(After reading the article on placebos in just-out National Geographic)…
Has anyone ever done an experiment in which both groups of patients were given a proven-working drug for their condition, but one was told that it was only given a placebo?
What would be the reason for such a study? What question would you be trying to answer?
The question it may answer would be - is there a “placebo effect” overlayed on/added to the effect of actual working medicines? That is, is the drug working better when the user knows it’s a drug that is supposed to work - as opposed to it working “objectively” without the user knowing that.
Yes. Patients receiving pain management drugs often get less of a benefit from the drugs if they think the drug has somehow lost its effectiveness even though it hasn’t, which isn’t quite the same as being told they are receiving a placebo but it’s basically the same thing. They think the drug they are taking doesn’t do anything (so they think it’s basically just a placebo) but the drug is still just as effective as it always was. This is known to have a negative outcome in surgical recoveries as well as pain management. Probably other things as well, but one of our resident medical dopers will have to give more detail on that. This isn’t my field of expertise.
I believe it’s lumped under the “nocebo” effect, though this often also refers to experiencing negative effects from a placebo.
Would not be permitted under current clinical-test protocols. You can’t /tell/ a person that they are being given only a placebo, then secretly give them a drug.
You can’t do the opposite either. You can’t claim to be giving a real drug, then secretly give a placebo.
And in most cases you can’t even tell patients that they “might” be given a placebo. Most drug trials compare two drugs (though pershaps one is known to be not very effective)
Yes, it’s fairly difficult to test some aspects of the placebo effect. In fact there is the suggestion (most people express it as a joke), that the placebo effect isn’t reall - it’s only a non-significant statistical anomaly. How would you know?
Oh, that’s easy. You do a study with two groups. The first group, you give a genuine placebo, while with the second group, you only tell them that they’re receiving a placebo.
Wikipedia: Nocebo
Tell the groups they may be all be given drugs or all be given placebos. Then tell them any group may be told they have all been given drugs or all been given placebos but that information may or may not be true. It will take more than two groups but if there’s an effect it should eventually be seen, but it may not be apparent as with common placebo tests where the subjects have no idea they are given a placebo.
ETA: And no effect may be seen at all. The point of the test is to find out if there is a no-placebo effect, not to look for predetermined conclusion.
This is not correct. Almost every drug tested before it reaches market in the US must be tested against a placebo via something called a double-blind study. Neither the study administrators (doctors, nurses, etc) nor the patients know whether the patients will receive the working drug or a placebo. The patients are all told before they enroll that this is the type of study that they are participating in. During the drug trial, data is gathered and analyzed to determine whether or not the drug is statistically more effective than the placebo. Yes, this means that a portion of the participants are not receiving a working drug, even in cases where you are testing for effectiveness for an illness.
If you read the in-box inserts that come with your medications, or look them up online, you can usually find information about how well the drug performed versus placebo. Double-blind studies are how this is determined.
The only exception to this type of testing is given for certain types of drugs in a life or death situation. You may recall the Ebola drugs that were used recently that had not undergone the full set of human subject testing. There are other emergency protocols that would allow drug companies to release certain drugs in specific circumstances, such as an experimental antibiotic that might extend someone’s life long enough to get an organ transplant. These are extraordinary situations and definitely occur out of the normal protocol. Exceptions are not always granted and people die.
Yes, sometimes people die from not getting the experimental drug, but let’s remember to keep that in context: There’s a reason that experimental drugs are experimental in the first place. People sometimes also die, or otherwise have their condition seriously worsened, from hitherto-unknown side effects of experimental drugs.
NM. I see that Sunny Daze already made the point I was going to. Should have reread the thread, Fred. Instead. So soon after getting out of bed. Still have a fuzzy head. My favourite colour is red. I’ll stop now.
Yes, quite true. I was trying to make the point that getting around the double-blind, drug-safety testing protocol requirements is quite difficult and not always possible, even in cases that might seem like “obvious” exceptions.
The testing requirements are so stringent for exactly the reasons you outline. The drugs being tested are experimental and their effects are unknown. Many drugs never make it to market precisely because they are worse for the patient than doing nothing (or because they have no net benefit). There are also prominent instances of drugs that did make it through testing and ended up being removed, or re-labelled, because continued data collection revealed serious drug problems once the drug was in the larger population.
To the original OP - it would taking some digging, but I’m pretty sure at some point in the 60’s or the 70’s, before patient safety protocols were so stringent, that some psychology team somewhere ran this experiment. It sounds like something we covered in Psych class in university. I’ll see if I can find anything.
Sunny Daze is quite right, although as I can’t help myself, we do sometimes do open label studies, where everyone knows who is getting what. This is common in my area, where the drugs we are bringing into the clinic have a certain inherent risk, and/or it would be obvious which are the drug. Radioactive drugs are a good example of this. You can’t give someone a radioactive placebo, and if you’re giving them something non-radioactive then they would know it’s the placebo.
Double-blind studies are not all done against placebo. Some were in the past. A few are even now. but I can’t remember the last time I read a double-blind study that was done with placebo as one arm.
Given the number of drug tests always happening, I’m sure that there are some happening right now where one arm is a placebo, and I’m sure that people who read a lot of drug studies will know lots of them: that’s why I qualified the third statement:
before going on to say
— which is how most double blind drug stidies are done.
Not related directly to the OP. but a useful fact all the same, expensive placebos are more effective than the cheaper kind.
Please do if you can. I have never seen anything like that, and it would be interesting to see. A study like this, if it shows a significant effect, would be applicable to all kinds of things. Like drug advertising or physician conversations with patients when discussing treatment.
Apologies for continuing the hijack, but as I followed up out of my own interest. I took a look at 4 recent new-drug launches in the US (Ameluz, Taltz, Adlyxin, and Onzetra). All four have placebo-controlled trials supporting their approval. I stopped there just because of time, but placebo-controlled trials look to be still quite common.