Are Placebos Really Sugar Pills?
asserttrue.blogspot.com
asserttrue.blogspot.com
For double and triple blind (the third blind comes from not even the pharma company knowing the blinding details until after the trial) placebo design is an extraordinarily complex undertaking. A placebo must match the drug in physical appearance, taste, texture, density, state (liquid/solid/gas), and anticipated side effects. Any material difference in any of these categories renders the trial completely and totally meaningless, because at a minimum it unblinds the doctors on the ground.
Pointing out a few random oversights out of the thousands of clinical trials that occur every year is not proof of evil on the part of pharma; it is a testament to the care that goes into their design. It represents a defect rate virtually unmatched in any other industry.
But sure, go ahead and advocate irresponsible alarmism over a non-issue, as if though drug trials aren't already expensive enough, retarding scientific progress and costing millions of lives from drugs that would have been otherwise developed.
> [...] at a minimum it unblinds the doctors on the ground.
It's very important that the person administering the drug/monitoring the patient can not tell the difference.
There was one study where a drug was designed to improve motor response in patients with retarded motor skills. The doctors had found a way to determine which patient was receiving which drug.
So when it came time for the doctors to measure efficacy of the drug, they would do the motor skills test. If the patient was on the actual drug, and didn't show improvement, the doctors had a tendency to say things like "Well, that can't be right. Let me run the test again."
Disaster. Hundred million dollar trial down the toilet.
But you're replying to a comment that lays out a scenario in which 8-9 figure clinical trials were ruined by the incidental outcome of reasonable honest doctors accidentally unblinding studies.
Also not saying specifically that it was malice, but that it's either malice or stupidity, or both.
Also, I have ethical and intellectual standards of doctors doing things that will impact hundreds of thousands of people.
Not sure why you said "Who cares if doctors will figure out which medication is the placebo?" - Since I didn't even insinuate that.
I agree. I think the main issue the author wants the reader to take away is that since the ingredients in placebos are not disclosed, what's to say they can't add ingredients that will skew the results? I'm not saying that this happens, but what is stopping them? They obviously have financial, and other, incentives for their drugs to succeed, so why not take that extra measure to maintain integrity in studies that use placebos?
It doesn't seem an undue burden to ask that the placebo ingredients be documented so that their effect on the study, which is critical by any measure, can be understood. How can a separate group reproduce the results of a study if they don't have this type of information?
What you seem to have said is - trust us. We won't tell you what's going on, you can't evaluate how well we are doing, but this is a totally hard problem so don't worry about it. This is disingenuous when the drug companies have vested interests in the outcome of these studies.
If the drug companies are truly doing such a great job, then they should be recognized for the excellent work. The only way to do that is to release the ingredients used in the creation of placebos.
When you submit a drug for approval at the FDA, you need to list in exacting detail exactly how trials were run, including how they were blinded. The ingredients and form of the placebo are critical for this.
The paper referenced in this blog post examined journal articles to determine if the composition of the placebo was reported. Journal article are notoriously lacking in details and they have to be since no one want to read an article with pages upon pages of details.
If his comment was "Hey, you guys who write papers on clinical trials, it would be awesome if you'd list the placebo composition, since I'd be more likely to believe your findings" I would say "fair point".
But the blog post says "Hey! Drug companies are trying to pull a fast on all of you! Beware of what's in your medicine cabinet" which is a completely ridiculous statement. This "missing information" is available to the people that matter (the FDA).
This is truly an example of making a mountain out of a molehill.
Yes, I was being a bit dramatic for impact.
and here: http://pharmastrategyblog.com/2010/10/whats-in-placebos-who-...
But all three article (including the Topic) seem to focus on the Journal articles and neglect to write anything about the reports to the FDA. I'd be really interested in more info on that.
EDIT: I have no opinion nor info about the quality of the links i posted, I just found them through cursory duckduckgoing. So beware, they might both be total cranks. :)
Actually what it says is that placebo ingredients should be documented in studies, and regulated by the FDA. Why is that such heresy? If placebos are sugar pills, they don't need regulation. If they are designed to cause side effects, they are drugs. And that will skew the results of the drug tests, too, and as he describes in the olive oil example, not always in favor of the drug company.
The blog overtly implies that drug companies use active placebos to minimize advertised side effects, and later suggests that active placebos be outlawed.
In fact, active placebos are a requirement for drug trials, because without them, the trial can't be effectively blinded.
FDA needs to review its policies on placebos and either outlaw "active placebos" or rigorously define acceptable conditions for their use.
That's like saying "The US Govt either needs to outlaw medicines OR rigorously define acceptable conditions for their use." Which is not a statement most reasonable people would disagree with… hence the FDA. Everybody would obviously choose the second measure. It's a classic persuasion technique: give an extreme choice so the other choice will seem more reasonable. It gets attention. Nobody is going to outlaw active placebos.
As for overtly implying drug companies manipulate their results, he had a small handful of examples and one of them is a counter example where the drug company looks to be undermining its own results by using a too-good placebo (olive oil) that made its drug look useless.
Finally he raises the question: If you kick people out of the study when they improve on placebo, are you still actually testing against placebo?
How can you know, if the trials don't document what's in the placebo, and the FDA doesn't regulate it?
FDA needs to review its policies on placebos and either outlaw "active placebos" or rigorously define acceptable conditions for their use.
That seems like a profoundly ignorant argument. The article doesn't account for the need to blind studies. Why does it come across as credible to you?
You should have chosen a different line to illustrate how shady the author thinks active placebos are.
The idea behind this statement is similar: if the FDA actually pre-commits to a policy saying "either we're going to figure out a huge set of requirements for publishing placebo composition data [huge hassle for us, honestly] or, if we haven't done that by January 2015, no more active placebos"--well, suddenly the drug companies are breathing down the FDA's neck to get those requirements drafted so they can comply with them before it's too late.
This preserves all of what you can currently get, but it frames the issue so that the average individual does not assume that an on-par comparison means all side effects are placebo effect.
But these are accusations made by a man who admits he had never heard of active placebos until a few days ago. He's reading reports that are summaries of the trials and apparently concluding that since the summary doesn't include the ingredients of the placebo then they're undocumented and not regulated by the FDA. How does he know that?
And to repeat my point just made in another comment all of it based on "Over the weekend I was reading some medical studies involving placebos.....then I started to wonder..."
The claim as made by the ads indicates to the average rational person that taking the drug has no higher chance of side effects than simple sugar pills. The reasonable conclusion for anyone who would be willing to take sugar pills is that this medication is safe, when in fact it easily might not be.
No, that's the entire point of placebos. Placebos are not supposed to be absolutely nothing they are supposed to account for all manner peripheral effects and determine efficacy.
The idea is to test your treatment against something that should have no or very low effectiveness. It's fine for them to be drugs and it's fine for them to have an effect - there's a name for that: placebo effects. People tend to get this confused with the well known "the placebo effect", but that is not the only effect placebos are intended to control for.
The example you point to is dubious to me. The study seems to be working as intended: if you have a drug that doesn't work better than olive oil - you don't have a useful drug.
We're talking about a cholesterol drug here, and it's not that the drug didn't do what it was supposed to do - control cholesterol - better than olive oil; it didn't improve mortality (for a specific subgroup)...something you'd expect with cholesterol control however that doesn't mean it's not effective in it's primary objective.
In fact Clofibrate was taken off the market for precisely that reason; it's effective in lowering cholesterol but it has side effects which lead to increased mortality.
I'm all for increased transparency in placebo use; but this article seems to have done little to clear the confusion about what placebos are and what they are intended for.
Why not?
Anyway, I found this comment useful. Thanks for replying.
According to the OP whose is (just checked) a tech evangelist at Adobe. Additionally the sum of their research is "Over the weekend I was reading some medical studies involving placebos".
What I'm seeing is that someone wrote something (the OP) and made some statements and additionally pointed to one journal article (which I'm sure only a few HN'ers have even bothered to click through and glance at.)
This is interesting to discuss and that's it (at this point at least).
"(a) A sponsor who intends to conduct a clinical investigation subject to this part shall submit an "Investigational New Drug Application" (IND) including, in the following order:
...
(c ) A brief general description of the composition, manufacture, and control of any placebo used in a controlled clinical trial."
http://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/CFRS...
Once the placebo has been designed to 'match the drug in physical appearance, taste, texture, density, state (liquid/solid/gas), and anticipated side effects' it should still be described in the final paper.
In fact, there are very few studies in which patients receive an "inert" placebo.
Consider a new drug that claims to treat heart disease. It would be obviously immoral to say, "we are going to take drugs away from the control group and give them sugar pills instead".
So a "Placebo" is usually the current standard-of-care, which might be a very powerful drug.
This works because we are not at all interested in the absolute effectiveness of a drug; we are only interested in whether or not it is better then the current alternatives.
> "Placebo" is usually the current standard-of-care
is correct, either as popularly understood or as actually used during drug trials. Instead, "Placebo" is a device to make the experience of not taking a drug identical to taking a drug, which means both that it's a physical pill and that it causes the appropriate side effects.
The post we're commenting on made a series of outlandishly ignorant assertions.
That there are reasonable challenges we might make to the way active placebos are handled does not make the article better. You're a great writer. Maybe you should write a good article on the perils of undisclosed/underdisclosed active placebos? Because this particular article doesn't appear to be good.
Another reason for disclosure is the ethical risk assessment. What is to prevent a misguided executive from arranging a placebo with minimal doses of active components that produce adverse effects... just to push the results of a marginally effective drug. I am sure that 99.9% of people in the industry would not do that, but there is no reason to avoid putting checks and balances in place, for the day when some random sociopath decides to push it a little bit and get away with it.
Doesn't that mean that it's poison by nature if said side effects are negative?
Screw wanting to take part in that.
"Here's a pill that will cause oily discharge and nothing else" Yum!
Let's take DMSO as an example. Treatments with DMSO leave a particular garlic-like taste in the mouth. This is a negative side-effect. But DMSO is not a poison during normal treatments any more than ethyl alcohol is a poison.
A good placebo for DMSO would have the same side-effects as otherwise a patient and perhaps even the doctor could tell the difference based on taste/smell compared to the anticipated taste/smell. I believe this still isn't possible, which makes testing DMSO difficult. One solution, for example, is to look for a dose-response curve, which is a more complicated signal to extract.
It always made me laugh reading about 'double blinded' trials of cannabis extract for MS... In this and many other cases the blind is a complete fiction.
Isn't testing a vital part of this development, and by extension a crucial part of what you call scientific progress? The fact that we need more and more sophisticated testing techniques is because we understand more about the intricacies of human body and the ways it can interfere with treatment process. Placebo is an obvious example of such interference.
To me, calling a scrupulous testing process a factor that's retarding scientific progress is like saying QA is retarding software progress.
This makes the following statement: "When compared to a placebo, this drug had rare incidents of the following side effects..." translate as "We guessed correctly at the incidents of side effects before we started the trials," correct?
If so, then that's false advertising.
Doctor hands patient sealed opaque pill blister pack, tells them to take two, nurse watching on a CCTV or via webcam sends a text message that they've verified the patient taking the pill.
Only 6% of clinical studies list what their placebo was, so the conclusion from that is that there is a concerted effort by drug companies to unscrupulously modified the outcomes of their studies by choosing biologically active placebos. That's a bit of a stretch.
Typically what is done for a study is that drug company will manufacture the drug for the study in tablet form, complete with fillers and binders (inert ingredients used to make the pill). Then they will manufacture the same tablet without the active ingredient. The drug has to be identical in shape and color or else the study isn't really blinded, physicians could tell that patients were getting two different drug. That's the reason why drug companies make their own placebos.
Also, the example of using olive oil instead of clofibrate has two possible explanations: 1) clofibrate is soluble in olive oil so they likely gave one set of patients clofibrate +olive oil and the other just olive oil (negating the effect of olive oil) OR the study was just poorly done (yes it happens) and it would be obviously to anyone familiar with clinical trials (including the people at the FDA).
I could go on, but it doesn't really seem necessary. The article speaks for itself.
However, I'm sure that, in some cases there are oversights in the placebo choice (we do make mistakes), and we don't know which ones.
There are also big financial stakes, and pharmaceutic firms are known to cheat while attempting to bring drugs to the market. The burial of negative studies come to mind.
If you don't provide enough information in your paper to repeat the experiment (and get the same results!!), then it's not good science. Good scientific studies need to be repeatable and need to provide enough information to do it.
And even if people don't want to repeat the full experiment, being able to review the details of how it was done will help them spot mistakes. You can't have things like "peer review" without saying what you did.
So this is a serious problem, even if the rest is breathless hype.
Did we read the same blog post? The author didn't claim any such thing. In fact, he referenced a study where (if you ascribe intent) it sounds like the drug company was undermining itself by using a placebo that worked better than the drug (the olive oil).
His actual conclusion, for non-readers, is:
Placebos aren't documented, and they're not included in studies, and many times they're active in causing side effects deliberately and yet they're barely regulated -- "current anarchy" -- and this casts doubt on the validity of studies.
His call to action sounds pretty reasonable to me:
> FDA needs to review its policies on placebos and either outlaw "active placebos" or rigorously define acceptable conditions for their use.
> The author didn't claim any such thing.
Actually, the author did claim any such thing:
> Active placebos are designed to mimic the side-effects of drugs under study. So for example, if a new drug is known (or thought by the drug company) to produce dry mouth, the drug company might use a placebo containing ingredients that produce dry-mouth. That way, of course, they can say things in their ads like "[drug name] has a low occurrence of side effects, such as dry mouth, which occurred about as often as they did with placebo."
The author here is accusing drug companies of using active placebos to make the side effects seem less critical, when there's no evidence of that presented.
I feel like the author just provided an example how the drug companies could be exploiting the fact that they do not have to disclose the information about the placebo, not necessarily accusing them of it - I know I wouldn't think of this immediately (especially sleepy me right now).
It is, I agree, unnecessarily alarmist, but the conclusion is true regardless of whether the drug companies are doing that.
Active placebos are designed to mimic the side-effects of drugs under study. So for example, if a new drug is known (or thought by the drug company) to produce dry mouth, the drug company might use a placebo containing ingredients that produce dry-mouth. That way, of course, they can say things in their ads like "[drug name] has a low occurrence of side effects, such as dry mouth, which occurred about as often as they did with placebo."
and
The current anarchy that prevails with regard to placebos calls into question the reliability not just of drug-company research but of virtually every placebo-controlled study ever done. Which is a hell of a thing to have to say, or even think about. In fact it's nauseating.
Sounds to me like he's saying drug companies are manipulating the outcomes of studies and implying that any study containing a placebo shouldn't be trusted.
"Placebo" must be a potent psychedelic or stimulator itself.
I can imagine amphetamine as placebo for MDMA, but not for LSD.
August 2009 "The Rise of Placebo Medicine" by Steven Novella, M.D.
http://www.sciencebasedmedicine.org/index.php/the-rise-of-pl...
December 2010 "Placebo Effects without Deception? Well, Not Exactly" by David Gorski, M.D.
http://www.sciencebasedmedicine.org/index.php/placebo-effect...
January 2013 "Is acupuncture as effective as antidepressants? Part 2. Blinding readers who try to get an answer" by James Coyne, Ph.D.
http://www.sciencebasedmedicine.org/index.php/is-acupuncture...
There are many more good articles about placebos and how they are used in clinical trials on the site where these articles come from. The issue the blogger whose post is kindly submitted here seems so upset about is simply an issue of making a placebo (sham) treatment indistinguishable from the treatment under investigation, so that doctors and patients are properly "blinded" during the trial. Another comment already posted here on HN has pointed out that the FDA does oversee what ingredients are put into placebos, whether or not those detailed ingredients are published in a peer-reviewed research study. The blogger's concerns are legitimate, but not proportionate to the actual problem.
Disclosure: I have been a subject of FDA-regulated medical trials. I was very impressed by thoroughness of data collection in those trials, and by my inability to distinguish whether I was receiving placebo or genuine medicine in one of those trials. My oldest son, now a hacker for a start-up, had work experience while in college at a medical device company, and he was impressed that every line of computer code he wrote during his summer job was reviewed line-by-line by FDA computer scientists as part of the review process for the medical device he worked on.
Is there any evidence that poor placebo design is a problem in practice, or is this a theoretical concern?
Yes, placebo design is extremely complicated and often involves active ingredients intended to simulate side effects and treatment experience without treatment effects.
There is quite a bit written in the comments here.
For instance, it would be quite easy for a company to fake efficacy accidentally or intentionally by using a placebo that inhibited natural healing or even made the treated condition worse. The control group would fare worse than the treated group, and the drug could advance toward the market. There is a massive economic incentive for this sort of cheating, and I don't think we can presume pharmaceutical companies are above economic incentives.
Pointing out a study where the pharma company might have used a placebo that made their drug harder to approve is hardly in keeping with the insinuation that they're using placebos for "disturbing" ends.
Seems more like a case of "throw all the mud and hope some of it sticks" to me. Sorry.
(Drugs should be tested against the best existing alternative treatment, not merely a placebo. We are not interested in whether a new drug has a non-zero effect - but whether it is an improvement on the state of the art in some way.)
Certain placebos, they add, may skew results in favor of the active drug. The researchers referenced a trial for a drug used to treat anorexia linked with cancer in which a lactose placebo was used. Since lactose intolerance is common among cancer patients, the fact that some suffered stomach problems from the placebo may have made the actual drug look more beneficial.
[1] http://articles.latimes.com/2010/oct/18/news/la-heb-placebo-...
[2] https://en.wikipedia.org/wiki/Placebo-controlled_studies#Pla...
From CFR 21 Part 312.23 (a)(7)(iv)(c): A brief general description of the composition, manufacture, and control of any placebo used in a controlled clinical trial. [is required to be supplied for an Investigational New Drug Application]
What is the difference? It seems the law clearly states that for new drugs, you must list the composition of your placebo to the FDA. It would seem the article is mistaken. I am not a lawyer nor do I work in the pharmaceutical industry. I just did a google search for "placebo composition site:fda.gov"
I don't know how that fits in with blinding everyone involved. As abtinf and refurb mention, designing placebo is a difficult process.
And you don't need to include anything to create side-effects; people taking sugar pills will happily report having a range of side effects.
This is also done, depends on the trial. For example, new vaccines are often tested against the existing one(s).
However, how can we prove that something is 100% biologically inert? We can't account for all the unknown effects, is there any such 'ingredient'? I guess that at least the used placebo should be proven to not affect directly anything directly related to the hypothesis.
First of all, most immporant studies don't use a placebo because it would unethical when there are standard treatments out there already and you have to show that your new drug is just as good or better than those treatments (the proper terms are supperior vs. non-inferior trials).
But considering trials that do use a placebo - the whole point is to prove that their drug is superior to the placebo. If anything, when they design the placebo, it would have structurally similar molecules that would give the same side effects, and potentially the same results <-- which is key. A research DOES NOT WANT the placebo to do as well as their drug. The placebo can only hurt their results, not help them cheat.
You could also argue that since it's not regulated, maybe the active component does actual harm to make the test drug look better! No. There are too many studies showing what the controlled group should look like (whether they are on a placebo or on nothing at all).
http://en.wikipedia.org/wiki/Calcium_metabolism#Interaction_...
But, according to the paper's author, Beatrice Golomb, MD, PhD, associate professor of medicine at the University of California, San Diego School of Medicine, this standard has a fundamental problem, "there isn't anything actually known to be physiologically inert."
http://www.sciencedaily.com/releases/2010/10/101018174335.ht...
So on average the effect of this small amount of sugar disappears.
I assume it's just labeled as homeopathic because they failed to fake some clinical trials to get FDA approval (or didn't even try), but maybe it's actually an effective marketing choice.
That said, I think capitalizing on that misconception is wrong and that it might be reasonable to take actions which prevent true but subtle statements like "drug had similar side-effects to a placebo" from misleading laymen.
i.e. would you subjectively report more pain relief when given ibuprofen (placebo) vs. vicodin (placebo).
But, pesky IRBs :(
Here are some Alzheimer placebo samples (pictures) from Pfizer, handed out to doctors in Switzerland. Sorry for the bad quality. Pictures taken with a phone.
For example, if you're testing fish oil, you wouldn't use a sugar pill since it's easy to know what you're taking by looking at it. You would use something like olive oil.
>That way, of course, they can say things in their ads like "[drug name] has a low occurrence of side effects, such as dry mouth, which occurred about as often as they did with placebo."
cooldeal point is that you need to mimic the side effects to make the control group believable. Whether it is needed to use inert components or not would depend on the nature of the drug in test.
That being said, the opacity of the manufacture process is concerning.
Studies with active placebos should have a third branch with inert placebos, to assess the incidence of side effects.
I think at best you could say it would amplify already present ethical issues.
1. Study participants are told that some of them will receive a "sham" (placebo) treatment. Participation is voluntary.
2. It is not known whether the drug they are testing is effective or comparable to the current standard of care. Determining that is the purpose of the study.
Take oncology drugs for example. Let's say you come up with a great new drug to treat Non-Hodgkin's lymphoma (NHL). Currently approved drugs already are very effective, pushing overall survival to close to a decade.
If you're a drug company, there is zero chance you'd get approved (by the independent IRB board who monitor ethical considerations) to run an NHL trial using a placebo. Even if patients agreed to take part in the trial, it would be unethical to deny effective treatments for a potentially fatal disease.
Hence, the FDA has softer data and efficacy requirements for oncology therapies. There is no requirement for a placebo arm unless there is no currently effective treatment available.
Also, I'm not a huge fan of more regulation, but scientific integrity seems to demand the placebo ingredients be disclosed.
From the discussion, it seems clear the doctor-scientists have to think about and carefully consider what to use for the control. A single, simple answer cannot work.
If the physician can effect the trial by knowing to whom they're administering the placebo it seems that the trial lacks in its design. Someone else said that the person administering the drug/control could repeat a test because they felt the result was wrong - why is such a person involved in a drug trial? If they lie about that then it seems they can pervert the trial any number of ways.
How do the patients know how the other members of the study are being affected? If they don't get a dry mouth, but the drug company thinks they should, how does the patient know?
Surely you need the physicians and patients to record exactly what happened, no more and no less.
It seems that you're doomed to fail if you don't know that this is what's happening: suppose you use an active control that dries the mouth (for 95% of patients), if you don't have the side-effect with the drug on test then you know you're part of the active group with a high enough confidence level to alter the results in a qualitatively similar way to using a non-active control (placebo).
So what you need is a placebo that matches exactly the profile of the side-effects of the drug. I'd warrant that's impossible - you don't know the side-effects of the drug that are expected in a wide scale longitudinal study because that's what you're testing isn't it?
What is entirely nefarious is any use of "no significant side-effects beyond that experienced under placebo" when in fact they don't mean placebo but "active-control".
Placebo may be synonymous with active control in Pharma circles but it's not in common usage and so this is still a fraud.