The placebo effect is the effect that arises from taking a placebo instead of an active ingredient. Placebo effects can take many forms and physiological effects are normal. They are just not caused by the active ingredient to be tested.
Undermining the Placebo Effect
- Thread starter scepticat
- Start date
-
- Tags
- placebo effect
Acleron
Master Poster
- Joined
- Oct 15, 2007
- Messages
- 2,290
Lets be clear here.
The placebo effect is the effect when a placebo trial, preferably double blind, is carried out and those who have taken the placebo show signs of improvement. No mechanism is supposed in this effect, it is just data that shows that people who are untreated and unaffected by the protocol of the trial show improvement. This is just a fact of life, it cannot be undermined unless someone is prepared to show it does not happen.
Even if a mechanism discloses why it happens, it is still a placebo effect.
The placebo effect is the effect when a placebo trial, preferably double blind, is carried out and those who have taken the placebo show signs of improvement. No mechanism is supposed in this effect, it is just data that shows that people who are untreated and unaffected by the protocol of the trial show improvement. This is just a fact of life, it cannot be undermined unless someone is prepared to show it does not happen.
Even if a mechanism discloses why it happens, it is still a placebo effect.
blutoski
Penultimate Amazing
- Joined
- Jan 10, 2006
- Messages
- 12,454
I'm just finishing off a presentation on this. I'll have it posted in a few hours but thought it would be valuable to contribute this: the expression "placebo effect" has two meanings, unfortunately, and they get confused.
1. the healing rate of the placebo group. eg: 30% of those on placebo got better, so we'd say that the placebo effect cured 30%. This is the situation Acleron is discussing.
2. the difference between the healing rates of the placebo group versus the non-treatment group, which suggests an effect from placebo itself. eg: if 30% of the placebo group got better, and 25% of the non-treatment group got better, then the placebo had 5% curing effect.
It is the conflation of these two definitions that have led to the urban myth that the 30% average recovery rate for placebo groups can be attributed to taking the placebo. In fact, there is very little evidence that placebo groups do better than non-treatment groups, and what evidence there is is isolated to conditions that require subjective reporting, and cannot be distinguished from reporting bias.
My impression is that your GP used the term in neither of the above definitions, and frankly, he used the wrong term to describe what he was thinking. My impression is that he figured you were just misunderstanding comparison bias. Consider: if you really did get an energy boost from drinking a stimulant, wouldn't you always expect to be 'more tired' after? Wouldn't 'normal' feel like 'tired'?
Here's an experiment you can do at home that I do to demonstrate comparison bias: prepare three bowls of water. One is hot, one is icewater, the other is lukewarm. Put both hands in the lukewarm bowl. They both feel the same. Then put one hand in the hot water and one hand in the icewater and leave them there for three minutes. Then put them in the lukewarm bowl. It will appear that they're feeling different temperatures, but you know intellectually that this sensation is merely an artefact of comparing to a previous state.
You can fall into the same trap when coming down from a stimulated episode, and misinterpret 'normal' as 'crashed'.
1. the healing rate of the placebo group. eg: 30% of those on placebo got better, so we'd say that the placebo effect cured 30%. This is the situation Acleron is discussing.
2. the difference between the healing rates of the placebo group versus the non-treatment group, which suggests an effect from placebo itself. eg: if 30% of the placebo group got better, and 25% of the non-treatment group got better, then the placebo had 5% curing effect.
It is the conflation of these two definitions that have led to the urban myth that the 30% average recovery rate for placebo groups can be attributed to taking the placebo. In fact, there is very little evidence that placebo groups do better than non-treatment groups, and what evidence there is is isolated to conditions that require subjective reporting, and cannot be distinguished from reporting bias.
My impression is that your GP used the term in neither of the above definitions, and frankly, he used the wrong term to describe what he was thinking. My impression is that he figured you were just misunderstanding comparison bias. Consider: if you really did get an energy boost from drinking a stimulant, wouldn't you always expect to be 'more tired' after? Wouldn't 'normal' feel like 'tired'?
Here's an experiment you can do at home that I do to demonstrate comparison bias: prepare three bowls of water. One is hot, one is icewater, the other is lukewarm. Put both hands in the lukewarm bowl. They both feel the same. Then put one hand in the hot water and one hand in the icewater and leave them there for three minutes. Then put them in the lukewarm bowl. It will appear that they're feeling different temperatures, but you know intellectually that this sensation is merely an artefact of comparing to a previous state.
You can fall into the same trap when coming down from a stimulated episode, and misinterpret 'normal' as 'crashed'.