'Publication bias' casts doubt on value of antidepressants for autism

That "publication bias" may mean that physicians believe the medications -- known as selective serotonin reuptake inhibitors (SSRIs) -- are more effective than they really are for children with these behaviors. Indeed, when researchers combined the data from published studies and those that never made it into print, the new analysis showed that SSRIs don't help repetitive behaviors much at all.

"At least from what we have right now, we need more information to determine if SSRIs are useful in treating repetitive behavior," said study author Melisa Carrasco, of the neuroscience graduate program at the University of Michigan in Ann Arbor. "I don't think we're trying to say they should not be used at all in autism. There is some compelling evidence toward their use in treating anxiety disorders in autism."

The study, in the May print issue of Pediatrics, appears online April 23.

The study also calls into question the effectiveness of current methods used to evaluate drugs, particularly pediatric drugs, in the United States. One expert said the findings have implications for countless other drug trials for other conditions.

It's long been recognized that drug trials that show the drug is effective are more likely to be published in peer-reviewed journals, where the results are widely read and disseminated to doctors, said Dr. Scott Denne, a professor of pediatrics at Indiana University School of Medicine, who wrote an accompanying editorial.

"Positive studies are exciting and potentially groundbreaking. Negative studies are not particularly exciting and at least in the estimation of both physicians and investigators, they don't really change anything, even though that isn't necessarily true," Denne said.

Current U.S. law requires that investigators submit a summary of the results of drug trials on ClinicalTrials.gov, a national registry of clinical studies. But often, researchers don't submit their results, and the information is never published on the government website, he said.

"A substantial number of trials are not having the results posted anywhere," Denne said. This deprives pediatric researchers and the public of valuable information, and may mean that trials are unnecessarily repeated.

Carrasco and her colleagues searched PubMed (a U.S. National Institutes of Health database) and ClinicalTrials.gov for randomized, double-blind, placebo-controlled trials (considered the gold standard of research) on using SSRIs to treat repetitive behavior in people with autism. Researchers identified five published trials and five unpublished trials marked as completed on ClinicalTrials.gov.

A meta-analysis (pooled analysis) of the published studies found a small but significant improvement in repetitive behaviors, including obsessions and compulsions, among autistic kids treated with SSRIs. When the unpublished studies were included, that benefit vanished.

About one in 88 U.S. children has autism, a neurodevelopment disorder characterized by problems with social interaction, communication and restricted interests and behaviors. That includes repetitive behaviors, such as arm-flapping or head-banging; having an obsessive interest in one topic; having a need to stick to a specific ritual or routine; and experiencing distress or agitation when that routine gets disrupted.

More information: To learn more about autism treatments, visit Autism Speaks.

Copyright © 2012 HealthDay. All rights reserved.