The Danish researchers questioned mothers of almost 100,000 children born in Denmark between 1997 and 2003 about their history of infection, influenza and antibiotic use during pregnancy.
The results initially seem significant:
Women who reported having the flu while pregnant were twice as likely to have kids who developed autism.
A week-long fever indicated a tripled risk. But, as the authors themselves point out, "misreporting of influenza is likely to be considerable," and "The results may be due to multiple testing; the few positive findings are potential chance findings."
Others questioned the comparisons drawn in the study: "The more comparisons you make, the more likely some difference will look important when it's not," Emily Willingham wrote for Forbes.