Association of Maternal Cigarette Smoking and Smoking Cessation With Preterm Birth

Abstract

Importance: Cigarette smoking during pregnancy increases the risk of preterm birth, low birth weight, and infant mortality.

Objective: To assess the probability of preterm birth among expectant mothers who smoked cigarettes before pregnancy and quit smoking at the start or during pregnancy.

Design, setting, and participants: This cross-sectional study analyzed information provided on live birth certificates from 2011 through 2017 that were obtained from US states that implemented the 2003 revision of the US live birth certificate. In total, 25 233 503 expectant mothers who delivered live neonates and had known prepregnancy and trimester-specific cigarette smoking frequency were included in the analyses.

Exposures: Cigarette smoking frequency (1-9, 10-19, and ≥20 cigarettes per day) 3 months prior to pregnancy and for each trimester during pregnancy.

Main outcomes and measures: Cigarette smoking cessation throughout pregnancy, after the first trimester, after the second trimester, and during the third trimester irrespective of first and second trimester smoking. Probability of preterm birth (<37 weeks' gestation).

Results: Of 25 233 503 expectant mothers who delivered live neonates between 2011 and 2017, the modal age at delivery was 25 to 29 years; 52.9% were non-Hispanic white, 23.6% were Hispanic, and 14.2% were non-Hispanic black women; 22 600 196 mothers did not smoke during the 3 months prior to pregnancy, and 2 633 307 smoked during the 3 months prior to pregnancy. The proportion of prepregnancy smokers who quit throughout pregnancy was 24.3% in 2011 and 24.6% in 2017. The proportion of prepregnancy smokers who quit during the third trimester was 39.5% in 2011 and 39.7% in 2017. High-frequency cigarette smoking often occurred among expectant mothers who smoked during pregnancy (eg, 46.9% of third-trimester smokers smoked ≥10 cigarettes per day in 2017). The probability of preterm birth decreased more the earlier smoking cessation occurred in pregnancy. For example, the probability of preterm birth was 9.8% (95% CI, 9.7%-10.0%) among 25- to 29-year-old, non-Hispanic white, primigravida and primiparous expectant mothers (ie, pregnant for the first time and not yet delivered) who smoked 1 to 9 cigarettes per day prior to pregnancy and maintained this frequency throughout their pregnancy. The probability of preterm birth was 9.0% (95% CI, 8.8%-9.1%) if smoking cessation occurred at the start of the second trimester (an 8.9% relative decrease), and 7.8% (95% CI, 7.7%-8.0%) if cessation occurred at the start of pregnancy (a 20.3% relative decrease).

Conclusions and relevance: Quitting smoking-and quitting early in pregnancy-was associated with reduced risk of preterm birth even for high-frequency cigarette smokers.

Figure 1.. Proportion of Expectant Mothers from 2011 through 2017 Who Smoked 3 Months Prior to Pregnancy but Quit Smoking Throughout Pregnancy, During the Second and Third Trimesters, During the Third Trimester, or in the Third Trimester

Analysis of 2011 through 2017 US live birth certificate data.^,,,,,,

Figure 2.. Frequency of Cigarette Smoking by Trimester Among Expectant Mothers Who Smoked During Pregnancy

Analysis of 2011 through 2017 US live birth certificate data.^,,,,,,

Figure 3.. Probability of Preterm Birth (<37 Weeks’ Gestation) by Cigarette Smoking Status 3 Months Prior to Pregnancy or During Trimester 1, 2, or 3

Analysis of 2011 through 2017 US live birth certificate data.^,,,,,,

Figure 4.. Estimated Probability of Preterm Birth (<37 Weeks’ Gestation) by Cigarette Smoking Frequency Among Expectant Mothers Who Smoked 3 Months Prior to Pregnancy but Quit Smoking at Various Points During Pregnancy and Among Expectant Mothers Who Did Not Smoke 3 Months Prior to or During Pregnancy

Analysis of 2011 through 2017 US live birth certificate data.^,,,,,, Expectant primigravida and primiparous women were aged 25 to 29 years, gave birth in 2017, attended at least some college, were married, did not receive Special Supplemental Nutrition Program for Women, Infants, and Children benefits, and were privately insured. The width of the 95% confidence intervals are smaller than the diameters of the circles denoting the point estimates and thus do not show in the figure.