The warnings have been ubiquitous for decades: drinking alcohol while pregnant could have severe impacts on the unborn child.

Now, a new study of thousands of first graders in four different regions of the United States contends the rates are much higher than previously believed – perhaps five times the most recent estimates.

The study, published in the Journal of American Medical Association, has already indicated that the effects of alcohol on fetuses are statistically comparable to other development disabilities.

“Our results suggest that the rate of FASD (fetal alcohol spectrum disorder) in children in the Untied States is as high or higher than autism spectrum disorders (ASD),” said Christina Chambers, co-principal investigator, a pediatrician at the UC San Diego School of Medicine.

The CDC rate of autism currently stands at 14.6 per 1,000 children the age of 8.

The new analysis of FASD finds the rate in the four regions studied was as low as 11, and as high as 50 children, per 1,000. (Those are the conservative estimates, they add).

The four sites were unnamed school districts: a Midwestern community of 172,000 total people; a Rocky Mountain community, pop. 60,000; a Southeastern site with 206,000 people; and a Pacific Southwest urban city of 1.4 million.

From 2010 to 2016, the study team assessed 13,146 first graders in the four regions. Of those, about half – 6,639 children – were selected to participate in the full assessments. Those assessments included measurements of the body (weight and height and head circumference), an examination for tell-tale facial features and other anomalies that can signal fetal alcohol syndrome, and neurodevelopmental examinations using cognitive and behavioral tests.

The correlation to alcohol consumed while the child was still in the womb was made through interviews with the mother, either in person or over the phone. (Consent rates overall ranged from 36.9 percent to 92.5 percent at the various sites – but the average was 59.9 percent).

Among the 6,639 children, a total of 222 children were classified with a fetal alcohol spectrum disorder. Only two of these children had previously been classified as such, the paper reports. (Although only 27 of the cases were the criteria for the most debilitating variety of the disorder, fetal alcohol syndrome).

The conservative rate at the four sites was between 11 per 1,000 and 50 per 1,000 children, they calculated. But the weighted rate is much higher still – from 31.1 to 98.5 per 1,000 first graders, they report.

The limitations include the sometimes-low consent rates, and the lack of a definitive biomarker for FASD – which could “overestimate” prevalence, they write.

“These findings may represent more accurate U.S. prevalence estimates than previous studies, but may not be generalizable to all communities,” they add.

An accompanying JAMA editorial contends the findings are a new call to completely cut out any drinking while pregnant, regardless of amount.

“As suggested by the American Academy of Pediatrics, the message about alcohol use during pregnancy to the public should be clear and consistent: there is no safe amount, time, or type of alcohol to drink during pregnancy or when trying to get pregnant,” the editorial argues.

But there is bound to be some skepticism. As early as 2000, some experts were calling the FAS “epidemic” a “moral panic” that had its roots in historical cycles of morality including the Temperance Movement. In a paper in the journal Alcohol and Alcoholism, Ernest L. Abel and colleague Elizabeth M. Armstrong argued that while FAS was a social problem, it was not a sweeping epidemic as had been popularly depicted.

“A small proportion of women of children-bearing age, especially those who are most disadvantaged by poverty, bear the greatest burden of risk for FAS,” the scientists wrote. “If we hope to reduce the incidence of this birth defect, we must reconstruct the problem not as a moral panic, but as a moral imperative to find and help those women most at risk of adverse outcomes.”