Placebos and Clinical Trials

Monday, January 09, 2012

Beryl Lieff Benderly discusses the potential dangers of relying on double-blind clinical trials, which she sees as damaging the chances for patients in dire need of getting treatments. She also talks about why she thinks too many researchers are looking at what placebos aren't doing, as opposed to what they are. Her latest article, "Head Games," is in the current issue of Miller-McCune.


Beryl Lieff Benderly

Comments [16]

Graham from Bronx

Four comments:
1) The segment was supposed to be about a critique of Double Blind studies, but the whole segment was about the placebo effect,
2) If both groups of patients (placebo and medication recipients) are told that they might be given a placebo then BOTH groups would be prone to the "placebo effect" therefore the pharmaceutical effect should still be the evident statistically,
3) The placebo effect always decays over time, and
4) The guest appeared to be unfamiliar with design of experiment concepts, and statistical anlaysis, so did she really mean that a placebo was being used or was a control being used (these are two different things).

Jan. 10 2012 10:24 AM
Angela W Stucker from Brooklyn

I have Parkinson's disease - I was diagnosed twelve years ago at age 38 with young onset PD. In 2001 I participated in a clinical trial for a very promising drug being developed by Guilford Pharmaceuticals, which handed responsibility for conducting the trial to Amgen. First to correct some statements made by listeners - phase I of a clinical trial is designed to test basic safety issues - participants are healthy individuals, who do not have the condition that the drug is intended to treat. Phase II, which is often conducted in two segments, is intended to test the drug for efficacy and safety, and in the case of blind and double blind studies, uses a placebo group. Most trials do use a placebo group.

The particular drug trial I was involved in was halted midway through the second phase of Phase II. The reason given was that even though some improvement was seen among the participants, it wasn't deemed sufficient by Amgen to continue testing the drug. I would agree that the placebo effect can definitely be a significant factor in testing, especially in PD patients, but I find it interesting that none of the professionals in this segment discussed the fact that there is an "objective" way of testing progression in PD - SPECT imaging, similar to a PET scan. A small subset of the trial group were also given SPECT scans during the trial. I was part of this group also.

When the trial I was involved in was dropped - it was dropped. Because the doctor heading the trial at the center where I was seen - Beth Israel in Manhattan - left around the time the trial ended, I wasn't told which group I was in, placebo or drug. Even my neurologist was unable to obtain the information for me. It took several attempts at communication from me before I finally received that information from the CEO of Amgen. Several patients tried to get more information about exactly why the trial was halted, after we had heard that the SPECT imaging provided some confirmation that the drug was at least somewhat effective. We were on the medication for six months, long enough that it was unlikely that the improvement was due to the placebo effect.

I am not suggesting any kind of conspiracy here, other than the "conspiracy" of greed that is inherent in the pharma industry in this country. Perhaps the drug being tested in the trial I was involved in was somewhat effective, but deemed not effective enough to justify the cost of development. I don't know. I do know that the profit motive very often takes precedence over concern for the well-being of patients. I would encourage anyone who is considering participating in a clinical trial to understand your rights as a participant, to read the fine print very carefully, to have a doctor who is not part of the trial, and most importantly, to understand that you will need to be your own best advocate.

Jan. 10 2012 01:30 AM
Elaine from usa

Grace, we do harness the placebo effect. The only time the placebo effect is separated out from the therapeutic benefit is during a double-blind, placebo-controlled trial. The biggest placebo effect of all, is seen when we know we are taking something that is sanctioned as therapeutic, which would encompass anything that's been approved by the FDA (and is used to treat something that is amenable to placebo effect, like pain, depression, asthma, and Parkinson's), and plenty of things that are over-the-counter, as well. One study showed that actual aspirin packaged with a well-known brand label provided more pain relief then the same aspirin packaged with no brand label, which in turn provided more pain relief than placebo packaged as if it were the branded aspirin, which in turn provided more pain relief than placebo packaged with no branding (Branthwaite 1981).

Jan. 10 2012 12:12 AM
Elaine MacDonough from Scranton, PA

Just adding to Dr. Alterman's comments, participants in phase 1 trials are usually healthy – the reason for this is because phase 1 trials are open label, i.e., both participants and researchers are aware that everyone is getting the real thing, which creates considerable potential for both placebo response and researcher bias. So, since phase 1 trials are only intended to determine safety, not efficacy, it makes sense to use healthy volunteers. However, when the therapy in question involves brain surgery, healthy volunteers no longer make sense, and so in these particular trials the participants had Parkinson's. Moreover, data on efficacy was collected and publicized, neither of which should have been done because of the large potential for placebo effect and bias. It seems to me the phase 2 trials functioned exactly as they are supposed to, and revealed that any benefit that had been seen to date was placebo effect. Thanks Dr. Alterman for explaining that.

Jan. 09 2012 10:28 PM
Lherman from Amherst NY

Beryl Benderly, and the Parkinson's patient advocates she talked about are not seeking to eliminate placebo controlled trials. They are saying that we need to bettter understand how the placebo effect (which is especially strong in Parkinson's Disease ) might effect the outcome of the trial. We need to know how long it could last. The usual 6 months for a phase II trial just might not be long enough. Otherwise we run the risk of halting trials prematurely and keeping potentially effective treatments from ever reaching clinical use. Type II errors (false negative outcomes) are especially devastating to People with Parkinson's because we have no effecive long=term or neuroprotective treatments available. We believe trial design should be adapted to harness the power of the placebo effect, resulting in more effective treatments.

Jan. 09 2012 03:10 PM
sue from NYC

Was the doctor who was on the line from Boston's Beth Israel Hospital a scheduled participant in the discussion or did he call in as a listener? I think he called in as a listener, and if this is correct, it underscores my point that this show did a terrible job educating listeners about the issues pertaining to testing drugs for efficacy. So many issues were completely ignored. For example, without double blind and other testing to make sure a drug really works, imagine chemicals would be pushed on the unknowing public to cure ailments, both serious and minor, and there'd be no reliable way to predict outcomes. I suppose this would work in Ron Paul's preferred universe----caveat emptor.
Leonard, what was THAT about?

Jan. 09 2012 01:27 PM
concerns-at-sloankettering from new york city

One must be very, very w(e)ary of the need to follow the [potential] money! Even high profile Institute(s) with non pareil r'aison d'etres and their powerful non-clinical researchers are apt to drop their guard, commit craven self-serving acts evincing an egregious conflict of interests and let the forces of commerce influence (potential) enrollment in clinical trials when large 'milestone achievement' royalty payouts are on the line.

“Sloan is pursuing a systemic approach to reducing expenses and increasing revenues […] One example of this is discouraging terminally ill patients from seeking initial treatment or second opinions from the cancer center […] the admission of such patients is counterproductive […] to Sloan Kettering.” [paraphrasing salient features, MSKCC, CFO/Chief Financial Officer]

Jan. 09 2012 01:14 PM
Milton from queens

@Grace: It's an ethical question, how do you disclose to your patient that you are putting them on placebo therapy. Frankly if there's a way to make money from something even placebos I'm sure the pharmaceuticals will find it (it should be noted how pharmaceutical companies are aggressively moving into the supplement market.)
There is a field in which the placebo effect works wonders, acupuncture. Placebos have an effect on certain things like pain, nausea, fatigue etc... but I have not heard of any effect that placebos have on "real disease" such as infections, HIV, cancer etc (outside of relieving symptoms like pain, nausea, fatigue etc...).

Jan. 09 2012 12:54 PM
Amy from Manhattan

Not to be (too) facetious, but if you want to study the placebo effect, what would you test it against?

Jan. 09 2012 12:41 PM

Hear, hear Grace.

Jan. 09 2012 12:39 PM
pete from maplewood

What this... uh... woman... uh... is saying... may be... the uhhhh... interesting thing... uh, uh. uh... but I uh... am turning... it... uh off because... I uh. um. um um... uh... sorry, but uh... I um can't uh... stand, uh um, um um, uh... it any more.

Jan. 09 2012 12:38 PM
ellen from NYC

Thalidomide is the reason I've been in a complete remission from chronic lymphocytic leukemia for almost 8 years.
Because of the tragic results of its former use, looking at the positive workings of that drug, namely preventing certain kinds of cell development, were not done for many, many years.

Jan. 09 2012 12:33 PM
Amy from Manhattan

Leonard, I'm sorry your friend died, & I'm wondering, do you know if the drug in the study she was in ever became an accepted treatment?

Jan. 09 2012 12:33 PM
grace from upper west side

why don't we hear of anyone trying to harness the placebo effect if it's so strong. No side effects, and it's free. Or is that the problem--that nobody stands to get rich from it?

Jan. 09 2012 12:25 PM
ellen from NYC

most phase 2 trials do not involve a placebo -- use of placebos are frowned upon in general. The comparison drug
is usually a drug that's currently in use in treatment of the malady. The point to see whether the new drug or combination of drugs is more effective.

In that way, no patients are denied treatment. But it also creates a problem: the FDR won't improve a trial if there isn't sufficient evidence that the new drug will accomplish something the older drug does not -- or that it can do the same thing, but do it in a way that is not as harmful -- i.e. the new drug lenalidomide was approved for trial because although it worked in the same way as thalidomide in treatment for certain types of leukemia (mine) it did not cause a particular side-effect that limited the use of thalidomide.

Jan. 09 2012 12:22 PM
Larry from Williamsburg

Your guest is doing a pretty good job of explaining how placebos might work. As a neuroscientist, I would like to add that it might be better to call it an "expectation" effect or "relaxation" effect. It appears that expecting benefits reduces stress/anxiety which can reduce symptoms (as we know stress can increase those same symptoms). Thus, placebo effects ARE biological effects. The control (phase 2) testing should be thought of as separating pharmacological (no biological) effects from expectation.

Jan. 09 2012 12:18 PM

Leave a Comment

Email addresses are required but never displayed.

Get the WNYC Morning Brief in your inbox.
We'll send you our top 5 stories every day, plus breaking news and weather.