The American Academy of Pediatrics has renewed advice to pregnant women: do not drink alcohol, not even a little bit, not at all.

The update of an old warning from the US surgeon general is meant as a best practices paper for clinicians, but also a warning to American mothers-to-be, the paper’s lead author said.

“The real measure [of how safe alcohol is] is how much are you willing to compromise the potential health and well-being of this baby,” said Dr Janet Williams, lead author of the survey to be published in the journal Pediatrics on Monday. She emphasized there is no known “safe” level of alcohol consumption.

Since the 1970s, researchers have associated prenatal drinking with neurocognitive and behavioral problems, as well as distinct facial deformities. The group of symptoms are known collectively as fetal alcohol spectrum disorders, or FASD (the most severe of which is fetal alcohol syndrome).

But the admonition to American women is muddled by a lack of knowledge about the disease and varying international standards. Perhaps most surprising, the highest prevalence of drinking among pregnant women remains among college-educated, middle-aged women.

“If a patient admits to drinking a glass of wine perhaps before she knew she was pregnant, there’s no reason for panic,” said John Hannigan, a professor of obstetrics and psychology at Wayne State University in Michigan. “And the correct and clinical recommendations are not to worry but stop your drinking.”

One in 10 women drink while pregnant (defined as having one drink in 30 days), and one in 33 pregnant women report binge drinking, according to the Centers for Disease Control and Prevention (CDC).

The syndrome has been blamed for a range of troubles, from specific facial deformities such as a thin upper lip and small eyes, to behavioral and developmental problems manifesting as ADHD, poor impulse control, impaired memory skills, problem-solving difficulties and trouble understanding mathematics.

But diagnostic tools are time-consuming to use, and the number of children affected is an estimate.

For public health researchers, the most contested question remains what amount of alcohol causes severe developmental impacts. In almost every study, binge drinking, or having more than four drinks in one sitting, is considered dangerous.

“There’s been some controversy over this whole issue,” said Dr Christina Chambers, a perinatal epidemiologist who is part of an NIH grant consortium. As a professor, she is working to establish the prevalence of FASD in San Diego at the University of California there.

“Women don’t want to be told to avoid alcohol entirely,” she said, echoing questions women often ask: “‘What’s the evidence that low levels of alcohol are harmful?’ and ‘You’re just telling me don’t do this, don’t do that,’ and some obstetricians seem to be comfortable with women drinking at low levels during pregnancy.”

“We really just don’t know a threshold below which it’s safe to drink, and trying to pick a number is not a safe thing to do,” Chambers said.

Internationally, standards vary for where to draw the line. Is one drink all right in the second trimester? One a week but only early in the pregnancy?

In the UK, the Royal College of Obstetricians and Gynecologists recommends women don’t drink at all in the first three months of pregnancy, but says up to two drinks per day twice per week seems acceptable. Danish articles, by comparison, found no cognitive differences between five-year-old children whose mothers drank moderately in early pregnancy and those whose mothers didn’t. The Australian National Health and Medical Research Council also recommends pregnant women don’t drink while pregnant or trying to become pregnant.

“There have been some studies recently coming out which look at ‘light drinking’, using a very specific definition of what that might be, has no detectable intellectual problems,” said Williams. “Well, that also might mean that we’re not capable of detecting [abnormalities]. We don’t have sensitive enough methods to detect what’s there. So, I think it’s a leap in faith to say it’s safe, it’s completely safe.”

The most prominent studies Williams referred to are by Danish researchers: a group of five papers, funded primarily by the CDC, that found low and moderate alcohol intake (between one and eight drinks per week) “is not associated with adverse neuropsychological effects”.

Those results go against strict, long-held advice by American doctors. The surgeon general’s first warning about drinking while pregnant was issued in 1981, followed by a 1989 congressional mandate that all alcohol bottles be labeled as potentially dangerous for expectant mothers.

In the field, most ongoing research funded by the National Institutes of Health (NIH) is looking for genetic markers that could predict whether some children are especially susceptible to, or protected from, FASD.

In the United States, a median of more than half of women of reproductive age drink alcohol, and 51% of all pregnancies are unintended. Even if a woman does not intend to drink while pregnant, she may do so unintentionally.

The federal government pours roughly $34m into NIH research on the disorders, but surprisingly few basic facts about how fetal alcohol spectrum disorders impact the American public are known.

The number of children affected is estimated at anywhere between 0.2 infants per 1,000 to nine children per 1,000. And there is no simple lab test for FASD – Canadian researchers found it can take up to 47 hours to diagnose one child with the range of symptoms that indicate the disease, because of the skill tests and interviews needed to do so.

The highest prevalence of drinking while pregnant is found among a surprising demographic: college-educated, unmarried women between 35 and 44 years old, according to the CDC.

Those findings seem to have held steady for years – in Elizabeth Armstrong’s book about fetal alcohol syndrome, Conceiving Risk, Bearing Responsibility, an analysis of 1988 data found women’s odds of drinking while pregnant increased with education and income level. At the time, a woman with a graduate degree was twice as likely to drink while pregnant as a woman who had not completed high school.

Still, the academy’s clinical bulletin recommends pediatricians be vigilant of FASD in children with a history of involvement with social services, and even reminded clinicians they are required to report cases of FASD to child protective services.

Anecdotally, some women said they felt “smart enough to question the American medical standard,” as Cosmopolitan’s Michelle Ruiz wrote in a piece proclaiming why she, and others who refused to go on the record, had had a glass of wine while pregnant.

“Family friends didn’t put a wine glass by my place because they assumed I wouldn’t be using it. ‘Why risk it?’ is a popular question posed to a pregnant woman sipping a half-glass of Cabernet,” Ruiz wrote. “That’s just it: I knew that drinking heavily during pregnancy would be a terrible thing and could cause Fetal Alcohol Syndrome, but I didn’t – and still don’t – think I was risking my baby’s health by drinking small amounts on a sporadic basis.”