pregnant person sitting at a table with cigarette in hand and glass of wine

Stillbirth Risk Triples When Women Smoke and Drink During Pregnancy

The combination of smoking and drinking during pregnancy nearly triples the risk of late stillbirth, according to a new study conducted by the Prenatal Alcohol in SIDS and Stillbirth (PASS) Network.  

The study was published online in JAMA Network Open

“Smoking and drinking are among the most harmful exposures to the fetus, and the combination of these two behaviors, which often go hand in hand, can be particularly lethal,” says William Fifer, PhD, professor of medical psychology in the psychiatry and pediatrics departments at Columbia University Vagelos College of Physicians and Surgeons and a co-author of the study. 

“The public health message is that women should stop smoking and drinking as soon as possible during a pregnancy and ideally before they even become pregnant,” adds Fifer. “Any exposure to smoking or drinking can be dangerous to the developing fetus.” 

Each year, about 24,000 stillbirths occur in the United States. Late stillbirths occur after 28 weeks of gestation.  

The new study is the first large prospective study of the effects of maternal smoking and drinking on pregnancy outcomes. The researchers followed nearly 12,000 pregnancies among more than 8,500 women in Cape Town, South Africa, and the U.S. Northern Plains between 2007 and 2015.  

The study found that the risk of late stillbirth was 1.6 times higher for women who drank after the first trimester and 2.2 times higher for those who smoked after the first trimester compared to those who quit after that period or never smoked or drank during pregnancy.  

Women who both drank and smoked after the first trimester were 2.8 times more likely to experience a late-term stillbirth compared to those who quit or didn’t drink or smoke. 

“The combination of drinking and smoking appears to have an additive effect on fetal health,” Fifer says. 

The researchers suspect that the combination of drinking and smoking is toxic to the developing brain and compromises placental function. “When the placenta doesn’t do its job, the fetus is starved of oxygen and nutrients, which can have a wide array of negative consequences,” says Fifer.  

In an earlier study, the PASS Network found that the risk of sudden infant death syndrome (SIDS) was 12 times higher in pregnancies exposed to a combination of maternal drinking and smoking past the first trimester. The network is currently looking at the effect of the combination on the risk for autism and other neurodevelopmental disorders. 

The study involved researchers from more than a dozen research institutions in South Africa and the United States. Fifer was involved in the creation of the PASS Network and led the physiological assessment team of the study with Michael Myers, PhD, professor of clinical psychiatry. Fifer and Myers are also with the New York State Psychiatric Institute. 


More information

The study is titled “Association of Prenatal Exposure to Maternal Drinking and Smoking With the Risk of Stillbirth” and was published Aug. 24 in JAMA Network Open.

Kimberley Dukes, Boston University School of Public Health, is the paper’s corresponding author. 

All authors: Hein Odendaal (Stellenbosch University, Cape Town, South Africa), Kimberly A. Dukes (DM-STAT Inc, Malden, Massachusetts and Boston University School of Public Health), Amy J. Elliott (University of South Dakota School of Medicine), Marian Willinger (National Institute of Child Health and Human Development), Lisa M. Sullivan (Boston University School of Public Health), Tara Tripp (M-STAT Inc and Boston University School of Public Health), Coen Groenewald (Stellenbosch University), Michael M. Myers (Columbia), William P. Fifer (Columbia), Jyoti Angal (University of South Dakota School of Medicine), Theonia K. Boyd (Harvard Medical School), Larry Burd (University of South Dakota School of Medicine), Jacob B. Cotton (Harvard Medical School), Rebecca D. Folkerth (Harvard Medical School), Gary Hankins (University of Texas Medical Branch), Robin L. Haynes (Harvard Medical School), Howard J. Hoffman (National Institute of Deafness and Other Communication Disorders), Perri K. Jacobs (Harvard Medical School), Julie Petersen (DM-STAT Inc and Boston University School of Public Health), Nicolò Pini (Columbia), Bradley B. Randall (University of South Dakota School of Medicine), Drucilla J. Roberts (Massachusetts General Hospital), Fay Robinson (DM-STAT Inc and PPD, Wilmington, North Carolina), Mary A. Sens (University of North Dakota), Peter Van Eerden (University of North Dakota), Colleen Wright (Stellenbosch University), Ingrid A. Holm (Harvard Medical School), and Hannah C. Kinney (Harvard Medical School) 

The research was supported by grants from the National Institutes of Health (U01HD055154, U01HD045935, U01HD055155, U01HD045991, and U01AA016501).  

The Columbia authors reported no conflict of interest disclosures.