- Author/Harry Collins, Trevor Pinch (Harry Collins, Trevor Pinch)
- Translator/Li Shangren
Since at least the 1950s, modern medicine has believed that the placebo effect has been established in science. Studies have shown that if a placebo is given, approximately 20% to 70% of patients seem to benefit from it.
Perhaps the most surprising of these is the placebo surgery: the patient is appropriately anesthetized and the skin is cut, but there is actually no meaningful operation; according to reports, such an operation is highly effective.
Sometimes fake surgery even seems to be more effective than real surgery. For example, it seems to be effective for certain types of chest pain and back pain. Studies in the mid-1990s showed that this is effective for knee arthritis; only cutting the patient’s knees has the same therapeutic effect as those of knee joints by scratching and washing; and the latter is generally considered to be the height of knee arthritis Effective standard treatment.
Unfortunately, these seemingly succinct findings are still controversial. Now we must pass through another Hall of Mirrors that is more distorted:People who are unwell may heal even if they don’t receive any treatment, And patients receiving placebo treatment and patients receiving a large number of medical interventions are also likely to recover on their own at roughly the same rate.
In other words, patients receiving placebo treatment may not be improved due to the placebo effect, but healed spontaneously; and medical treatment is also ineffective, and patients receiving surgical treatment actually healed spontaneously.
In this case, it’s not that the placebo effect is as good as real surgery, butThe placebo effect is no better than real surgery, and both are equally ineffective.
Experiment to confirm the “placebo effect”
To understand whether there is a placebo effect, another experiment must be done: compare a group that received a placebo with another group that did not receive a placebo at all. In this case, if there is a placebo effect, the treatment status of patients receiving placebo must be better than that of patients not receiving treatment.
Two Danish physicians (Hrobjartsson and Gotzsche) collected research papers comparing placebo-treated and untreated patients, and analyzed them in 2001. Among the 114 trials, only a few were directly designed to test placebos; most of the remaining conditions were that doctors examined three groups of patients: patients receiving medical treatment, patients receiving placebo treatment, and none at all. Patients receiving treatment. They found that in terms of improvement in treatment conditions,There was no significant difference between the patients who received placebo and the patients who did not receive treatment.。
This sounds like a decisive study, and the Danish physician’s report seems very convincing at first glance. The number of studies they analyzed and the number of patients are very large. This research seems to overturn a major preconception.But if you carefully examine the cautious proviso at the end of the paper, you will find thatThe conclusion is not so unbreakable。
First of all, the data shows that placebo has a small effect on the experience of pain. In addition, placebo may have a considerable effect on a small number of patients or certain diseases, although not all patients or all diseases have an effect. The statistical analysis methods used by Danish researchers can easily cover up these minor effects and a small number of diseases and patients. What is even more worrying is the complex logic discussed below. To illustrate this logic, more and more exclamation marks must be used at the end of sentences.
Whether it’s a placebo or other therapies, it’sCan’t compare blindly with untreated conditions! Both the patient and the therapist will know who is not receiving treatment; in fact, the fact that there is no treatment cannot be concealed, otherwise it is not “not receiving treatment” but receiving a placebo.
Now things get more complicated
If physicians and patients know who is not receiving treatment, we would expect this to bring expected and report effects; if placebo is effective, we would expect the difference between placebo and untreated patients to be more significant! In other words, patients who have not received treatment should feel pessimistic about their prospects, and those who perform treatment should expect that there will be no improvement in this group of patients; therefore, we will think,Regardless of the person performing the treatment or the person receiving the treatment, there will be a strong report effect, and it is expected that the effect will strengthen both.
All in all, even if there is no placebo effect, in these non-blind experiments, because there is no negative reporting bias and expected effect of the treatment group, the experimental results should see a placebo effect.In this wonderland of Alice in sleep, this should be a kind ofAn experiment that will never fail! Regardless of whether there is a placebo effect, the result should always look like a placebo effect! !
The results of these experiments now do not have a significant placebo effect, which means that there are no expected or reported effects in these experiments, which showsThere must be something wrong with these experiments! ! ! Just like Mendel’s famous experiment on hereditary characteristics, the results of the experiment were so beautiful that it seemed to be a fake!
In response to these doubts, the Danish researchers argued that since most experiments have three groups of patients instead of two groups, patients and analysts are not interested in the difference between the placebo group and the untreated group, and this may be Reduce the reporting effect and expectation effect. However, this argument still seems weak.
In any case, even if the lack of expectation and reporting effects are not the decisive factors, there are other reasons why we do not trust the conclusions of this study. As mentioned earlier, the group of patients who are not receiving treatment will inevitably know that they are not receiving treatment.If their disease is serious, then they may feel that since they did not receive any treatment in this study, they wouldDecide on your own to seek other treatments in ways unrelated to this research(See similar claims about vitamin C testing in Chapter 4). This does not apply to the placebo group, because patients in this group think they are receiving treatment. The result is,The difference caused by whether there is self-treatment may cause the success rate of different groups to be not much different。
Unexecuted double-blind experiment
Considering the above two arguments against the conclusions of the Danish researchers, we don’t know what position to take. This situation often arises in difficult statistical sciences. We only knowDon’t take the placebo effect for granted as in the past, But it is still difficult to be sure that it does not exist.To solve this problem, a double-blind experiment must be conducted between the placebo group and the non-treatment group. However, thisImpossible by definition(I have to add an exclamation mark at the end of this sentence)!
Despite these academic debates, pharmaceutical companies and drug trial executives and even critics of pharmaceutical companies believe that the placebo effect is real. Critics point out that the so-called double-blind is usually impossible to implement, because if the drug has side effects such as dizziness or dry mouth, patients can often guess whether they are taking the real drug or a placebo. This means that even if the drug outperforms the placebo effect in a randomized controlled trial, it may just be because the drug has side effects and a stronger placebo effect!
Pharmaceutical companies and the units that perform trials for them attach great importance to the authenticity of the placebo effect. In fact, they even evaluate the sensitivity of the patients under the trial to the placebo effect, trying to exclude those patients who are susceptible to hints (hidden mentality). Treatment) and so on. We can say this: in terms of how the placebo effect affects our thinking about medicine, it is true.
——This article is excerpted from “The Tension of Medicine”, September 2021, Left Bank Culture.
