Do non-alcoholic substances cause a genuine placebo-effect?

[EndME]

Is drinking one alcohol free beer and getting a "relaxed feeling" without any drunkeness a genuine placebo-effect? If not, what is it actually?

My understanding is that "placebo-effect" is commonly used as a phrase that encompasses all sorts of effects that are not due to a substance or something else being effective, for example including regression to the mean or using inadequate subjective outcome measures. But that strictly speaking a genuine placebo-effect would more refer to something that is related to believing in something that has no actual efficacy as substance but still getting an effect from that belief alone.

So now my question would be under which category would drinking alcohol free beer and getting a relaxed feeling from that fall?

Is the relaxed feeling a consequence of the hypothalamus or whatever else goes on in the body sensing a substance that is familiar to it's alcoholic brother and as a consequence sending out the same signals and as such the relaxed feeling being no different to when one drinks alcoholic beer in which case most people would argue that it is not a placebo-effect? Or does it have to do with the "mind being tricked somehow" due to learned associations or even social rituals? Why does the relaxed feeling come before any feelings of drunkeness? Why does the body not realize that no drunkeness is setting in and then remove the feeling of relaxedness if that's the situation that's occuring i.e. if the feeling comes from anticipating the effects of alcohol, why does it persist?

To me it seems it would not be a genuine placebo-effect especially because even if I convince myself that I should not be relaxed and am living with the idea that alcohol free beer should not make me relaxed, I'll still be relaxed, but I'm not quite sure if I can even come up with an accurate definition of what a genuine placebo-effect is supposed to be and how to interpret the given situation. What would happen to a person that never had alcoholic beer or those that break down alcohol so quickly that they can't really get much effects from beer? Nothing, because their body doesn't "misinterpret" the molecules associated to alcohol or still something because the signals don't actually come from the alcohol itself but rather from social settings or relaxing effects from things like hops?

If it is not a genuine placebo-effect, is not still the consequence of the body being tricked? Would it be dualist to say it's not a placebo-effect because it is the body doing the misguiding rather than the mind?

I guess this is something the @jnmaciuch's and @Jonathan Edwards's of this world can answer easily.

 

[Eddie]

If the way you feel changes, then the part of your brain that manages the relaxed feeling in your brain was affected by something. I would argue that a placebo effect just means we don't have good reason to think that there is a causal pathway between the chemical in the substance and physical change in the brain. But that doesn't make the affect a property of the "mind" it just means that there is something else interacting with your brain that doesn't stem back to the chemical interactions caused by the substance.

There are 1000s of potential explanations for what is making you relaxed and it is impossible to fully know which one is occurring. But in every case the final step will be some physical process in which a certain neurons receive whatever signal produces a relaxed feeling. It could be the case that whenever you touch a glass bottle of non-alcoholic beer, certain receptors in your fingers sent a message to your brain that triggers a sequence of events culminating in a relaxed message being sent to certain neurons. Maybe seeing the colour of the beer activates just the right neurons to have them send a relaxing message to some other neurons. Or maybe the neurons that identify whether you are in a certain social setting are triggered and send a message to other neurons to relax. Some of those specific examples are unlikely, but whatever is happening there is still a real effect, it might just not be caused the the substances in the beer itself.

Believing in something is a physical process that causes certain neurons to fire and might change all sort of physical parameters. The firing of neurons that represents the thought that the alcohol was poisonous would probably result in the release of adrenalin and cause all sorts of physical changes. That would be a placebo effect, but unless you believe in a separate mind, it is still just stuff interacting with stuff.

 

[EndME]

I would argue that a placebo effect just means we don't have good reason to think that there is a causal pathway between the chemical in the substance and physical change in the brain.

That seems sensible to me at first (maybe brain is even too narrow and has to be replaced by "physical change in the body"?), but it seems every definition of the placebo effect actually involves the belief in something. To me that would mean that people reporting improvement as part of homeopathic trials without thinking that the drug does anything would not be considered as part of a genuine placebo-effect whilst the belief in them with the same reported improvement would be considered a placebo-effect. I'm not quite sure whether that's a sensible definition or how it can actually ever work in practice, because capturing beliefs seems very hard, but it seems to be the definition which one stumbles across everywhere. I can also understand why one would want to involve something like belief in such a definition or more accurately not keep the above definition as general as above, as it seems to simply be what I referred to in my post by the “colloquial version of placebo-effect” which can’t separate between things such as regression to the mean and Hawthorne effect which are distinct phenomena which a more specific definition would probably want to separate in some sense. My impression is that all things placed together in the bucket of "lack of a causal pathway between chemical and result" are captured under “placebo response” but not under “placebo effect”. The easiest way to then define "placebo effect" would be along the lines of "placebo effect=placebo response-Hawthorne effect-regression to the mean-other known effects" in which case one doesn't have to involve something like "belief" at all, but it would deviate from the classical definition one usually sees.

Believing in something is a physical process that causes certain neurons to fire and might change all sort of physical parameters. The firing of neurons that represents the thought that the alcohol was poisonous would probably result in the release of adrenalin and cause all sorts of physical changes. That would be a placebo effect, but unless you believe in a separate mind, it is still just stuff interacting with stuff.

But that's not quite the question or is it? Sure, it's all just a result of neurons firing but the question is surely why they fire? Is it a result of the social setting, the substance, a combination or something else? That doesn't seem too hard a question to be answered to me. Blindfolding people or giving them unlabelled drinks would answer one bit of the question, removing them from a social setting another bit and it seems to me in both cases they would stay relaxed. Of course one can argue that removal from a social setting is insufficient if the "feeling of relaxed" is not a direct consequence of the social setting itself but rather a result of the associtation between a chemical and a calm setting. The social setting idea works quite nicely for me as I have no idea how to answer the alternative: Why would the neurons continue firing as a consequence of a chemical, why don't they receive signals that the substance is different to the alcoholic one after a while?

 

[Eddie]

If a placebo effect has been defined as having to involve a belief in the treatment then that is fine. But even if someone believes in the treatment and experiences a placebo effect, we don't know that it was the belief that caused the effect and not some other factor. In any given example we don't currently have the ability to know if the 'belief' neurons impacted the 'feel relaxed' neurons. So at this stage I think we should keep it broad and say there is some impact on the 'feel relaxed' neurons but we don't know what specifically caused that impact.

As you mentioned testing people in different situations might be a good way to try and see what is causing the effect. The reason I think it is a difficult question is we don't know and can't rule out all the possible contributing factors to the effect. When you try to control for one factor, the fact that you are running an experiment at a different moment of time will have changed many other things. This isn't a huge problem when looking for clear significant effects, but when things like the impact of beliefs are involved I think it poses challenges too great to drawn any real conclusions.

We don't know why some specific neurons would continue to fire because we don't know enough about the brain. Maybe the neurons responsible for 'feeling relaxed' communicate in ways where a initial signal causes an effect that lasts for a longer period of time. If you want to try and narrow down which factor(s) are responsible, testing it in different scenarios might help to tentatively exclude some factors. But also is it really important to know exactly what is causing the effect when you can reliably bring about the desired outcome?

 

[EndME]

If a placebo effect has been defined as having to involve a belief in the treatment then that is fine. But even if someone believes in the treatment and experiences a placebo effect, we don't know that it was the belief that caused the effect and not some other factor. In any given example we don't currently have the ability to know if the 'belief' neurons impacted the 'feel relaxed' neurons. So at this stage I think we should keep it broad and say there is some impact on the 'feel relaxed' neurons but we don't know what specifically caused that impact.

As part of most observations, for instance medical trials, I think you can probably almost never tell whether something is a genuine "placebo effect" or rather just something part of the general bucket of "placebo response" and you can never determine if some experienced a "placebo effect" or "placebo response". Quite possibly the same is the case here. But I don't think that should matter to definition of "placebo effect" or "placebo response", especially when you can probably estimate whether something is more likely to be impacted by some effect other than a "genuine placebo effect". For the PACE trial for example there is no need for "genuine placebo effects" to exist to yield positive results, it can all already be covered by other effects that have nothing to do with treatment efficacy.

As you mentioned testing people in different situations might be a good way to try and see what is causing the effect. The reason I think it is a difficult question is we don't know and can't rule out all the possible contributing factors to the effect. When you try to control for one factor, the fact that you are running an experiment at a different moment of time will have changed many other things. This isn't a huge problem when looking for clear significant effects, but when things like the impact of beliefs are involved I think it poses challenges too great to drawn any real conclusions.

Very possibly, but would anybody ever suggest that feeling relaxed from an alcoholic beer is a placebo-effect? In general I wouldn't be expecting a precise answer to the question but if nobody would suggest it would the result for alcoholic beer on what basis would one make a different conclusion for non-alcoholic beer? Perhaps not any neuroscience is need to answer the question to begin with and it can be shown to all be a fairly simple correlation to the half-life of hops? I don't know.

But also is it really important to know exactly what is causing the effect when you can reliably bring about the desired outcome?

No, none of this is any way important. It's purely supposed to be an example for me to better understand what a "genuine placebo-effect" is supposed to be, how one would go about defining it and what it isn't supposed to be, even if it can't be determined in all practical cases and perhaps not even in this case.

 

[Kitty]

Is drinking one alcohol free beer and getting a "relaxed feeling" without any drunkeness a genuine placebo-effect?

Possibly, because you feel the way you expect to feel. If you take part in a trial of a drug or intervention, you also expect something to happen.

Maybe it's not as simple as that, but it must be part of it?

 

[EndME]

Possibly, because you feel the way you expect to feel. If you take part in a trial of a drug or intervention, you also expect something to happen.

Possibly, but my anticipation was the opposite: I thought I wouldn't feel anything.

I'm not sure you always expect something to happen as part of any trials, but something will happen to you independently of that simply because of all the other effects already (Hawthorne effect, Regression to the mean, Observer bias, ...).

Perhaps, @Eddie is quite right in that it is more accurate to say you can almost never determine whether something is really due to a "genuine placebo-effect" and that it's more about trying to get a handle for which effects dominate.

 

[Eddie]

As part of most observations, for instance medical trials, I think you can probably almost never tell whether something is a genuine "placebo effect" or rather just something part of the general bucket of "placebo response" and you can never determine if some experienced a "placebo effect" or "placebo response". Quite possibly the same is the case here. But I don't think that should matter to definition of "placebo effect" or "placebo response", especially when you can probably estimate whether something is more likely to be impacted by some effect other than a "genuine placebo effect". For the PACE trial for example there is no need for "genuine placebo effects" to exist to yield positive results, it can all already be covered by other effects that have nothing to do with treatment efficacy.

Very possibly, but would anybody ever suggest that feeling relaxed from an alcoholic beer is a placebo-effect? In general I wouldn't be expecting a precise answer to the question but if nobody would suggest it would the result for alcoholic beer on what basis would one make a different conclusion for non-alcoholic beer? Perhaps not any neuroscience is need to answer the question to begin with and it can be shown to all be a fairly simple correlation to the half-life of hops? I don't know.

No, none of this is any way important. It's purely supposed to be an example for me to better understand what a "genuine placebo-effect" is supposed to be, how one would go about defining it and what it isn't supposed to be, even if it can't be determined in all practical cases and perhaps not even in this case.

I think you are getting caught up on the labels. If we have "placebo response" as the category that includes all effects not directly tied to casual pathway of the drug then that is the level at which we can make conclusions. The "placebo effect" would be a subcategory that is hypothesized to exist but can't be shown to contribute in any case. If we have good reason know what factors did impact the placebo response then it might make sense to call it something else, but I am somewhat skeptical we can know what caused the response in these types of cases.

It might be a bit like trying to determine if the cause of cancer in any one case was from consuming bacon. That might be a reasonable hypothesis since we know it can impact cancer risk, but there are so many factors at play that in a specific case it would be impossible to have any confidence. It wouldn't make sense to then have a category based on bacon caused cancer, its just cancer. Once we develop a way to reliability determine the cause of cancer we can then call some of those cases genuine bacon caused cancer but until then its a useless label.

 

[Kitty]

Possibly, but my anticipation was the opposite: I thought I wouldn't feel anything.

I guess it depends whether all aspects of expectation are in the thinking brain? (I don't know the answer to that.)

In this specific example it may also depend on experience and how much alcohol was left in the beer. If the beer is brewed via the traditional process—rather than being a non-brewed, beer-flavoured soft drink—it's difficult and expensive to remove it all.

"Alcohol-free" could mean "alcohol content is low enough to meet the standard, but not necessarily low enough that your body can't detect a substance it's met before".

 

[EndME]

I think you are getting caught up on the labels. If we have "placebo response" as the category that includes all effects not directly tied to casual pathway of the drug then that is the level at which we can make conclusions. The "placebo effect" would be a subcategory that is hypothesized to exist but can't be shown to contribute in any case. If we have good reason know what factors did impact the placebo response then it might make sense to call it something else, but I am somewhat skeptical we can know what caused the response in these types of cases.

I'm not quite sure. If you have a trial of a drug and a placebo and the placebo does better than the drug, but the higher response rate is caused by one clinican who predominately prescribed the placebo and who had a much higher response rate in both substances then you can probably have good reason to believe that the effect is not caused by a placebo-effect but rather by recruitment bias or things falling under the Hawthorne effect label. You can probably also witness that the placebo-response rate is greater in conditions that naturally remit, that to has nothing do with a "placebo-effect". Similary you can probably expect a greater placebo-response in pain conditions where regression to the mean plays more of a role than say in cancer. In most practical scenarios you can probably never put a number on anything specifically and it doesn't matter either but in more sophisticated settings quite a few things can probably be estimated to some extent depening on the context. I'd think that's all that matters in practice. I do think there is also evidence for a genuine "placebo-effect" even if most trials can't seperate individual effects. In all these cases you can make useful conclusions that go beyond "placebo-response" even if you can't delineate everything at more detail.

It might be a bit like trying to determine if the cause of cancer in any one case was from consuming bacon. That might be a reasonable hypothesis since we know it can impact cancer risk, but there are so many factors at play that in a specific case it would be impossible to have any confidence. It wouldn't make sense to then have a category based on bacon caused cancer, its just cancer. Once we develop a way to reliability determine the cause of cancer we can then call some of those cases genuine bacon caused cancer but until then its a useless label.

Sure, you can probably never determine the cause of a single case of cancer, but that's purely a consequence of causality as we know it almost never really working in medicine in a straightfoward way. Yet we speak of smoking as the cause of lung cancer because of the strength of relationship, rather than anything else. I think it's quite fine that language is contextual and that cause and effect are typically not the same in medicine as elsewhere. I think everybody understands that it's possible to get lung cancer without smoking and or that it's possible to smoke 2 packanges of cigarrettes for 70 years and never develop lung cancer.

Perhaps one should ask "what does one estimate, which effects dominate the relaxedness following non-alcoholic beer" but isn't that already implicit in the way "cause" is used in medicine?

 

[EndME]

In this specific example it may also depend on experience and how much alcohol was left in the beer. If the beer is brewed via the traditional process—rather than being a non-brewed, beer-flavoured soft drink—it's difficult and expensive to remove it all.

"Alcohol-free" could mean "alcohol content is low enough to meet the standard, but not necessarily low enough that your body can't detect a substance it's met before".

Possibly, but I'm fairly convinced I don't get the same effects from a glass of juice with some drops of vodka, but I could be wrong.

 

[Kitty]

Possibly, but I'm fairly convinced I don't get the same effects from a glass of juice with some drops of vodka, but I could be wrong.

But you can't test it because it's not the same thing. Your brain may respond differently to a suggestion of beer than to a suggestion of vodka.

There may also have been numerous other factors that influenced your response at the time, which you couldn't reproduce for a vodka experiment. Mood is one of them, but also how recently you ate, what the lighting conditions were like, whether you were alone or with someone else, what you'd been thinking about and doing that day, what the weather was like, whether the situation triggered any memories...

 

[EndME]

But you can't test it because it's not the same thing. Your brain may respond differently to a suggestion of beer than to a suggestion of vodka.

But in that case it's not the suggestion of alcohol that matters, but the suggestion of beverage, which brings us back to where we already were.

There may also have been numerous other factors that influenced your response at the time, which you couldn't reproduce for a vodka experiment. Mood is one of them, but also how recently you ate, what the lighting conditions were like, whether you were alone or with someone else, what you'd been thinking about and doing that day, what the weather was like, whether the situation triggered any memories...

In my case I have the impression that it can be reliably reproduced independent of the circumstance (but admittedly I see no point of trying drops of vodka with juice, so I cannot be sure of anything on that front). Of course the association may be to circumstances in which beer is usually consumed but I see that as part of the question for which I have no answers to begin with.

 

[jnmaciuch]

I’m not sure there really is a broadly agreed upon definition of placebo. In the context of a clinical trial, the question you’re asking is whether clinical improvement is due to the drug mechanism or anything else not attributable to the drug mechanism, so “placebo” in this case would be a catch-all. The specific cause doesn’t really matter

The association with belief/expectation is probably more due to historical context and psychological speculation rather than anything concrete. I think most scientists would consider those as factors which might bias their results without a specific thesis of why/how. Even psych studies designed specifically to assess the role of belief/expectation aren’t really capable of ruling out other factors, let alone making any mechanistic claims

 

[Utsikt]

But that strictly speaking a genuine placebo-effect would more refer to something that is related to believing in something that has no actual efficacy as substance but still getting an effect from that belief alone.

I think you need to define what placebo is supposed to mean. Everything will have an effect in the purest sense of the word, otherwise you’d break causality.

So placebo might be defined as achieving an outcome through a different mechanism than what you’d expect to occur based on the intervention you believe you’re exposed to.

Although that definition assumes ignorance about how the «real» intervention works, because presumably the «real» intervention could also cause an effect through the same mechanisms.

If a placebo pill lowers the reported pain, then you’d assume that the real analgesic lowers the reported pain through both the pharmacologically targeted mechanisms and the «placebo» mechanisms.

So in this instance, you could divide the «effect» into that caused by nerve signals being pharmacologically blocked (if that’s what the analgesic is designed to do), and that caused by the other mechanisms.

And the placebo would be the effect of the other mechanisms, whatever they are.

 

[Jonathan Edwards]

I guess it depends whether all aspects of expectation are in the thinking brain? (I don't know the answer to that.)

I am pretty sure we have lots of evidence for 'expectation' being much more general than just parrt of 'thinking'. I forget the exact experiments but my memory is that people like Paco Calvo who work on plant behaviour, have shown that you can get a plant like a blackberry to 'expect' there to be light in a certain direction - to which it will grow - using something the plant has previously found to be associated with more light (maybe wind direction). There is an amazing video of a bean plant that expects there to be a pole in a particular direction, possibly because of shadows or something like that, and in response throws itself (or its growing tip) to exactly the right place. It may be that plant cells think but we can be pretty sure that whole plants do not think, in the sense of sensing what they are doing.

 

[bobbler]

placebo
a substance that has no therapeutic effect, used as a control in testing new drugs.
a measure designed merely to humour or placate someone.

placebo effect
a beneficial effect produced by a placebo drug or treatment, which cannot be attributed to the properties of the placebo itself, and must therefore be due to the patient's belief in that treatment.

The term placebo response technically refers to the full range of outcomes and changes a patient experiences after receiving a placebo (an inactive or sham treatment). This is a broad measure used in clinical trials and encompasses several factors, not just the specific "placebo effect" itself.
it would be down to what the person was attributing to what really.

if the person drinking non-alcoholic beer wasn't being deceived into thinking it was alcoholic then is it placebo effect, I think no?, rather than them knowing/not really either thiking about or caring what the person who put it into their hand is believing, but just thinking it is a nice drink that is making them feel relaxed

I think it is a pretty hard one to validly test tbh because if blinded but told it could be either and then asked if they feel confident to get behdin the wheel of a car (or any other activity) it is more a measure of people's confidence and risk etc as anything else. And people get tired and relaxed in the evening or times when they'd normally drink any kind of beer, be it social or after-work, shoes and work gear off and favourite film on or whatever. It's a bit of a weird question to ask that I'm not sure most are qualified to answer it is so irrelevant to daily life.

So I'm not even sure you could easily do the placebo response on it, but definitely not the placebo effect and suggest it is someone's belief in the drink itself.

I mean what happens when because it is 'going down so well' and you actually had somehow deceived a beer drinker it wasn't non-alcoholic (unlikely as someone who used to work in a bar, people know when it isn't their normal beer and if it isn't their normal beer then they are drinking it in a different sense to when it is one they are 'in a routine with') then how do you deal with the issue that they managed to drink a 5th one remarking they normally finish at 3 etc

 

[MrMagoo]

I once drank a whole bottle of rosė schloer and felt a bit dizzy and guilty as I poured the last glass, I’d been really enjoying it and feeling relaxed. I then remembered that it isn’t wine, it’s just grown-up fizzy juice made to taste a bit like wine.

My point being - if your brain recognises the taste and smell of wine, and wine = relaxed, giddy, dizzy, tipsy, why shouldn’t something which smells and tastes a lot like wine induce the same feelings for a few minutes? Same with non-alcoholic beer.

If I smell a cigarette now, it makes me feel ill but when I smoked the smell of someone else’s made me feel good, a little pick-me-up. If I smell vanilla perfume it reminds me of cake (which I love) I can very well later think “oh I’ve already had a treat today” except I didn’t eat a cake, I smelled it!

I think the question here is less about placebo and more about olfactory senses and how they trigger responses in us.

 

[bobbler]

I think wine might be an easier one to do it with and to do dose effects on because what choice people have to drink the exact same drink every time is less with wine and even the same wine probably varies a bit.

But I’m no wine drinker and have no idea where low or no alcohol is at for wine and how good it is.

I have drunk schloer interestingly and it’s as close to the wine taste as I’d like so I assume vice versa for those liking drier ie not sweet wine (assuming that tastes as I think) it would be as you described where you kind of know but enjoy the taste spectrum being ‘wine like’ etc,