Placebos cannot replace antipsychotics

As I’ve discussed previously, placebo effects are mostly a myth, and if a new drug has an effect barely above that of a mythical placebo effect, it’s considered a failure. In a recent article in Reuters Health, Rising placebo response seen in schizophrenia trials, Amy Norton states that clinical trials of anti-schizophrenia drugs, in a class of drugs called antipsychotics, are finding lesser effects because patients are responding positively to placebos (that presumably does not contain anything but sugar).

Treating schizophrenia or any psychosis is difficult because different patients respond in different ways to each drug. For some individuals, these drugs can treat many of the symptoms of schizophrenia, including hallucinations and delusions, which allows them to live relatively normal lives. But for some individuals these same drugs have significant side effects, including sedation, weight gain, and hyperglycemia (which can be serious for a diabetic). Eventually, individuals stop using the drugs because of the side effects and their psychotic symptoms return.

Norton also reports that current clinical trials of 2nd generation antipsychotics, which emerged about 20 years ago and are most prescribed for these symptoms, have smaller treatment effects compared with the results of the initial trials from the early 1990s. In a meta analyses recently published in the Journal of Clinical Psychiatry, researchers at the Food and Drug Administration (FDA) looked at 32 clinical trials that were submitted to the agency between 1991 and 2008 as a part of the New Drug Applications submitted to the FDA for clearance.

The researchers discovered that trials done in recent years have turned up smaller benefits than was shown in older studies. In the 1990’s, the overall success rate for these drugs was around 85%, but dropped to around 74% during the 2000’s. Furthermore, an increasing placebo response and a diminishing treatment effect (drug-placebo difference) was observed in trials comparing the 1991-1998 period versus the 1999-2008 period, with a decrease in treatment effect of over 45%. It wasn’t that the drugs were actually less effective, but because the placebos were showing a bigger positive response.

But placebos don’t have any clinical value, so it’s almost inconceivable that suddenly we have a better quality of placebo. It’s still a sugar pill, so what is happening?

I think the first thing to look at is study design. There are some differences in the population of schizophrenics between trials. Some have more women, different ethnic groups or other medical differences in the study group. Maybe the trials chose patients with stronger or lesser symptoms of schizophrenia–those with worse symptoms might actually improve slightly with a placebo, just out of randomness in a study. Are we testing for the right “end points” of the study? For example, maybe it’s more than just relieving of symptoms but better standard of life (all subjective). 

But it also could be that the original studies were too small, allowing bias to inter the design. In other words, the study was developed to meet our expectations for the drug rather than testing for a reasonable hypothesis.

In the end, there’s still not a placebo effect.

The Original Skeptical Raptor
Chief Executive Officer at SkepticalRaptor
Lifetime lover of science, especially biomedical research. Spent years in academics, business development, research, and traveling the world shilling for Big Pharma. I love sports, mostly college basketball and football, hockey, and baseball. I enjoy great food and intelligent conversation. And a delicious morning coffee!