in the disease?

in the previous post We started with the understanding that comparison treatment recipients for such who do not receive treatment It is problematic, since the improvement in the condition of the treatment recipients may be largely attributed to the placebo effect of the treatment and not to the treatment itself.

But amid the great enthusiasm (or great fear) of the placebo effect, it seems that another significant bias has been forgotten: how do we know if the improvement in the condition of each of the groups is not due from natural recovery or natural temporary improvement in the disease? The comparison with a group that does not receive any treatment adds extremely important information!
Consider the following situation: Let's say we are testing a new medicine against a runny nose. The experimental group receives the drug and their runny nose goes away within a week. The control group receives a placebo, and their runny nose goes away within... a week. Are we witnessing an amazing placebo effect? Or maybe even a group of people who would not have received any treatment would have recovered within a week? At least in the case of a runny nose, this is the case.

In other words, to get a complete picture we must perform an experiment that contains three Groups: those who accept Real care, those who receive Sham treatment And such who do not receive treatment at all. According to the differences between the recovery rates in the different groups, it is possible to know the size of the placebo effect and the relative advantage of the tested treatment over the placebo.
The following diagram illustrates the state of things (the recovery rates are just an example of course. The numbers are different in each case):

A triumphant blow to the placebo effect
When the token fell, which due to enthusiasm for the placebo effect was "forgotten" on the comparison side with people who did not receive treatment at all, the researchers went back to check past conclusions regarding the strength of the effect.

Kienle and Kiene We revisited Beecher's seminal 1955 paper, the one that effectively founded modern placebo research. They examined the data on which he relied when he came to his conclusion about the tremendous power of the placebo, and concluded that There is no evidence of any placebo effect in any of the studies he based his article on! The improvements in the health of the placebo recipients was real, but due A host of other factors We will detail them shortly.
to find out more about the situation Compiled by Danish researchers Asbjørn Hróbjartsson and Peter C Gøtzsche About 200 controlled studies in 60 different treatment areas, in which a control group was also included in the experiment did not receive treatment.
Overall No significant advantage was found for the placebo treatments compared to those who received no treatment at all!
in a press interview Says Hróbjartsson: "In our opinion, the high levels of placebo effect that have been reported so many times in many articles are the result of flawed research processes."

Allegations of research failure of the same type were also made against the sensational findings regarding the tremendous placebo effect in the field of antidepressants. The claim is that these conclusions were based mainly on comparisons between drug recipients and placebo recipients, but not between placebo recipients and those who receive no treatment at all, that is, in fact what the studies Kirsch referred to show is Both the placebo and the drug have little effect, And that the depressive states mostly pass by themselves (an equally depressing conclusion regarding the antidepressants).

Beyond the misinterpretation regarding the power of the placebo, Kirsch's works suffer from many other methodological flaws, and are far from being objective. The antidepressants seem to help after all. They may not be as effective as the drug companies would have us believe, but they are certainly more effective than Kirsch would have us believe. Those interested in more details are welcome to read HERE, HERE וHERE for example.

And what about the knee surgeries and the thoracic artery ligation surgery? What is the correct conclusion in these cases? the mother The placebo is amazingly powerful? Or the surgery is simple Amazingly useless And is this a natural recovery in both cases?
Only a few surgical procedures stand up to scientific scrutiny (at least in the US). Who knows how many other useless surgical procedures are practiced today...

Neither treatment nor placebo
This is the place to mention a host of possible reasons for receiving a report of an improvement in the medical condition, reasons that do not stem from any treatment, nor from the placebo effect, at least not in its accepted definition.

A natural improvement in the condition
Original diagnosis is incorrect - A certain percentage of those who are diagnosed with some medical problem, are diagnosed that way accidentally, meaning that they are healthy people. If such people receive any treatment, and upon further examination it turns out that they are healthy (how surprising), it is easy to make a mistake and attribute their "recovery" to the treatment.

natural recovery - Our body has highly sophisticated systems for self-repair and healing. He does this constantly, usually without our knowledge, and in most cases he will manage to deal with the problem even without external help. It is easy to mistake and attribute a beneficial effect to treatment (real or simulated), in cases where the person is naturally healthy.

Natural fluctuations in the disease / What Comes Down Must Go Up - The symptoms of many chronic diseases worsen and improve in waves. Back pain can disappear for a few months and then reappear. Joint pains, migraines - all these are usually not constant. Naturally, people turn to treatment when the suffering is maximum, that is, during the low period of the disease. This will also be the time when they will tend to apply and participate in trials of new drugs. andFrom a low point the situation can only get better. Again, it is easy to make a mistake and attribute the improvement in the condition to the treatment (real or sham) given at the same time, although this is not the case.

The effect of the subject's knowledge that he is part of an experiment

Another interesting family of biases arises from the very fact that the person knows that he is participating in research. This can be expressed in several forms.

Changing life habits - The subjects may be more careful about nutrition, increase physical activity, be careful about taking other medications more obediently, etc. These factors may improve their health regardless of the specific treatment they receive during the trial.

The knowledge that they are being watched - She alone may bring about changes. The phenomenon It was first discovered at the Hawthorne plant of the Western Electric company in the United States in an experiment conducted at the end of the twenties of the last century. The researchers examined which changes should be made in the work environment in order to increase employee productivity. In one experiment they discovered that Increasing Light intensity increases productivity. In another experiment they discovered that Reduction The intensity of the light causes an increase in output. The confused researchers continued to test many parameters, and finally came to the conclusion that There is actually no importance to the change they will make. The very fact that the workers in the factory knew that they were participating in an experiment and were being monitored in some way led to an increase in productivity.

Desire to please the therapist/researcher - Sometimes subjects try to please the researcher, because of appreciation, reverence, a general desire to help, to be liked, etc. When such a subject is asked if there has been an improvement in his condition, he may answer in the affirmative, even though in fact there has been no real improvement. Usually the questions are not yes/no questions, and then it is even easier to "push" the assessment up one grade, just to please. And the result - both the real treatment and the sham treatment accumulate a larger gap compared to those who did not receive treatment (or compared to patients who will receive treatment under normal life circumstances, and not as part of a study).

Unlike the first biases we mentioned, which can be offset if compared to a group of people who received no treatment at all, this bias cannot be offset that way. It exists by the very existence of the study.

The motivation to feel better - selective attention and interpretation - People who are aware that they are receiving treatment (real or simulated) may pay more attention and remember moments of improvement, and pay less attention and forget less successful moments. Even motivated therapists can contribute a lot to the illusion that the treatment is indeed successful. "You look much better than the previous visit", "I think there is an improvement, I don't remember the pain appearing so much in the last week", etc. Is this a legitimate part of what we used to call the placebo effect? I think this is a borderline case. It is not an objective improvement in the situation, nor a subjective improvement, but In selective memory and biased interpretation regarding the success of the treatment.

methodological biases
Sometimes we tend to forget this, but quite a few studies are not planned correctly, are not performed correctly, their results are not analyzed correctly, and the conclusions do not always follow from the results. Not all studies are equally reliable.

But the story is not over
Do these factors and biases explain the כל The placebo effect as documented for many years? Was all the hype surrounding the placebo effect unfounded? As we have seen, some believe that this is the case.
And some don't. Quite a few researchers have devoted their energy to learning about the placebo effect through experiments That was their goal, and not as Lui's conclusions from clinical trials aimed at testing the effectiveness of new drugs and treatments.
And they discovered interesting things. About that in the next section.

____________________________________________________________________

Want to receive an email update every time I post something new? Sign up at the top right (you can always cancel).
Think others might be interested? Send them the article or share on Facebook!