The share of pregnant women reporting recent alcohol use increased in the U.S. over more than a decade, according to a new national study.
Reported alcohol use rose from 9% in 2011-2012 to a peak of 15% in 2021-2022. It remained near that level at 14.5% in 2023-2024, suggesting the long-term increase may have recently stabilized but has not clearly reversed.
The study, published in JAMA, also found increases in binge drinking and heavy drinking. By 2023-2024, 4.6% of pregnant women reported binge drinking and 2% reported heavy drinking during the previous month.
The results do not show that everyone counted in the analysis knowingly drank after learning they were pregnant. The survey did not collect information about pregnancy stage, when participants recognized their pregnancies or exactly when during the previous month they consumed alcohol.
That distinction matters because some alcohol exposure occurs before a person knows they are pregnant.
Researchers at Columbia University Mailman School of Public Health analyzed data from the Behavioral Risk Factor Surveillance System, a telephone survey sponsored by the Centers for Disease Control and Prevention.
The study included 33,729 pregnant women ages 18 to 49 who participated between 2011 and 2024 and answered questions about alcohol use during the previous 30 days. Researchers also examined trends among nonpregnant women of reproductive age for comparison.
Current alcohol use was defined as consuming at least one drink during the past month. Binge drinking meant having four or more drinks on one occasion, while heavy drinking was defined as consuming eight or more drinks per week.
Because the “current use” category included anyone who reported at least one drink, it combined a wide range of drinking patterns. Someone who drank once before recognizing a pregnancy would be counted in the same broad category as someone who drank more regularly.
“Alcohol consumption in the general population has changed considerably in recent years, particularly during the COVID-19 pandemic, but national estimates of prenatal alcohol use did not extend beyond 2020,” said Emilie Bruzelius, the study’s first author and a postdoctoral research fellow in epidemiology at Columbia.
When the researchers examined trends across successive two-year periods, they found that the odds of current alcohol use, binge drinking and heavy drinking increased over the full study period.
The results do not mean alcohol use rose steadily every year. Current drinking reached its highest reported level in 2021-2022 and declined slightly in the final two-year period. Still, the most recent estimate remained considerably higher than the 2011-2012 level.
The analysis also found substantial overlap between higher-risk drinking and tobacco use. Among pregnant women who reported binge or heavy drinking, 68% also reported tobacco use.
That finding may help identify people who could benefit from support addressing more than one substance. It does not show that tobacco use causes alcohol consumption or that alcohol causes tobacco use.
No amount of alcohol has been established as safe during pregnancy. Alcohol can pass through the placenta, and prenatal exposure can affect fetal development. The level of risk can vary depending on factors including the amount consumed, the drinking pattern and the timing of exposure.
“While no amount of alcohol has been established as safe during pregnancy, these increases across all patterns of drinking are concerning,” said Silvia Martins, the study’s senior author and a professor of epidemiology at Columbia.
The study did not examine pregnancy outcomes, fetal alcohol spectrum disorders or child development. It therefore cannot show how many pregnancies involved harmful exposure or connect the reported drinking patterns with specific outcomes.
The findings also rely on self-reported information. Because drinking during pregnancy is sensitive and stigmatized, some participants may have reported less alcohol use than they actually consumed. Others may not have remembered their drinking accurately.
Telephone surveys may also miss people who are difficult to reach or less likely to participate. Although the data were weighted to produce national estimates, the percentages should be understood as estimates based on reported behavior rather than exact measurements of all prenatal alcohol exposure.
The study followed population trends over time, not the same individuals throughout the 13-year period. Different groups of women participated during each survey cycle.
The results should not be used to shame people who consumed alcohol before realizing they were pregnant. Pregnancy recognition does not always happen immediately, and unplanned pregnancies are common.
Instead, the findings support routine, nonjudgmental conversations about alcohol in reproductive and prenatal health care. Those discussions may be especially helpful before pregnancy, early in pregnancy and when someone is trying to stop or reduce drinking.
They also highlight the importance of making support accessible. People may continue drinking for many reasons, including alcohol dependence, mental health needs, stress, social environments or a lack of clear information about pregnancy.
For someone who drank before learning they were pregnant, stopping after pregnancy recognition can prevent further exposure. A health care professional can help assess the situation without assuming that one reported drinking occasion led to a particular outcome.
The study cannot explain why reported alcohol use increased. Changes in drinking patterns during the COVID-19 pandemic may have contributed, but the analysis was not designed to identify the causes of the trend.
It also cannot determine whether the recent leveling represents a lasting shift. Additional years of data will be needed to see whether reported alcohol use during pregnancy begins to decline or remains elevated.
The study was funded by a training grant from the National Institute on Drug Abuse within the National Institutes of Health.