A fascinating study just published in the medical journal Pain examined the side-effects reported by patients taking placebos in clinical trials to test migraine drugs. It found that side-effects from placebo were almost as common as from the actual drug, but most interestingly, were specific to side-effects you would expected from the comparison medication.
In other words, the side-effects you get from a sugar pill in a study on anticonvulsant drugs closely resemble side-effects you get from anticonvulsants and are different from the side-effects you get from a sugar pill in a study on pain killers, which more closely resemble pain killer side-effects.
The researchers, led by neuroscientist Martina Amanzio, looked at trials for three type of migraine drugs: NSAIDs (like aspirin), triptans that work on the serotonin system, and anticonvulsant drugs more often used to treat epilepsy.
Side-effects from placebo are known as the nocebo effect and just the combined list of side-effects from the placebo groups in this study is surprising enough:
abdominal pain, anorexia or/and weight loss, attention difficulties, burning or/and flushing, chest discomfort, chills, diarrhea, dizziness, dry mouth, dyspepsia, fatigue, heaviness, injection side reaction, insomnia, language difficulties, memory difficulties, nasal signs and symptoms, nausea, numbness, paresthesia or/and tingling, pharyngitis, somnolence or/and drowsiness, stinging or/and pressure sensation, taste disturbance, tinnitus, upper respiratory tract infection, vomiting, weakness
It turns out that when placebo was being compared to an anticonvulsant, side-effects more common in these drugs – like fatigue, reduced appetite, sleepiness and tingling sensations – were more common in the placebo. In contrast, stomach upsets and dry mouth were more common in the placebo group when the comparison was with NSAID painkillers, which more often cause these symptoms themselves.
One explanation may be that before taking part in a clinical trial, patients are informed of the possible side-effects that the active drug may cause, regardless of whether they are going to be given placebo or the actual medication.
Information on the possible side-effects will be specific to the real medication, and, as we know that expectation plays a big part in the placebo effect, it is probably also shaping the nocebo effect and leading to the production of symptoms through expectancy.
Link to PubMed entry for study.
Mind Hacks 
Your post made me wonder about how we learn about medications and their side effects. Then I thought about what I see on TV. Could it be that modern drug advertisements with there recitations of a lengthy catalog of side effects, create an unfavorable expectations? Might people now hold an unconscious litany of possible disasters? If the possibility of taking medication (real or placebo) triggers anxiety then couldn’t the placebo effect work in reverse?
A question more to the point, what would be the neuroscience behind the placebo effect? For me, the emotionality of treatment setting (including receiving a script for the placebo) and the likelihood of frequent subsequent recall of the treatment event(s). Emotions lead to feelings lead to moods, and better moods have demonstrable impact on treatment success.
More placebo stuff from sciencedaily.com: http://www.sciencedaily.com/releases/2009/10/091001081303.htm
This is indeed an very interesting point. It actually links very closely with Jastreboff’s Tinnitus Treatment model of TRT, where he connects peoples side effect (the tinnitus occurring) with their knowledge of the disorders symptoms. Once again very interesting post and more food for thought.