A 2026 analysis of nearly 30,000 mothers in the United States found that babies born to mothers who reported a health care provider’s diagnosis of anxiety weighed slightly less at birth on average. The difference was substantially larger among babies whose mothers reported tobacco use during pregnancy.
The findings describe associations rather than proving that anxiety or tobacco exposure caused a particular baby’s birth weight. The analysis also could not determine whether treating anxiety changes birth outcomes.
What the analysis found
In the fully adjusted model, which accounted for maternal age, education, race and ethnicity, marital status, household income and e-cigarette use, babies of mothers with a reported anxiety diagnosis weighed an average of 81.6 grams, or 2.9 ounces, less than babies of mothers without that report.
The corresponding difference linked with tobacco use during pregnancy was 213.1 grams, or 7.5 ounces. The associations remained when anxiety and tobacco use were included in the same model.
In the data overall, babies of mothers who used tobacco weighed about 2,886 grams, or 6.4 pounds, on average, compared with about 3,104 grams, or 6.8 pounds, among babies of mothers who did not report tobacco use.
The analysis described the anxiety-related shift as small for an individual baby. Across a population, however, even a modest average difference could move more babies toward the threshold for low birth weight, defined as less than 2,500 grams, or about 5.5 pounds.
How the study was conducted
The analysis used 2021–2023 data from the Pregnancy Risk Assessment Monitoring System, a CDC-run survey of women who recently gave birth that is linked with each baby’s birth certificate.
Researchers began with 59,343 records. The final analysis included 29,288 mothers with complete answers from 11 places: Delaware, Kansas, Michigan, Missouri, Montana, New Jersey, Pennsylvania, Utah, Washington, Wisconsin and New York City.
Anxiety was measured with a yes-or-no question asking whether a health care provider had told the mother she had anxiety before or during pregnancy. Twenty-nine percent answered yes. Tobacco use was also measured with a yes-or-no question, and 6.4% reported using tobacco during pregnancy.
Birth weights came from birth certificates and were recorded in 250-gram bands. The researchers used the midpoint of each band, a method that adds some imprecision to estimates of differences as small as 2 to 3 ounces.
A multilevel statistical model separated differences between individual mothers from differences between states and allowed the anxiety association to vary among states. About 6.3% of the variation in birth weight was between states rather than between mothers.
Why anxiety and tobacco use need to be considered together
The analysis does not treat anxiety and tobacco use as unrelated factors. The researchers noted that some pregnant women may smoke as a way of coping with stress or anxiety. This means anxiety could be associated with birth weight both directly and indirectly through tobacco use.
Possible explanations discussed in the analysis included tobacco or other coping behaviors, changes in the body’s stress response and differences in prenatal-care use. The study did not test these pathways directly.
Earlier evidence points in a similar direction. A 2018 meta-analysis of 29 prospectively collected studies associated anxiety during pregnancy with an average birth-weight reduction of about 56 grams, along with higher odds of low birth weight and preterm birth. A 2022 meta-analysis of 55 cohort studies associated smoking during pregnancy with 89% higher odds of low birth weight.
What the results cannot show
The study was observational, so it cannot establish cause and effect. Factors such as prenatal care, other health conditions, medications, the timing of delivery or social support could explain part of the observed differences.
- The anxiety measure did not record severity, timing or treatment.
- Women with undiagnosed anxiety were counted as having no anxiety.
- The tobacco measure did not record how much someone smoked, whether they quit or whether they experienced secondhand smoke exposure.
- Self-reported tobacco use may have been underreported.
- Birth timing was not modeled, so some of the weight difference could reflect earlier delivery.
- About half of the initial records were excluded because of missing data.
- The sample did not include southern states and was not nationally representative.
The strength of the anxiety association was highest in Pennsylvania, Kansas and Utah and lowest in Washington, Wisconsin and New Jersey. Because only 11 places were included, the researchers presented those differences as descriptive rather than as a state ranking that can be generalized more broadly. Overall state rates of anxiety and tobacco use did not predict birth weight.
Together, the findings support examining mental health and tobacco exposure as related parts of pregnancy research, while leaving important questions about mechanisms, treatment and individual risk unanswered.
AI tools were used to assist with the preparation of this article.
Leave a Reply