American Journal of Epidemiology
Copyright C 1997 by The Johns Hopkins University School of Hygiene and Public Health
AH rights reserved
Vol 146, No 1
Printed In U SA.
Exposure to Environmental Tobacco Smoke and Birth Outcome: Increased
Effects on Pregnant Women Aged 30 Years or Older
Indu B. Ahluwalia,1 Laurence Grummer-Strawn,2 and Kelley S. Scanlon3
The purposes of this study were to examine the association between setf-reported environmental tobacco
smoke (ETS) exposure during pregnancy and birth weight, prematurity, and small-for-gestational age infants
and to determine whether these associations differ by maternal age. Data from the Pregnancy Nutrition
Surveillance System from two states that collected data on both passive and active smoking for the penod
1989-1994 were analyzed. ETS exposure was defined as reported exposure to the cigarette smoke of a
household member. Multiple logistic and linear regression analyses were used to evaluate the association
between ETS and birth outcomes. The mean adjusted birth weight among infants of nonsmoking mothers age
30 years or older was 90 g less among infants exposed to ETS than among infants not exposed. No significant
association was found among infants of younger nonsmoking mothers. Similarly, the nsks for low birth weight
(adjusted odds ratio (OR) = 2.42, 95% confidence interval 1.51-3.87) and preterm delivery (adjusted OR =
1.88, 95% confidence interval 1.22-2 88) were elevated among older nonsmokers exposed to ETS, but not
among younger nonsmokers exposed to ETS (adjusted OR = 0.97, 95% confidence interval 0.76-1.23;
adjusted OR = 0.92, 95% confidence interval 0.76-1.13, for low birth weight and preterm delivery, respectively). These findings indicate that the association between ETS exposure and adverse pregnancy outcomes
appears to be modified by maternal age. Am J Epidemiol 1997; 146:42-7.
birth weight; environmental pollution, tobacco smoke; infant, premature; infant, small-for-gestational
age; maternal age
Studies have shown that smoking during pregnancy
elevates the risk of delivering a low birth weight
(LBW), preterm, or small-for-gestational age (SGA)
infant and that this risk increases with increasing maternal age (1-5). The literature on the association
between environmental tobacco smoke (ETS) and
pregnancy outcome has shown that exposure to ETS
also elevates the risk of adverse birth outcomes. Martin and Bracken (6) demonstrated that ETS exposure
for at least 2 hours per day resulted in a mean birth
weight reduction of 85 g and a twofold increased risk
of LBW among the infants of nonsmokers. Since then,
several other studies have demonstrated that infants
born to women exposed to various levels of ETS have
lower birth weights than do infants not exposed to ETS
(7-14); a few studies have demonstrated no adverse
effect on birth weight among infants exposed to ETS
(15, 16). The differences in birth weight ranged from
16 g to 120 g. Studies have also shown that women
exposed to ETS were slightly more likely to deliver an
SGA infant or to experience preterm delivery (10, 11,
17). One study found no association between ETS
exposure and SGA (18). The preterm component of
LBW has not shown a consistent association with
ETS. One study found a mild association between
ETS exposure in the work environment and preterm
birth (10); several others found no association (6,
12, 17).
Although the relations in several of these studies
were not statistically significant, the results nonetheless report an increased risk of adverse birth outcomes
among those exposed to ETS (10, 11, 17). From the
existing literature, it is not clear which component of
LBW is most impacted by maternal exposure. It is
clear, however, that overall birth weight is negatively
affected by ETS exposure during pregnancy. To our
knowledge, no study has examined the effect of maternal age on the association between ETS exposure
Received for publication September 26, 1996, and in final form
Apnl 10, 1997.
Abbreviations. ETS, environmental tobacco smoke; LBW, low
birth weight; SGA, small-for-gestational age.
1
Epidemic Intelligence Service, Division of Nutrition and Physical
Activity, National Centers for Chronic Disease Prevention and
Health Promotion, and Epidemiology Program Office, Centers for
Disease Control and Prevention, Atlanta, GA
2
Division of Nutrition and Physical Activity, National Centers for
Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, Atlanta, GA.
Repnnt requests to Dr. Indu B. Ahluwalia, Division of Nutrition
and Physical Activity, Maternal and Child Health Branch, Centers for
Disease Control and Prevention, 4770 Buford Highway NE, Mailstop
K-25, Atlanta, GA 30341-3724.
42
Environmental Tobacco Smoke and Birth Outcome
and the occurrence of LBW, preterm delivery, or
SGA.
In this study, we have assessed the relation between
maternal exposure to ETS and the occurrence of LBW,
preterm delivery, and SGA and have determined the
effect of advancing maternal age on the association
between exposure to ETS and pregnancy outcomes.
MATERIALS AND METHODS
Study subjects
The subjects for the study were women with singleton pregnancies for whom ETS exposure and birth
outcome data were reported to the Centers for Disease
Control and Prevention Pregnancy Nutrition Surveillance System from the two US states (Arizona and
North Dakota) that collected ETS data from 1989 to
1994. Subjects were low-income women who received
services from publicly funded maternal and child
health clinics. Our study included 17,412 subjects who
reported ETS exposure and a birth outcome.
Definition of variables
Exposure to ETS was defined as self-reported exposure to a household member's cigarette smoke during pregnancy. Dichotomous (yes/no) response to the
question, "Does anyone other than you (the pregnant
woman) currently smoke cigarettes in the household?"
was used to obtain information about ETS exposure. A
smoker was anyone who reported during the first
clinic visit that he or she smoked any number of
cigarettes. The outcome variables considered in this
analysis were mean birth weight, LBW, preterm delivery, and SGA. LBW was defined as an infant
weight of less than 2,500 g at birth, and preterm
delivery was defined as a gestational age of less than
37 weeks (19). The gestational age used to define
preterm birth was calculated from our estimate of the
expected date of delivery, which is based on the date
of last menstrual period. SGA was defined as a birth
weight of greater than two standard deviations below
the reference median birth weight for the infant's
gestational age on the basis of gender, race, altitude,
and gestational age-specific fetal growth reference
(Yip R, McLaren N, unpublished manuscript, 1992)
(20).
Statistical methods
We performed multiple linear regression analyses to
assess the association between ETS exposure and
mean birth weight after controlling for potential confounders. We used multiple logistic regression to calculate adjusted odds ratios and related 95 percent
Am J Epidemiol
Vol. 146, No. 1, 1997
43
confidence intervals for LBW, preterm delivery, and
SGA and performed statistical analyses using the Statistical Analysis System (Cary, North Carolina). Factors considered in the regression models included
mother's ethnicity (i.e., white, black, Hispanic, Native
American, and Asian), education (i.e, ^ 8 , 9-<12, 12,
13-<15, and >15), parity (i.e., 0, 1, 2, and >3),
marital status (i.e., married or not married), weight
gain during pregnancy (i.e., pounds), prepregnancy
body mass index (i.e., kg/m2), state of residence, use
of alcohol (i.e., yes or no), and use of tobacco (i.e., yes
or no). The altitude of the clinic was also taken into
account. Because maternal age showed an obvious
effect in the descriptive analyses on the relation between ETS and birth weight, women were divided into
two age groups: those younger than age 30 years and
those aged 30 years or older. We constructed separate
models for the two groups.
RESULTS
Overall, of the 17,412 study subjects, 31 percent
reported exposure to ETS, and 22 percent were smokers. Forty-seven percent of those exposed to ETS and
11 percent of those not exposed to ETS were smokers
(table 1). Compared with women not exposed to ETS,
those exposed were more likely to be younger, to be
white, to have completed fewer years of schooling, to
be primigravida, and to report alcohol use.
Figure 1 illustrates mean birth weight by maternal
age among ETS-exposed and nonexposed women. The
data are further stratified by maternal smoking status.
Among nonsmokers less than age 30 years, there is
little difference in infant birth weight between ETSexposed and nonexposed women. However, for nonsmoking women over age 30 years, birth weight was
lower in infants of those exposed to ETS. This birth
weight difference between exposed and nonexposed
women appears to increase with increasing maternal
age. Among smokers, those exposed to a household
member's cigarette smoke show consistently lower
infant birth weights compared with smokers not exposed to additional cigarette smoke in the home. This
difference among women over age 30 years also increases with increasing maternal age. The upward
surge in birth weight for smokers age 30 years or older
who were not exposed to additional ETS and a downward surge among smokers exposed to ETS reflect the
small number of births in these age categories.
After observing that the relation between reported
ETS exposure and infant birth weight was modified by
maternal age, we stratified the results by age less than
and equal to or greater than 30 years. Stratification
revealed that among older nonsmokers, the mean adjusted birth weight of infants was 90 g lower for those
44
Ahluwalia et al.
TABLE 1. Sododemographlc characteristics of the study
population by exposure to ETS*, PNSS*, 1989-1994
Characteristic
ETS exposure
(n«. 5,446)
Non-ETS exposure
(n = 11,866)
No.
%
No
%
2,563
2,860
47.3
52.7
1,254
10,637
10.5
89.5
1,557
1,816
1,199
581
196
27
29.0
33.8
22.3
10.8
3.6
0.5
2,661
4,120
2,953
1,445
540
106
22.5
34 8
25.0
1£2
46
0.9
Ethnicity
African American
Asian
Hispanic
Native American
White
121
38
1,042
589
3,656
2.2
0.7
19.1
10.8
67.1
499
145
4,773
1,047
5,500
4.1
1.2
39.9
8.8
46.0
Education
£8
9-11
12
>12-14
£15
585
1,785
2,045
744
220
10 9
33.2
38.0
13.8
4.1
1,764
2,966
4,256
1,896
753
15.2
25.5
36.6
16.3
6.5
Parity
0
1
2
£3
2,083
1,463
939
745
39.8
28 0
18.0
14.2
3,802
3,268
2,233
2,042
33 5
28.8
19.7
18 0
Marital status
Married
Not married
2,576
2,752
48.3
51.7
5,981
5,671
51.3
48.7
956
4,460
17.7
82.3
1,549
10,372
13.0
87.0
Smoker
Yes
No
DISCUSSION
Maternal age (years)
15-19
20-24
25-29
30-34
35-39
£40
Alcohol use
Vfea
No
ETS was observed, but because the confidence intervals are wide, we cannot rule out chance. Exposed
smokers showed little or no increased risk of delivering SGA infants (table 3), although the estimated
relative risk of SGA was slightly elevated among
younger smokers. However, again, the confidence intervals are wide.
* ETS, environmental tobacco smoke; PNSS, Pregnancy
Nutrition Surveillance System.
exposed to ETS than for those not exposed (table 2).
Among smokers, birth weight was lower in infants of
both younger and older women exposed to ETS.
Among nonsmokers, we observed an increased estimated risk of LBW and preterm deliveries among
older women but not among younger women (table 3).
The estimated relative risk of SGA was elevated
among older ETS-exposed women, but the confidence
intervals are wide, and therefore we cannot rule out the
possibility that the observed effect is entirely due to
chance. Among smokers, the estimated risk of LBW
was elevated for both younger and older pregnant
women exposed to ETS (table 3). A small increased
risk of prematurity among older smokers exposed to
In this study, we demonstrated an association between ETS exposure and the occurrence of LBW and
preterm delivery among older nonsmoking women.
We also observed an increased risk of SGA among
older women exposed to ETS, but because the confidence intervals are wide, the observed effect could be
due to chance. Our inconclusive findings regarding
SGA are consistent with those of other studies (17,
18). Our findings on the association between ETS
exposure and preterm births are more in accord with
findings of the study by Ahlborg and Bodin (10),
which found a mild association between ETS exposure
in the workplace and preterm delivery. However, we
observed the increased preterm risk only among older
exposed women. Among smokers, we observed an
increased risk of LBW among both younger and older
women exposed to ETS; however, the confidence intervals are wide. The estimated risk for LBW was
higher among older smokers exposed to ETS. Exposed
smokers showed little or no increased risk of delivering a preterm or an SGA infant.
These findings should be considered in light of the
following limitations. We cannot exclude the possibility that factors other than ETS may explain the findings. For example, we were not able to control for
economic and social factors (e.g., income and household size) that may be associated with both pregnancy
outcome and exposure to ETS. However, we adjusted
for education level, ethnicity, and marital status, each
of which are important indicators of socioeconomic
status. Furthermore, our study population included
only low-income women. An additional limitation is
that we were not able to assess the extent of underreporting of cigarette smoking or ETS exposure and did
not have cotinine levels on pregnant women, which
would have helped to assess exposure and provide
insight about the extent of misclassification that may
have occurred. Cotinine is the main metabolite of
nicotine, and blood serum levels of cotinine are often
used as an indication of an individual's exposure to
tobacco smoke. We also did not know the length of
exposure or history of tobacco use among our study
subjects, although, in terms of having a history of
tobacco use, findings from several studies of birth
weight among smokers who quit before conception or
Am J Epidemiol
Vol. 146, No. 1, 1997
Environmental Tobacco Smoke and Birth Outcome
3600
45
ETS nonexposed
Nonsmokers
^ 3500
CO
I 3400
£ 3300
| 3200
t
m
3100
c
I 3000
2900
15-19
2800
20-24
25-29
30-34
35-39
40+
Maternal Age
FIGURE 1. Adjusted mean birth weight (g) of infants by maternal age, smoking status, and environmental tobacco smoke (ETS) exposure,
Pregnancy Nutrition Surveillance System, 1989-1994. Birth weight adjusted for ethnicity, education, marital status, parity, state, alcohol use,
weight gain, prepregnancy body mass index, and altitude.
TABLE 2. Adjusted* mean birth weight (In grains) of Infants,
by mother's age and exposure category, PNSSf, 1989-1994
Nonsmokers
<30
£30
ETSf
exposure
(weight)
Non-ETS
exposure
(weight)
Mean
dfference
3,395.9
3,407.8
3,387.1
3,497.8
-8.8
90.0
-43 7 to 26.1
-0.8to180.9
3,168.1
3,114.8
3,226 0
3,187.3
57.9
72.5
003to115.8
-68.3 to 213.3
95%Clt
Smokers
<30
£30
* Adjusted for ethnicity, education, marital status, parity, state,
alcohol use, weight gain, prepregnancy body mass index, and
altitude.
t PNSS, Pregnancy Nutrition Surveillance System; ETS,
environmental tobacco smoke; Cl, confidence interval.
early in their pregnancies have shown that smokers
deliver infants with birth weights similar to those of
infants of nonsmokers (21-23). It could be argued that
the cumulative effect of exposure could be age dependent, with older women being affected differently than
younger women. None of the studies have examined
these associations by maternal age. Another explanation for the findings could be that women classified as
exposed to ETS may have been former smokers and,
because they live in households with smokers, may be
more likely to relapse into smoking after the initial
data collection. We examined this issue, and our data
show that only 2.8 percent of the women relapsed: 2
percent among those who were not exposed to ETS
and 3 percent among those who were exposed. Our
study findings are consistent when we exclude the
small percent of participants who relapsed. We had no
information about exposure to ETS in environments
other than the home, which could affect the intensity
of exposure (24). Finally, because data were available
Am J Epidemiol
Vol. 146, No. 1, 1997
for only two states, our findings are not generalizable
to all low-income pregnant women.
Despite these limitations, the study has several
strengths. The sample size allowed us to examine the
effect of maternal age on the association between
reported ETS exposure and birth outcomes. In addition, we were able to examine the association between
ETS exposure and outcomes associated with birth
weight— LBW, SGA, and preterm delivery.
ETS may affect birth outcomes through several
mechanisms, especially among older women. There is
evidence that smoking restricts the utero placental
blood flow, and the placentas of smokers show vascular lesions (25, 26). Fetal growth is known to be
retarded by the direct toxic effects of nicotine, carbon
monoxide, and other substances generated by burning
cigarettes (27). Literature on smoking shows that older
pregnant women who smoke during pregnancy are at
increased risk for adverse birth outcomes, such as
delivering premature, LBW, and SGA infants (1-5).
There has been discussion of decreased placental sufficiency among older women (28, 29). The decreased
placental sufficiency among older pregnant women
may combine synergistically with ETS exposure to
affect the birth weight and length of gestation, similar
to what is hypothesized to occur among smokers.
Thus, it could be that placental capacity to respond to
the components of cigarette smoke is compromised
among older women compared with younger women
(5). Studies show an association between maternal
smoking and both SGA and preterm births, with the
association with SGA being stronger. We observed a
stronger association between ETS exposure and preterm delivery among older women. It could be that
among older women there are other risks for adverse
46
Ahluwalia et al.
TABLE 3. Associations between ETS* exposure and low birth weight, pro term delivery, and small-forgestational age deliveries, PNSS*, 1989-1994
Maternal age (years)
Nonsmoters
Smotere
<30
LBW*
P re term births
SGA*
<30
£30
OR*,t
95% Cl*
ORt
95% Cl
ORf
95% Cl
ORt
95% Cl
0.97
0.92
0.97
0.76-1.23
0.76-1.13
0.75-1.26
2.42
1.88
1.28
1.51-3.87
1.22-2.88
0.76-2.15
1.39
1.00
1.20
1.01-1.93
0.73-1.37
0.83-1.75
1.69
1.42
0.93
0.95-3.02
0.75-2.70
0.51-1.69
• ETS, environmental tobacco smoke; PNSS, Pregnancy Nutrition Surveillance System; OR, odds ratio; Cl,
confidence interval; LBW, low birth weight; SGA, small-for-gestabonal age.
t Odds ratios adjusted for ethnicity, education, marital status, parity, state, alcohol use, weight gain,
prepregnancy body mass index, and altitude.
pregnancy outcomes that may combine with ETS exposure synergistically (30).
The findings of this study are particularly important
because many women above age 30 years are having
children, and the proportion of births occurring in this
age group is increasing. In the United States, approximately 30 percent of births occur among women aged
30 years or older (31). A recent study showed that 37
percent of US adults who do not use tobacco live in a
household with at least one smoker or are exposed to
ETS at work (32). Given the proportions of older
women giving birth in the United States and adult
exposure to ETS, it is possible that upwards of
300,000 pregnancies among nonsmokers could be affected by ETS exposure, which has implications for
the family and for the child's long-term growth and
development. Lastly, LBW and preterm infants incur
higher health care costs.
Our study adds evidence to the literature demonstrating the adverse effect of ETS on pregnancy outcomes and suggests that behavioral interventions to
stop smoking as well as to avoid ETS exposure should
target women. The pregnant women in this study were
seeking services from publicly funded maternal and
child health clinics. Such clinics may be useful for
delivering behavioral interventions to low-income
women who are exposed to ETS in their homes (33).
From a public health point of view, policies are needed
to limit pregnant women's exposures to ETS.
ACKNOWLEDGMENTS
The authors thank the two states that collect data for the
Centers for Disease Control and Prevention-based surveillance system upon which this research was conducted: the
Arizona Department of Health Services, Community and
Family Health Services, Office of Nutrition Services and the
North Dakota Department of Health, Division of Maternal
and Child Health, Preventive Health Section, Women, Infants, and Children Program.
REFERENCES
1. Cnattingius S, Axelsson O, Eklund G, et al. Smoking, maternal age, and fetal growth. Obstet Gynecol 1985;66:449-52.
2. Cnattingius S. Does age potentiate the smoking-related nsk of
fetal growth retardation? Early Hum Dev 1989;20203-ll.
3. Cnattingius S, Forman M, Berendes HW, et al. Effect of age,
parity, and smoking on pregnancy outcome: a populationbased study. Am J Obstet Gynecol 1993;168 16-21.
4. Wen SW, Goldenberg RL, Cutter GR, et al. Smoking, maternal age, fetal growth, and gestational age at delivery. Am J
Obstet Gynecol 1990;162:53-8
5 Fox SH, Koepsell TD, Dalmg JR Birth weight and smoking
during pregnancy-effect modification by maternal age. Am J
Epidemiol 1994;139:1008-15.
6. Martin TR, Bracken MB Association of low birth weight with
passive smoke exposure in pregnancy Am J Epidemiol 1986;
124:633-42
7. Rubin DH, Krasilnikoff PA, Leventhal JM, et al Effects of
passive smoking on birth-weight. Lancet 1986,2.415-17.
8 Campbell MJ, Lewry J, Wailoo M. Further evidence for the
effect of passive smoking on neonates. Postgrad Med J 1988;
64.663-5.
9 Lazzaroni F, Bonassi S, Manruello E, et al. Effects of passive
smoking during pregnancy on selected perinatal parameters.
Int J Epidemiol 1990;19:960-6.
10 Ahlborg G Jr, Bodin L. Tobacco smoke exposure and pregnancy outcome among working women. A prospective study
at prenatal care centers in Orebro County, Sweden. Am J
Epidemiol 1991;133:338-47.
11. Zhang J, Ratcliffe JM. Paternal smoking and birth weight in
Shanghai. Am J Public Health 1993;83:207-10.
12 Martinez FD, Wright AL, Taussig LM, and the Group Health
Medical Associates. The effects of paternal smoking on the
birth weight of newborns whose mothers did not smoke Am J
Public Health 1994,84:1489-91.
13. Rebaghato M, Bolumar F, Florey C du V Assessment of
exposure to environmental tobacco smoke in nonsmoking
pregnant women in different environments of daily living.
Am J Epidemiol 1995;142:525-30.
14. Rebagliato M, Florey C du V, Bolumar F. Exposure to environmental tobacco smoke in nonsmoking pregnant women in
relation to birth weight. Am J Epidemiol 1995;142:531-7.
15. MacArthur C, Knox EG. Passive smoking and birth weight.
Lancet 1987;l:37-8.
16 Chen Y, Pederson LL, Lefcoe NM. Passive smoking and low
birth weight Lancet 1989,2.54-5.
17. Fortier I, Marcoux S, Brisson J. Passive smoking during
pregnancy and the risk of delivering a small-for-gestationalage infant Am J Epidemiol 1994;139.294-301.
18 Chen LH, Peutti DB. Case-control study of passive smoking
and the risk of small-for-gestational-age at term Am J Epidemiol 1995;142:158-65.
Am J Epidemiol Vol. 146, No. 1, 1997
Environmental Tobacco Smoke and Birth Outcome
19. Dunn PM. The search for perinatal definitions and standards.
Acta Paediatr Scand Suppl 1985;319:7-16.
20. Adams MM, Delaney KM, Stupp P. The relationship of interpregnancy interval to infant birth weight and length of
gestation among low-risk women, Georgia. Paediatr Perinat
Epidemiol 1997;Suppl 11:48-62.
21. MacArthur C, Knox EG. Smoking in pregnancy: effects of
stopping at different stages Br J Obstet Gynaecol 1988;95:
551-5.
22. Mamous AG, Hueston W. The effect of smoking cessation
during pregnancy on preterm delivery and low birth weight.
J Family Prac 1994,38:262-6.
23. Lieberman E, Gremy I, Lang JM, et al. Low birth weight at
term and the timing of fetal exposure to maternal smoking.
Am J Public Health 1994,84:1127-31.
24. Samet JM. Editorial Commentary New effects of active and
passive smoking on reproduction? Am J Epidemiol 1991;133.
348-50.
25. Naeye RL. Effects of maternal cigarette smoking on the fetus
and placenta. Br J Obstet Gynaecol 1978;85:732-7
26. Lehtovuta P, Fross M. The acute effect of smoking on intervillous blood flow of the placenta. Br J Obstet Gynaecol
1978;85:729-31.
Am J Epidemiol
Vol. 146, No. 1, 1997
47
27. Bosley ARJ, Sibert JR, Newcombe RG. Effects of maternal
smoking on fetal growth and nutrition. Arch Dis Child 1981;
56:727-9.
28. Bender HG, Wemer C, Greis HP. Asynchronous maturation of
the placenta in dependence on the parity and the age of the
parturient. (Abstract translated from German). Z Geburtshilfe
Pennatol 1977; 181:421-31.
29. DeSesso JM. Maternal factors in developmental toxicity. Teratog Carcinog Mutagen 1987;7:225-40.
30. Cunningham GF, Leveno KJ Childbearing among older
women—the message is cautiously optimistic. N Engl J Med
1995;333:1002-3.
31 Ventura SJ, Martin JA, Taffel SM, et al. Advance report of
final natality statistics, 1993. Hyattsville, MD: National Center for Health Statistics, 1995. Monthly Vital Statistics Report
September 21, 1995;44(3S).
32. Pirkle JL, Flegal, KM, Bernert JT, et al Exposure of the US
population to environmental tobacco smoke. J Am Med Assoc
1996;275:1233-40.
33. Kendrick JS, Zahniser SC, Miller N, et al Integrating smoking
cessation into routine public prenatal care, the smoking cessation in pregnancy project. Am J Public Health 1995;85'
217-22