The conclusion came in an 81-page report requested by two federal agencies to address the doubts raised about the safety of childhood vaccines.
The 14-person panel urged more research on autism, but said further pursuit of possible links between vaccines and the neurological disorder probably is not worth the money or effort.
Critics said the final proof may come if autism diagnoses drop now that thimerosal has been virtually eliminated in routine childhood vaccines.
Autism is a complex developmental disorder best known for impairing a child's ability to communicate and interact with others. Recent data suggest a tenfold increase in autism rates in the past decade, although it's unclear how much of the apparent surge reflects better diagnosis.
Reports published in 2001 by the Institute of Medicine found no connection between the MMR vaccine and autism, and insufficient evidence to draw conclusions about thimerosal, a mercury-based preservative added to multiple-dose vials of vaccine. Since then, enough new studies have been published to reject both theories, the panel said.
Especially convincing were a Danish study showing no difference in the rate of autism between children who got thimerosal-containing vaccines and those who did not; and a British study showing no relationship between the introduction of MMR and autism rates, or between the timing of a vaccination and the onset of autism symptoms.
"The vaccine hypotheses are not currently supported by the evidence," wrote the panel, consisting of physicians, neuroscientists, epidemiologists, statisticians and a nurse.
In a telephone briefing, the chairwoman of the committee, Marie McCormick, of the Harvard School of Public Health, said her advice to parents is that children "should be getting their vaccines" because the life-threatening infections they protect against "are only a plane ride away."
While high doses of mercury can cause neurological damage, there's no evidence this type of damage causes the symptoms specific to autism, the panel concluded.
On the other hand, genetics play a role in autism, and several studies show clear signs of prenatal onset of the disorder, the report noted.
The panel said five observational studies "consistently provided evidence of no association" between thimerosal and autism in Sweden, Denmark, the United States and Britain. A similar number that do suggest a link are small, uninterpretable, unpublished or poorly designed, it said. Fourteen studies, including nine controlled ones, found no association between MMR and autism. Three that found a connection are poorly designed or offer very indirect evidence that could be explained in other ways, the authors said.
A leading vaccine skeptic, Rep. Dave Weldon, R-Fl., a physician, criticized the report as "premature, perhaps perilously reliant on epidemiology, based on preliminary incomplete information."
The Institute of Medicine committee has produced eight studies on vaccine safety at the request of the National Institutes of Health and the Centers for Disease Control and Prevention. This is the last, and the group will disband.
Material from The Associated Press is included in this report.