Skip to main content

In Defense of Sham Treatment

There is an ethical dilemma in neurosurgical trials regarding the use of sham surgeries as placebo controls. There have been countless instances of treatments that showed promise during preliminary trials, but failed to move past phase II clinical trials when the treatments proved ineffective compared to sham controls (Freed et al., 2001 & Olanow et al., 2003). The article “Why Fake It? How ‘Sham’ brain surgery could be killing off valuable therapies for Parkinson’s disease” discusses the issues regarding using shams in neurosurgical studies particularly in Parkinson’s disease.

There is one group that encourages the use of shams as the only true way to statistically prove that a treatment is effective. Without proper double blind, control studies, how can you move a drug past clinical trials and not be worried it won’t actually help anyone?

The other group discourages the use of shams arguing that they are expensive, yield abundant false negatives, and can potentially cause detriment to the patients. Sham controls needed in clinical trials are extremely expensive. Every part of the surgery needs to be identical for a sham except for drug infusion into the brain. Therefore, money is spent to bring the patient in, attach a stereotaxic frame to the skull, pay a full staff in the operating room complete with turning on machines to make noises present during surgery and drill holes into the skull of the patient. There are also huge hoops to jump through ensuring no one outside of the OR knows the treatment the patient received.

Many treatments given in phase I trials to a handful of patients show promising results that then do not show effectiveness compared to controls in phase II trials. It is argued that sometimes it takes several years for treatments to become effective and that some treatments are being prematurely discredited because of these trials (Politis et al., 2010). Maybe if some of these trials were extended, more treatments might prove effective.

The last main argument is that there may be psychological effects of un-blinding that are detrimental to the patients. Some patients receive the sham and will need to be told at the end of the trial that they were not given the treatment. These patients may have been feeling better due to the placebo effect, where just the belief you could be getting an effective treatment helps you, but when told they did not get the treatment their condition can worsen due to different neurochemical effects. A benefit to consider is that if there were a positive effect of drug seen, these patients would already have the holes drilled needed to administer the drug at a later date. It is unfortunate that some patients may have negative effects of un-blinding, but all patients take a risk during a clinical trial that they will be given the placebo.

One reason this is being to fervently argued is that the placebo effect has especially strong effects in Parkinson’s patients. Sham treatments cause placebo effects in these patients because their expectation to receive treatment yields a release of dopamine in the brain, which is lacking in this disease state. This is why so many drugs are discredited. You cannot see an effect of the treatment if the placebo effect alone is contributing to fixing the issue in patients. Some doctors noted that they don’t want to include controls because they don’t want to exclude the placebo effect. It is, after all a natural part of treatment and any individual in a trial will show the effect of placebo because they do not know what their treatment was.

This article focused on the reasons to stop using shams, but I felt that some of these arguments leant themselves just as nicely to why we should continue to use sham controls. Just as the doctors argue that they do not wish to exclude placebo effects and therefore, do not wish shams, I believe that if we keep using shams to find out what treatments are most effective we may one day discover some combination of placebo and treatment that is doubly effective.

Secondly, doctors say all patients will show some placebo effect, even a patient taking a pill over the counter. They use this as an argument against shams, but I find it one of the better arguments to continue using them. If all patients have some placebo effect, and in phase I trials patients feel better, but in phase II trials the control and treatment groups are the same, you cannot argue that both groups were just experiencing the placebo effect. This is the only way to know for sure that the benefit your treatment is yielding is not due to the placebo effect.

However, I strongly believe if the treatment is ineffective and the placebo effect is the only reason patients are feeling better, treatment should continue with a placebo. Why perform treatment surgeries due to cost if the placebo works just fine? No one may need to purchase/manufacture that drug if sham patients are seeing real improvement. The placebo effect is a natural part of treatment and maybe sometimes its benefit alone is enough to help a very sick patient. Just treat with shams.

–Lauren DeBrouse
Neuroscience Graduate Program

Want to cite this post?

DeBrouse, L. (2011). In Defense of Sham Treatment. The Neuroethics Blog. Retrieved on
, from


Emory Neuroethics on Facebook