On the night of my 32nd birthday, my husband and I enjoyed a delicious dinner while on vacation in Orvieto, Italy. To complement my pasta, I enjoyed a single glass of red wine, my first since learning I was pregnant four months earlier. Even now my indulgence inspires periodic pangs of guilt: Did I stunt my son’s potential by sipping that Sangiovese?

Nobody questions the notion that heavy drinking during pregnancy is harmful. It can cause facial abnormalities, central nervous system problems and stunted growth. But evidence regarding the effects of light or occasional drinking is mixed. In five epidemiological studies published in 2012, medical psychologist Erik Mortensen of the University of Copenhagen and his colleagues found that five-year-old children born to women who had one to four drinks a week during pregnancy displayed no deficits in general intelligence, attention or other types of higher-order thinking. On the other hand, in 2011 psychiatrist Nancy Day of the University of Pittsburgh and her colleagues reported that teens born to women who averaged more than one drink a week during pregnancy were twice as likely as those born to nondrinkers to have conduct disorder, a condition characterized by theft, deceit or violence.

The truth is hard to discern because research on the issue is fraught with problems. The ideal type of experiment is not ethical: scientists cannot randomly assign one group of women to drink during pregnancy and compare the outcome with those instructed to abstain. As a result, they must compare what happens to women who choose to drink during pregnancy with those who do not, and these women often differ in important ways. All things considered, having an occasional drink during those nine months—say, one or two a week—probably poses little, if any, harm. Still, some experts warn, light or sporadic drinking may have effects we do not know how to measure.

Messy Methodologies
The thorniest problems with this research involve the ways in which women who drink during pregnancy differ from those who do not. A woman who drinks moderately or heavily is more likely to smoke, use drugs and physically abuse her child than a nondrinker, all of which could worsen her son’s or daughter’s prognosis independent of alcohol. This woman might also have a genetic background predisposing her—and her children—to behavioral problems.

Meanwhile women who drink lightly while pregnant may have protective characteristics. Compared with abstainers, they “are often the more affluent moms, the more educated moms, and the smarter moms as well,” perhaps because they belong to a higher socioeconomic class than teetotalers, says epidemiologist Ron Gray of the National Perinatal Epidemiology Unit at the University of Oxford. This effect could explain why, in a 2010 study, clinical psychologist Monique Robinson and her colleagues at the Telethon Institute for Child Health Research in Perth, Australia, discovered that children of mothers who drank two to six drinks a week while pregnant were less likely to have behavioral problems at age 14 than mothers who abstained. As with most such studies, the researchers tried to account for systematic differences in the groups, but as Robinson acknowledges, “these unmeasured factors may have influenced our results.”

Another complexity: most studies assess maternal drinking through interviews, and pregnant women might lie about or underestimate their consumption out of embarrassment or shame. Lower estimates can mask harmful effects if they cause light drinkers to be incorrectly categorized as abstainers and put in the comparison group. They might, however, inflate the perceived risk if heavy-drinking mothers of children with deficits get incorrectly categorized as light drinkers.

Moreover, definitions of light, moderate and heavy drinking can vary enormously. Investigators who consider light drinking to be up to a drink a day tend to conclude that the practice is more harmful than those who say light drinking means up to one drink a week. (The alcohol content in “one drink” can also differ.) And many studies are based on an “average” number of drinks per time interval, which can lump together dramatically different drinking patterns—for example, imbibing one a week and binging on five in one night in a month. And yet, Day says, “there’s a lot of literature that shows that binge drinking may have a bigger effect.” One 2009 study, for example, tied a single binge during pregnancy to hyperactivity and attention problems in children. Most studies of periodic light drinking, on the other hand, find that it has no effects.

Teasing Out the Truth
Because individual studies so often conflict, coming to a consensus about how much alcohol—if any—is safe for an expectant mother is a tall order. Nevertheless, one psychologist couple, Sandra and Joseph Jacobson of Wayne State University, has assessed a good deal of the published literature in an attempt to answer the question. They say that no obvious neurobiological deficits have been detected in children whose mothers consume less than one drink a day during pregnancy. Still, the data do not prove that a drink a day in pregnancy is without effect, Sandra Jacobson warns; it could be that researchers have not been looking at the right outcomes or that their tools are too insensitive to pick up any changes.

Gray believes the latter to be true. Although he and his colleagues have found that moderate drinking during pregnancy (which they define as an average of two to six drinks a week) has no overall effect on child IQ at age eight—the kids of the drinking moms in his cohort actually had higher IQs than the abstainers’ kids—they did find, using a new genetic technique, some potential for harm.

Humans have multiple genes for proteins that break alcohol down into harmless by-products. Some people have mutations in these genes, causing alcohol to linger in their bloodstream. In a November 2012 paper Gray and his colleagues reported that the more mutations children of maternal drinkers have in these genes, the lower their IQs are. (The effect is weak; the IQs of children with four mutations were only 3.5 points lower than those of kids with two.) More important, however, the mutations have no effect on the IQ of children born to nondrinking women, suggesting that there is at least a minimal danger to prenatal drinking for certain individuals.

Because of such findings and all the other uncertainties, policy makers are unlikely to ever give the green light to occasional drinking during pregnancy. “We will never, ever, ever know how much is safe for every individual,” says biologist Kathleen K. Sulik of the University of North Carolina at Chapel Hill. What is harmless for a woman of one race, weight, nutritional status and genetic background may be dangerous for another. That said, the literature is reassuring to the many women who socially drank before realizing they were pregnant and to those who, like me, had one celebratory glass.

Better Safe Than Sorry. K. K. Sulik, S. K. O’Leary-Moore and E. P. Riley in British Journal of Obstetrics and Gynecology, Vol. 119, No. 10, pages 1159–1161; September 2012.

The Effect of Different Alcohol Drinking Patterns in Early to Mid Pregnancy on the Child’s Intelligence, Attention, and Executive Function. U. S. Kesmodel et al. in British Journal of Obstetrics and Gynecology, Vol. 119, No. 10, pages 1180–1190; September 2012.