With as many as 1 in 10 children with autism ultimately losing their diagnosis, a new government study suggests some kids placed on the spectrum shouldn’t be.

Looking at data from a federal survey of parents of over 1,500 kids diagnosed with autism, researchers found that the developmental disorder may be overdiagnosed in more than 9 percent of cases.

The study, published online this month in the journal Autism, was produced by researchers at the U.S. Centers for Disease Control and Prevention, the University of Washington, the National Institutes of Health and the federal Health Resources and Services Administration.

Researchers looked at parent responses from the 2011 Survey of Pathways to Diagnosis and Service on 1,420 kids with a current autism diagnosis and 187 previously diagnosed children, all of whom were between the ages of 6 and 17.

About 13 percent of kids initially identified on the spectrum subsequently lost the diagnosis, the study found. Most often, parents reported that children shed their label due to new information.

Several factors may be contributing to overdiagnosis, researchers said. It could be the result of imprecise screening and evaluation processes or difficulty distinguishing kids with language issues from those with developmental delays.

In some cases, however, parents admitted that clinicians gave their child an autism diagnosis simply because that label would make services more readily available.

Despite the relatively high rate of children who lost an autism diagnosis, however, researchers said that just a fraction of these kids had in fact “recovered.” Only 3 percent of children who shed their diagnosis did so as a result of “treatment or maturity,” the study found, and it remains uncertain if all of those children were correctly labeled to start.

It’s also unclear whether instances of children losing their diagnosis are becoming more common, according to Stephen Blumberg of the CDC’s National Center for Health Statistics and his colleagues. What’s more, researchers indicated that overdiagnosis alone remains insufficient to explain dramatic increases in autism prevalence in recent years.

“This study confirms that ASD diagnoses can and sometimes do change as children mature and overcome delays, and as new information is assimilated by their health care providers,” Blumberg and his colleagues wrote in their findings. “These changes over time can complicate the use of surveys and retrospective surveillance methods to estimate the current prevalence of ASD, the characteristics of children who currently have ASD and the adequacy of services for this population.”