Consumer Reports on Thursday said its report on infant car seat safety was not accurate because the test crashes were run at speeds higher than stated.

The magazine said on Jan. 4 that the car seats "failed disastrously" in crashes at speeds as low as 56 km/h, but, in a statement, the organization said it was pulling its original report, noting it would publish a new study with possible revisions.

"We withdrew the report immediately upon discovering a substantive issue that may have affected the original test results," the organization said in a statement.

"Our tests were intended to simulate side crashes at the NCAP speed of 38 mph [61 km/hr]. The new information raises a question about whether the tests accurately simulated that speed, however, so we are now reviewing our tests and the resulting article."

Nicole Nason, an NHTSA adminstrator, said Consumer Reports was right to withdraw its study.

"I was troubled by the report because it frightened parents and could have discouraged them from using car seats," she said.

The original report said seats came off their bases or twisted in place. The magazine said in one case, a test dummy was hurled more than nine metres.

In mid-January, Transport Canada issued a statement saying that its investigations and analyses did not support the Consumer Reports findings.

"At this time, Transport Canada has no evidence to suggest a safety issue exists," the federal agency said.

With files from the Associated Press