1526: Placebo Blocker

Explain xkcd: It's 'cause you're dumb.
Jump to: navigation, search
Placebo Blocker
They work even better if you take them with our experimental placebo booster, which I keep in the same bottle.
Title text: They work even better if you take them with our experimental placebo booster, which I keep in the same bottle.


This comic is a joke about the difficulty of testing a drug that is supposed to block the placebo effect.

A placebo experiment is used for testing a drug candidate. It has two groups: one that gets a real drug candidate, and one that gets a fake. The placebo effect describes the observation that the group that gets the fake often show signs of having received a working drug - though commonly weaker than in the group that gets an effective real drug.

Cueball states to Hairbun, with a citation from the real world, that his team created a Placebo Blocker, a drug designed to prevent the placebo effect. Cueball begins to design a test for this new drug. Following typical experimental design, patients would be split into two groups: a control group, and the group that receives the treatment.

Cueball knows that the treatment given to the control group is supposed to be designed so that it is not influenced by the variable trying to be isolated. As the placebo effect is the effect under investigation, a placebo can not be used as a control treatment as a comparison with a placebo blocker. Cueball tries to design around this. In his test, both groups would receive a placebo as a treatment for an unspecified condition (the Treatment Placebo); in addition the test group would receive the Placebo Blocker drug, while the control group would get a placebo pill instead (the Placebo-Blocker Placebo). If this works as expected, the Treatment Placebo would be blocked by the Placebo Blocker in the test group, while in the control group, the Placebo-Blocker Placebo may have a placebo effect in blocking the placebo effect of the Treatment Placebo, and the difference between these effects can be measured to test the effectiveness of the Placebo Blocker.

Cueball and Hairbun think about this trial until they both develop headache from frustration. Cueball then kindly offers Hairbun a sugar pill. While this might have helped cure the headache via the placebo effect had he told her it was a headache treatment, by revealing the pill as merely a sugar pill, it may reduce the effect (though it has been shown that placebos tend to work even if the subject is aware that they are placebos).

In the title text, Cueball mentions that his sugar pills against headache works even better together with the new experimental placebo boosters. Incidentally, he indicates that he keeps those in the same bottle with his sugar pills. Assuming someone believes placebo boosters are in the jar this would allow them to take the sugar pills and receive a greater placebo effect, as the placebo effect is based upon faith in the treatment, regardless of whether there are placebo boosters in the jar.

It is possible but unlikely that:

  • Cueball's sugar pills are, in fact, the Placebo Blockers themselves and that, seeing Hairbun has a headache, Cueball is inspired to somehow use the opportunity as an experiment to test the Blockers
  • Cueball is suggesting Hairbun take a "placebo booster" which is really a "placebo blocker", thus testing the blocker he mentioned earlier in the comic.

Questionable neuroscience research is also discussed in 1453: fMRI.


The placebo effect refers to the phenomenon in which patients given an inactive treatment such as a sugar pill can still show improvement relative to an untreated patient. The placebo effect is thus very important to consider when testing new drugs, since even ineffective drugs can have a positive effect on the patients due to the placebo effect. Modern drug experiments are hence conducted as double-blind trials, where the patients are randomly given either the treatment or a placebo without either they or the administering doctors knowing who receives the new drug and who received the placebo pill. (It is important that the doctor does not know, as if they did, it may affect the way they interact with the patient.)

Generally the patients need to believe that they are receiving an active treatment, but one study showed that the effect can occur even if the patients are told that they are receiving a placebo pill. The key factor seems to be that the patients must believe that a positive effect will occur. For example, (1) patients experience a greater effect if they believe that the treatment is expensive and (2) patients who know that they have not been given an active treatment will experience the effect if they are told that placebos can have a positive effect through the power of the mind. Furthermore, the placebo can increase the effectiveness of treatments which seem larger (this is why over-the-counter pain medication is often administered as two half-doses rather than just one full dose).

Several reasons for the placebo effect have been proposed, from study artifacts - such as under-reporting of negative outcomes by patients who think they are being treated, to neurological explanations for how mental state can translate into physical outcomes.

Placebo-blockers do actually already exist. A side-effect of the opiate antagonist Naloxone is that it blocks the placebo effect.

It should be noted that placebo does not actually improve the objective condition, only the patient's subjective perception of it (i.e. the patients do not get better more than they randomly would, but the placebo makes them think they do).[actual citation needed]

Mechanisms of the placebo effect[edit]

The placebo effect is one of the greatest mysteries in modern medicine. It is typically found that the placebo effect is an effective treatment in itself in addition to the effectiveness of drugs and other treatments, and it has been found to cause small improvements to cancer outcomes. In other cases such as pain relief, the placebo effect is claimed to be comparable with the effectiveness of the drug itself - but this is a misunderstanding: this is not evidence of placebo working, but of the drug not working.

The comic refers to the recent study by Kathryn T. Hall, Joseph Loscalzo, and Ted J. Kaptchuk. (2015) Genetics and the placebo effect: the placebome. Trends in Mol Medicine. Volume 21, Issue 5, May 2015, Pages 285–294 - however, bear in mind that one has to treat studies very carefully Kaptchuk vs Placebo

It is possible to test the placebo blocker using three groups: a test group who receive a placebo and a placebo blocker, a control group who receive a placebo but no blocker, and a second control group who receive no treatment whatsoever, as a lack of treatment is the variable that an actual placebo is designed to control for. Still it might be hard to determine if the pills are having a negative effect or blocking the placebo effect, so multiple trials with multiple illnesses may have to be carried out.


[Hairbun is standing in front of Cueball who does all the talking. Below them is a footnote.]
Cueball: Some researchers* are starting to figure out the mechanism behind the placebo effect.
Cueball: We've used their work to create a new drug: A placebo effect blocker.
Footnote: * Hall et al, DOI: 10.1016/J.MOLMED.2015.02.009
[Zoom in on Cueball who now holds his arms out.]
Cueball: Now we just need to run a trial! We'll get two groups, give them both placebos, then give one the real placebo blocker, and the other a...
Cueball: ...wait.
[Hairbun holds her chin, while Cueball just stand there for a beat panel.]
[Hairbun looks again at Cueball who begins to take the lid off of a medicine bottle.]
Hairbun: ...My head hurts.
Cueball: Mine too.
Cueball: Here, want a sugar pill?

comment.png add a comment! ⋅ comment.png add a topic (use sparingly)! ⋅ Icons-mini-action refresh blue.gif refresh comments!



The title of the referenced paper introduces the 'Placebome', the collection of genes which lead to the placebo effect. This is an absolutely ridiculous word, and would be worthy of Jonathan Eisen's Worst New Omics Word Award. Quantum7 (talk) 08:31, 18 May 2015 (UTC)

Title text bottle

It seems more plausible to me that the "they" and "same bottle" in the title text refer to the sugar pills for headache. The title text would then be an organic continuation of the immediately preceding dialogue. Angew (talk) 09:01, 18 May 2015 (UTC)

Agreed. Take two sugar pils. The second will boost the effect of the first. It could work if you believe it.--Kynde (talk) 12:14, 18 May 2015 (UTC)
Agreed, updated. -- Phyzome (talk) 13:47, 18 May 2015 (UTC)

I'm confused about why this explanation is a stub. Personally, I think it explains the comic well, but I'll refrain from removing the incomplete tag in case most people think that the explanation isn't adequate. Caeleste Alarum (talk) 15:21, 18 May 2015 (UTC)

Wouldn't you find a malady that can be treated via placebo, like a headache, give control group A a headache pill, control group B a placebo and tell them it was a headache pill and give the test group a placebo blocker as the placebo and tell them it is a headache pill?

A placebo blocker would be really useful in medical testing to find out which medicines are actually effective and which are simply producing a stronger placebo effect through a noticeable side effect. 15:46, 18 May 2015 (UTC)

How to do it.

Compound two sets of placebos. The control set is just sugar pills. The other set would be the blocker. Unless the active dose is massive, it'd also be partially a sugar pill already.

Present both as a possible treatment for some malady.

Each group would then only get one pill, and be ignorant that there was potentially a placebo blocker in their dose. 17:09, 18 May 2015 (UTC)

I would guess that there would have to be at least 6 groups. Groups 1 and 2 would be the traditional experiment with a drug and a placebo, groups 3, 4, 5, and 6 are given two pills one of which they told is a drug, the other is a placebo blocker which may prevent the first drug from helping you. groups 3 and 4 are given the real placebo blocker, groups 5 and 6 are given another placebo. This would be an interesting experiment in that you could test the psychological effects of telling someone who took a real drug that "it may not work." 18:36, 18 May 2015 (UTC) Veggiet

I think we've entirely overlooked the idea of using a control group that doesn't know what the word placebo means. With such a control group, one could not tell them any lies at all "I've invented this new drug called placebo that will cure your rheumatism".Seebert (talk) 18:57, 18 May 2015 (UTC)

Why do you have to include the world placebo? Any new made up name like turgidax or pragmanol or frogans or vulpix or bligdrine will work. What is wrong with saying to the patient in the control group,"You are testing a pill of bligdrine." The patient may or may not know what placebo is, but they will certainly be not aware of bligdrine. 20:21, 18 May 2015 (UTC)BK201

I once removed the last paragraph regarding speculations on the title text as of what could happen if there were two different pills in the bottle. But it is to me clear that there is just sugar pills, as is already explained above. And thus this paragraph is overkill. But it was inserted again after I deleted it. I vote for it to be deleted, but will let someone else do this, as not to make it a personal edit war... --Kynde (talk) 16:05, 19 May 2015 (UTC)

Here's an experiment done all the way back in 1978: Levine, J.D., Gordon, N.C., and Fields, H.L. (1978). The mechanism of placebo analgesia. Lancet 2: 654-657. Summary: The researchers recruited volunteers who were undergoing tooth extraction. After surgery, one group received morphine-based drugs (most, but not all, reported pain relief). Others received placebo (about a third reported pain relief). Others yet were given placebo AND naloxone, a drug that blocks the action of opioids (none reported pain relief). The researchers concluded that administration of placebo caused the release of endogenous opioids in some patients, and naloxone worked as a placebo blocker. More recent research with functional brain imaging has confirmed that opioids and placebos activate the same brain regions (Petrovic, P., Kalso, E., Petersson, K.M., and Ingvar, M. (2002). Placebo and opioid analgesia imaging--A shared neuronal network. Science 295: 1737-1740).CLSI (talk) 17:07, 19 May 2015 (UTC)

How to run the experiment

I am going to comment out (but not delete yet) the explanation for how it is "actually quite simple" to test the placebo blocker. I feel that the experiment is flawed in that testing the two drugs simultaneously would potentially impact the results in respect to either one.

The experiment suggests: Group 1 would receive the new drug and a placebo pill; group 2 would receive the new drug and a Placebo Blocker; group 3 would receive two identical placebo pills; and group 4 would receive one placebo pill and one Placebo Blocker.

However, we still don't truly know how placebos work. Are the people in this study told what the second pill is for (a placebo blocker?) maybe that knowledge will negate the placebo entirely. For example, what if in a simple double-blind test of Drug A, the placebo group had the same results as the drug group. However, Drug B (the blocker) doesn't work. In the proposed experiment of both Drug A and Drug B, between Group 1 and Group 3, what if the placebo effect works on Placebo B, and so the placebo effect Placebo A had in the simple experiment is negated. Thus Drug A appears to be effective even though its not better than a placebo.

The bottom line is that it's not that easy to design an experiment to test two variables at once. TheHYPO (talk) 16:07, 25 May 2015 (UTC)


The sentence "My head hurts" doesn't mean a headache, but Cueball takes this too literally. It's more like a reaction on some stupid experiments witch hurt my head. --Dgbrt (talk) 20:53, 8 June 2015 (UTC)

I've just gone through the article, fixing grammar. I'm happy that spelling and grammar is good now. However I've also updated the explanation quite a bit; hopefully people agree it's an improvement, but I know a lot of people have taken an interest in this article. Edit or roll back if you like! Cosmogoblin (talk) 23:32, 14 July 2015 (UTC) P.S. In explaining the placebo, the placebo blocker, the placebo blocker placebo ... my head started to hurt too! (Could we have a "placebo-blocker-placebo blocker"? Or would that be taking things too far?)

If I'm thinking about it correctly, then a "placebo-blocker-placebo blocker" would be the same thing as a simple placebo-blocker. The only difference is that while giving them the sugar pill, instead of telling them "this pill will get rid of your headache," you'd be saying "this pill will get rid of your placebo effect." So if the blocker worked, then the patient would NOT be fooled ie they would NOT be convinced that the fake blocker works, which is kind of pointless because you DID end up giving them the blocker afterwards! Ok I agree my head hurts (talk) (please sign your comments with ~~~~)

As of the writing of this comment, there's an incomplete tag. While I agree that it's a bit bigger than necessary, it was good enough for me. However, there are two things that I disagree with (plus the typo that I fixed):

  • I don't think the headache is because they're frustrated about not being able to conduct the experiment. I think it's because they're confused about the placebo inception.
  • I don't think the sugar pill is meant to cure the headache by placebo. I think it's a comforting action to offer someone a good tasting food when someone's distressed, plus the humor that the pills Cueball's holding are probably his placebo blocker pills and that they're actually sugar pills. 22:53, 23 December 2015 (UTC)