Smoking during pregnancy has been associated with a range of adverse birth outcomes,1 with smoking-attributable neonatal expenditures costing the United States an estimated $366 million in 1996.2 Intrauterine growth restriction (IUGR) with correspondingly small-for-gestational-age (SGA) newborns and preterm birth are two of the most well-documented sequelae of smoking during pregnancy.3–6 Small-for-gestational-age and preterm newborns have increased morbidity and mortality,7–9 as well as long-term adverse health and developmental outcomes.10–12 Prior studies of smoking cessation during pregnancy have demonstrated decreases in the incidence of low birth weight, SGA, and preterm newborns if mothers quit early; however, these studies have been limited in scale and number.13–17 Additionally, older expectant mothers who smoke appear to have an increased risk of preterm and SGA newborns beyond that seen with increased age alone,18–20 suggesting that quitting may carry an even greater benefit with increasing maternal age.
For the many pregnant women who continue to smoke, additional information about the benefits of quitting after the onset of pregnancy may provide an incentive to quit. The purpose of this study was to use the newly revised birth certificate to assess the impact of smoking cessation on the risk of delivering preterm and SGA newborns in a large population-based sample of U.S. births.
MATERIALS AND METHODS
This study is a retrospective cohort analysis of U.S. birth certificates that examines the association between delivery of a preterm and/or SGA newborn and maternal smoking status throughout pregnancy. The most recent (2003) revision of the U.S. birth certificate contains additional levels of detail on several maternal and perinatal risk factors, compared with the prior revision in 1989. This new revision has been gradually implemented across the country. We analyzed U.S. birth certificate data from 2005 for the eleven states that used the most recent revision that year: Idaho, Kansas, Kentucky, Nebraska, New Hampshire, New York (not including New York City), Pennsylvania, South Carolina, Tennessee, Texas, and Washington. Because we analyzed publicly accessible, deidentified data, this study was exempt from review by an institutional review board.
The new U.S. birth certificate assesses maternal tobacco use by asking the mother for the average number of cigarettes smoked per day for each trimester. Any number of cigarettes smoked greater than zero designates the mother as a smoker for that respective trimester. In this study, maternal smoking status was defined as “never smoked” (no smoking in any trimester), “quit in the first trimester” (smoked in first trimester but not in second or third), “quit in the second trimester” (smoked in first and second trimesters but not in third), and “smoked throughout” pregnancy (smoked in all three trimesters). Smoking cessation could only be determined if the mother reported smoking in one trimester, but not in the subsequent trimester(s). Smoking cessation in the third trimester could not be assessed because information about maternal smoking at or after birth was not collected. Pregnant women who did not smoke for one trimester but smoked in a subsequent trimester were considered “intermittent smokers,” and births to these mothers were excluded. The main analysis excluded women who delivered before 28 weeks of gestation because births occurring before the third trimester did not have third-trimester smoking data and second-trimester quitting could not be ascertained. To analyze the association between the outcomes and first-trimester smoking cessation among women with deliveries before 28 weeks of gestation, we conducted a sensitivity analysis described below.
We defined four possible combinations of preterm/SGA newborn status: preterm non-SGA newborn, term SGA newborn, preterm SGA newborn, or term non-SGA newborn (referent). For the main analysis, a preterm newborn was defined as any live newborn born between 28 and less than 37 completed weeks of gestation and a term newborn was any live newborn born at or after 37 completed weeks of gestation. Gestational age was determined as the number of weeks from the mother's last menstrual period, unless it was implausible for the newborn's birth weight, based on a combination of two established methods of size-for-gestational-age determination.21,22 For implausible values based on last menstrual period, the clinical estimate of gestational age was used instead. If this was also implausible, the newborn was excluded from the study population. Determination of SGA was based on the usual cut point of weight less than the 10th percentile for the respective gestational age of each newborn.21
Several potential confounders pertaining to the mother were selected based on prior knowledge and adjusted for in our analysis. Maternal age was categorized as less than 20, 20–29, 30–39, and 40 or more years. Race/ethnicity was categorized as non-Hispanic white, non-Hispanic black, Hispanic, and all others. Education was defined as inadequate-for-age, adequate-for-age, and beyond adequate-for-age, using the available categories of educational attainment on the birth certificate. To be considered adequate-for-age, mothers aged less than 15 years should have achieved an education level of “8th grade or less,” those aged 15–18 years should have been at least in “9th through 12th grade,” and those aged more than 18 years should have completed high school. Marital status was defined as married or unmarried. Prenatal care was categorized as beginning in the first trimester or not. Prior preterm birth history incorporated parity, and was categorized as no prior births, prior birth(s) but none of them preterm, or prior birth(s) with at least one preterm. We aimed only to examine the relationship between the exposure and outcomes and therefore did not adjust for intermediary conditions (eg, placental abruption) found in the causal pathway.
The study population (n=915,441) was limited to singleton live births to U.S. residents in the 11 states using the revised birth certificate in 2005. From the initial sample of 984,811, we excluded 3,407 (0.3%) with implausible gestational age. We then excluded 11,885 newborns (1.2%) with unknown (0.7%) or intermittent (0.5%) maternal smoking status, and 50,114 (5.1%) with missing data for either outcome (preterm or SGA newborn) or any other covariate in our analysis. An additional 3,964 newborns (0.4%) born before 28 weeks of gestation were excluded from the main analysis to examine newborns that had maternal smoking data for all three trimesters.
Multinomial logistic regression analyses were performed to calculate adjusted odds ratios (aORs) and 95% confidence intervals (95% CIs) for three adverse birth outcomes (preterm non-SGA newborn, term SGA newborn, and preterm SGA newborn, with term non-SGA newborn as the referent) among pregnant women who never smoked during pregnancy, those who quit smoking in the first trimester, and those who quit smoking in the second trimester, compared with women who smoked throughout pregnancy (referent). Additional aORs and 95% CIs were calculated after stratification of the study population by maternal age. Statistical analyses were performed using SAS softwareV.9 (SAS Institute Inc., Cary, NC) for Windows.
As noted previously, preterm newborns born before 28 weeks of gestation were excluded from the main analysis. To determine whether this exclusion appreciably affected the association between first-trimester maternal smoking cessation and the outcomes, a separate sensitivity analysis was conducted. This analysis was performed in an identical manner to the main analysis but included only those 3,964 (0.4%) preterm newborns at less than 28 weeks of gestation and all term newborns. Newborns delivered between 28 and less than 37 weeks of gestation were excluded from the sensitivity analysis.
Among the study population of births, 8% were preterm non-SGA, 8% were term SGA, and 1% were preterm SGA. During pregnancy, 88% of the expectant mothers reported not smoking, 2% reported quitting during the first trimester, less than 1% reported quitting during the second trimester, and 10% reported smoking throughout (Table 1). After categorizing the births by preterm/SGA status, maternal risk factors with a greater proportion of preterm and SGA newborns included age less than 20 years, non-Hispanic black race/ethnicity, and unmarried status, while education beyond adequate-for-age and onset of prenatal care in the first trimester were risk factors with a smaller proportion of SGA newborns.
The proportion of women delivering preterm or SGA newborns differed by maternal smoking status (Table 2). When compared with those who smoked throughout pregnancy, women who quit smoking in the first trimester had smaller percentages of preterm non-SGA (8% compared with 10%), term SGA (9% compared with 15%), and preterm SGA newborns (1% compared with 2%).
Compared with the referent group of women who smoked throughout pregnancy, women who quit smoking in the first trimester had lower adjusted odds of delivering preterm non-SGA, term SGA, and preterm SGA newborns similar in magnitude to those who never smoked (Table 3). Quitting in the first trimester lowered the odds of delivering a preterm non-SGA newborn by 31%, a term SGA newborn by 55%, and a preterm SGA newborn by 53% compared with smoking throughout pregnancy. Women who quit in the second trimester also had lower adjusted odds of delivering preterm non-SGA and term SGA newborns than smokers but to a lesser magnitude than first-trimester quitters or nonsmokers. Second-trimester quitters did not have significantly lower odds of delivering preterm SGA newborns than smokers.
After stratifying by maternal age, lower adjusted odds of delivering preterm non-SGA, term SGA, and preterm SGA newborns were observed among first-trimester quitters, as well as nonsmokers in all age strata (Table 4), although the reductions were not statistically significant for preterm non-SGA newborns born to nonsmokers aged less than 20 years and preterm SGA newborns born to first-trimester quitters aged 40 and older. In general, this reduction in odds was more pronounced among strata of older mothers, particularly among women aged 40 and older. The trend of lower odds with increasing maternal age was not observed among second-trimester quitters.
When we conducted the sensitivity analysis to examine the impact of first-trimester quitting on the delivery of SGA and preterm newborns at less than 28 weeks of gestation, in contrast to the main analysis, the reduction in the odds of delivering preterm non-SGA newborns was not significant for first-trimester quitters, and no reduction in the odds of delivering preterm SGA newborns was observed (Table 5). The odds of delivering term SGA newborns remained the same.
These findings not only confirm prior research that has shown an increased risk of delivering preterm and SGA newborns among pregnant smokers,3,6,23–25 but demonstrate that those who quit in the first trimester can achieve the same lower risk of adverse birth outcomes as women who never smoked during pregnancy. Although pregnancy is the most common reason women quit smoking,26 smoking is a difficult addiction to break. Added incentive to quit may come with further evidence of the benefits of smoking cessation even as pregnancy progresses. Our results show that first-trimester quitters have a risk of delivering a preterm or SGA newborn comparable to those who never smoked during pregnancy, and second-trimester quitters also have a lower risk of these outcomes, although not to the same magnitude as first-trimester quitters. Additionally, older mothers who quit in the first trimester appear to have a greater reduction in the risk of delivering preterm non-SGA and term SGA newborns than younger mothers. Although our sensitivity analysis did not show a significant reduction in the risk of delivering preterm newborns at less than 28 weeks of gestation (either SGA or non-SGA) for first-trimester quitters, the subgroup of very preterm newborns whose mothers quit smoking was quite small, limiting our ability to observe a difference.
The results we observed are biologically plausible. With regard to SGA newborns, our findings are consistent with prior studies on the effects of smoking and IUGR20,25,27 and support the premise that smoking impedes growth and fetal weight gain, which occur primarily in the third trimester.14,15 Regarding IUGR, smoking appears to have a greater impact on older pregnant women, although the explanation for this is not clear. Regulatory factors involved with placental blood flow and corresponding fetal growth may be more sensitive to smoke exposure in older women, and eliminating this exposure would be expected to improve these measures.18–20 Before implementation of the 2003 birth certificate revision, U.S. birth data showed that older women who smoke during pregnancy tend to smoke more cigarettes per day than younger women.28 In our data, we observed a similar positive correlation between maternal age and number of cigarettes smoked (data not shown). This may partially explain the greater benefits derived from quitting among older mothers. Smoking similarly increases the risk of preterm labor via effects on the placenta and increases the risk of antepartum hemorrhage, abruptio placentae, placenta previa, and preterm premature rupture of membranes.29–31 Although these adverse outcomes occur later in pregnancy, their origins occur early. The earlier a pregnant smoker quits, the more likely the impacts of smoking may be mitigated. Our results that show second-trimester quitting may still be beneficial, although less so than first-trimester quitting, are consistent with this premise.
We recognize that using a 10th percentile cut point for determination of SGA status will include some newborns who are constitutionally small,32 and use of live-birth weight standards instead of ultrasonography-derived fetal growth standards can overestimate the proportion of preterm SGA newborns.33 To address these concerns and focus attention on more severely impacted newborns, we restricted the definition of SGA and repeated the analysis using a third percentile weight-for-gestational-age standard34 as the SGA cut point. Despite these changes, the results were similar to the main analysis (data not shown).
This study was able to measure changes in outcomes associated with smoking cessation in a large population of births. The U.S. birth certificate allows for relatively uniform collection of data on a standardized form. The 2003 revision assesses smoking status by trimester, allowing for improved surveillance of smoking during pregnancy,8 as well as assessment of quit status, which was not previously possible.
This study was subject to a few limitations. First, although we analyzed a large population, it is not nationally representative, given that it included births in only 11 states (approximately 25% of U.S. births). The study population also excluded multiple births, those with implausible gestational age, and those with missing covariate data, as well as births at less than 28 weeks of gestation, resulting in a preterm birth rate that was 3.5% lower than the national rate. However, the births in our study were nearly identical to the entire population of U.S. resident singleton births of plausible gestational age that same year with respect to maternal age, race/ethnicity, marital status, prior preterm births, and the outcomes of preterm and SGA newborns (data not shown). Second, because maternal smoking status is self-reported, misclassification bias may have occurred, either through bias in reporting or recall. Mothers may have been reluctant to report smoking while pregnant or may have been more apt to report quitting when they did not. However, this underreporting is likely to dilute any positive associations, as this type of misclassification would classify some true smokers as nonsmokers.35 Third, recall bias may have affected designation of maternal smoking status among quitters if the mother did not recall which trimester she stopped smoking. Fourth, U.S. birth certificate data are limited in type and quality. Information on several risk factors that may act as potential confounders could not be adjusted for in our study, such as passive smoke exposure and body mass index. Finally, as for any analysis using U.S. birth certificates, the quality of gestational age data may be affected by maternal recall or misidentification of the last menstrual period.8
Among pregnant women who smoke, quitting in the first trimester reduces the risk of delivering both preterm and SGA newborns to the level of nonsmokers. Smoking cessation during the second trimester also confers significantly lower risk of delivering a preterm non-SGA and term SGA newborn, although to a lesser degree than first-trimester quitting. For older mothers, the protective benefit against these adverse birth outcomes observed among first-trimester quitters appears to be even greater than for younger mothers. These findings provide further evidence of the benefits of smoking cessation in a large U.S. subpopulation and serve as an added incentive to quit.
1. Women and smoking: a report of the Surgeon General. Executive summary. MMWR Recomm Rep 2002;51(RR-12):i–iv;1–13.
2. Centers for Disease Control and Prevention (SDS). State estimates of neonatal health-care costs associated with maternal smoking–United States, 1996. MMWR Morb Mortal Wkly Rep 2004;53:915–7.
3. Chiolero A, Bovet P, Paccaud F. Association between maternal smoking and low birth weight in Switzerland: the EDEN study. Swiss Med Wkly 2005;135:525–30.
4. Hammoud AO, Bujold E, Sorokin Y, Schild C, Krapp M, Baumann P. Smoking in pregnancy revisited: findings from a large population-based study. Am J Obstet Gynecol 2005;192:1856–62; discussion 1862–3.
5. Pollack H, Lantz PM, Frohna JG. Maternal smoking and adverse birth outcomes among singletons and twins. Am J Public Health 2000;90:395–400.
6. Raatikainen K, Huurinainen P, Heinonen S. Smoking in early gestation or through pregnancy: a decision crucial to pregnancy outcome. Prev Med 2007;44:59–63.
7. Garite TJ, Clark R, Thorp JA. Intrauterine growth restriction increases morbidity and mortality among premature neonates. Am J Obstet Gynecol 2004;191:481–7.
8. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Menacker F, Kirmeyer S, et al. Births: final data for 2005. Natl Vital Stat Rep 2007;56:1–103.
9. Mathews TJ, MacDorman MF. Infant mortality statistics from the 2004 period linked birth/infant death data set. Natl Vital Stat Rep 2007;55:1–32.
10. Hack M, Klein NK, Taylor HG. Long-term developmental outcomes of low birth weight infants. Future Chile 1995;5:176–96.
11. Pallotto EK, Kilbride HW. Perinatal outcome and later implications of intrauterine growth restriction. Clin Obstet Gynecol 2006;49:257–69.
12. Saigal S, Doyle LW. An overview of mortality and sequelae of preterm birth from infancy to adulthood. Lancet 2008;371:261–9.
13. Jaddoe VW, Troe EJ, Hofman A, Mackenbach JP, Moll HA, Steegers EA, et al. Active and passive maternal smoking during pregnancy and the risks of low birthweight and preterm birth: the Generation R Study. Paediatr Perinat Epidemiol 2008;22:162–71.
14. Lieberman E, Gremy I, Lang JM, Cohen AP. Low birthweight at term and the timing of fetal exposure to maternal smoking. Am J Public Health 1994;84:1127–31.
15. MacArthur C, Knox EG. Smoking in pregnancy: effects of stopping at different stages. Br J Obstet Gynecol 1988;95:551–5.
16. Mainous AG, Hueston WJ. The effect of smoking cessation during pregnancy on preterm delivery and low birthweight. J Fam Pract 1994;38:262–6.
17. Wen CP, Cheng TY, Lin CL, Wu HN, Levy DT, Chen LK, et al. The health benefits of smoking cessation for adult smokers and for pregnant women in Taiwan. Tobacco Control 2005;14(suppl 1):i56–61.
18. Cnattingius S. Maternal age modifies the effect of maternal smoking on intrauterine growth retardation but not on late fetal death and placental abruption. Am J Epidemiol 1997;145:319–23.
19. Salihu HM, Shumpert MN, Aliyu MH, Kirby RS, Alexander GR. Smoking-associated fetal morbidity among older gravidas: a population study. Acta Obstet Gynecol Scand 2005;84:329–34.
20. Wen SW, Goldenberg RL, Cutter GR, Hoffman HJ, Cliver SP, Davis RO, et al. Smoking, maternal age, fetal growth, and gestational age at delivery. Am J Obstet Gynecol 1990;162:53–8.
21. Alexander GR, Himes JH, Kaufman RB, Mor J, Kogan M. A United States national reference for fetal growth. Obstet Gynecol 1996;87:163–8.
22. Kiely JL. What is the population-based risk of preterm birth among twins and other multiples? Clin Obstet Gynecol 1998;41:3–11.
23. Delpisheh A, Kelly Y, Rizwan S, Attia E, Drammond S, Brabin BJ. Population attributable risk for adverse pregnancy outcomes related to smoking in adolescents and adults. Pub Health 2007;121:861–8.
24. Shah NR, Bracken MB. A systematic review and meta-analysis of prospective studies on the association between maternal cigarette smoking and preterm delivery. Am J Obstet Gynecol 2000;182:465–72.
25. Villalbi JR, Salvador J, Cano-Serral G, Rodriguez-Sanz MC, Borrell C. Maternal smoking, social class and outcomes of pregnancy. Paediatr Perinat Epidemiol 2007;21:441–7.
26. Klesges LM, Johnson KC, Ward KD, Barnard M. Smoking cessation in pregnant women. Obstet Gynecol Clin North Am 2001;28:269–82.
27. Horta BL, Victora CG, Menezes AM, Halpern R, Barros FC. Low birthweight, preterm births and intrauterine growth retardation in relation to maternal smoking. Paediatr Perinat Epidemiol 1997;11:140–51.
28. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Menacker F, Munson ML. Births: final data for 2002. Natl Vital Stat Rep 2003;52:1–113.
29. Castles A, Adams EK, Melvin CL, Kelsch C, Boulton ML. Effects of smoking during pregnancy. Five meta-analyses. Am J Prevent Med 1999;16:208–15.
30. Fredricsson B, Gilljam H. Smoking and reproduction. Short and long term effects and benefits of smoking cessation. Acta Obstet Gynecol Scand 1992;71:580–92.
31. Seltzer V. Smoking and women's health. Int J Gynaecol Obstet 2000;70:159–63.
32. Reeves S, Bernstein IM. Optimal growth modeling. Semin Perinatol 2008;32:148–53.
33. Ott WJ. Intrauterine growth retardation and preterm delivery. Am J Obstet Gynecol 1993;168:1710–5; discussion 1715–7.
34. Oken E, Kleinman KP, Rich-Edwards J, Gillman MW. A nearly continuous measure of birth weight for gestational age using a United States national reference. BMC Pediatr 2003;3:6.
35. The 2004 United States Surgeon General's Report: The Health Consequences of Smoking. N S W Public Health Bull 2004;15:107.