An article at the NY Times looks at research into the "nocebo" effect. Named after the placebo effect, it's the term for when patient expectations do harm, rather than good. "When a patient anticipates a pill’s possible side effects, he can suffer them even if the pill is fake." The article describes several instances of patients getting the placebo in a drug trial, but reporting the expected side effects of the drug, rather than the benefits or nothing at all. Quoting: "Consider the number of people in medical trials who, though receiving placebos, stop participating because of side effects. We found that 11 percent of people in fibromyalgia drug trials who were taking fake medication dropped out of the studies because of side effects like dizziness or nausea. Other researchers reported that the discontinuation rates because of side effects in placebo groups in migraine or tension drug trials were as much as 5 percent. Discontinuation rates in trials for statins ranged from 4 percent to 26 percent. ... In one remarkable case, a participant in an antidepressant drug trial was given placebo tablets — and then swallowed 26 of them in a suicide attempt. Even though the tablets were harmless, the participant's blood pressure dropped perilously low."