International Journal of Clinical Medicine, 2011, 2, 318-324
doi:10.4236/ijcm.2011.23055 Published Online July 2011 (http://www.SciRP.org/journal/ijcm)
Maternal Smoking during Pregnancy and Sudden
Infant Death Using the National Maternal and
Infant Health Survey: A Case-Case Study
Kathleen Franciska Brookfield1, James Dean Wilkinson1, Barbara Luke2, Kristopher Arheart1,
Eleni Sfakianaki1
1Miller School of Medicine, University of Miami, Miami, USA; 2Michigan State University, East Lansing, USA.
Received January 12th, 2011; revised March 1st, 2011; accepted April 1st, 2011.
ABSTRACT
We utilized data from the National Maternal and Infant Health Survey (NMHIS) to analyze the risk of SIDS and other
infant deaths among women who smoke during pregnancy adjusting for potentially modifiable risk factors such as sec-
ondhand smoke exposure and breastfeeding. The following variables were assessed with respect to risk for SIDS and
other infant deaths: smoking exposure, level of education, infant and maternal age, infant and maternal birthweight,
maternal BMI, gender, secondhand smoke exposure, breast feeding, prenatal vitamins, WIC, multiple gestation, sleep
apnea monitor prescription, sleep apnea incidents and maternal alcohol use. Univariate analysis and multivariate lo-
gistic regression were performed to identify variables significantly associated with the odds of mortality from SIDS.
Analysis utilized weighted estimates using SUDAAN 9.0.0 to adjust for design effects. A p-value < 0.01 was cons idered
statistically significant. Women who smoked during pregnancy were 1.83 times more likely to give birth to an infant
that died from SIDS versus some other cause of death, OR (95%) = 1.83(1.33, 2.51). Other Race infants and Black in-
fants were more likely to suffer SIDS mortality than White infants, but the result was not significant in the fina l model.
Other modifiable risk factors, such as secondhand smoke exposure and breast feeding, were not significant predictors
of SIDS mortality. Independent o f sociod emo graph ic va riables and o ther poten tial risk fa cto rs for SIDS death, ma tern al
smoking was associated with an increased risk of SIDS death versus other death. This study highlights the importance
of screening all pregnant women for tobacco use and emphasizes the importance of smoking cessation to decrease the
risk of infant death from SIDS.
Keywords: Sudden Infant Death, Smoking, Pregnancy, Passive Smoking, Breastfeeding
1. Introduction
Sudden Infant Death Syndrome (SIDS) is the leading
cause of postneonatal death among all racial and ethnic
groups, representing nearly one-third of such deaths.
The incidence of SIDS peaks in the second or third
month of life, and subsequently decreases over the first
year of life. Two components of cigarette smoke poten-
tially responsible for a role in SIDS pathophysiology are
carbon monoxide (CO) and nicotine. These toxins,
which cross the placental barrier, have been demon-
strated to alter the physiologic development of organs
and tissues most susceptible to hypoxia damage, includ-
ing the brain and heart in animal models [1-4]. Altered
autonomic nervous system function has been described
in infants born to mothers who smoked during preg-
nancy [5-7]. It has been hypothesized that CO exposure
from maternal smoking may have an effect on postnatal
electrophysiological remodeling which could predispose
the infant heart to fatal arrythmias [3]. Exposure to CO
in concentrations comparable to that of human smokers
induced a statistically significant delay in action poten-
tial (AP) duration in a rat model [3]. The potential effect
of CO exposure on AP duration is a plausible explana-
tion for an arrhythmia resulting in sudden death.
In addition to CO exposure, nicotine exposure has
also been implicated as a key cigarette component re-
lated to the cardiorespiratory events characteristic of
SIDS [8]. Although not associated with restricting uter-
ine blood flow, the cardiovascular effects of nicotine
increase the fetal heart rate while decreasing fetal heart
rate variability [9,10]. Studies demonstrating that pre-

Maternal Smoking during Pregnancy and Sudden Infant Death Using the National Maternal and 319
Infant Health Survey: A Case-Case Study
natal exposure to cigarette smoke alters lung mechanics
in infants and children have been bolstered by animal
experiments linking nicotine to altered alveolar archi-
tecture, pulmonary hypoplasia, increases in collagen
deposition and decreases in elastin content [11]. Nico-
tine is also known to adversely affect neural networks
controlling respiration by acting as a fetal neuroterato-
gen [2].
While many toxic, infectious, metabolic, physiologic
and socioeconomic factors have been proposed as causes
of SIDS, less attention has focused on the role of modi-
fiable behaviors such as smoke exposure (from maternal
and secondhand exposures) and breast-feeding [12,13].
Smoking during pregnancy continues to be a significant
public health problem, with 13.8% of women in 2005
admitting to this behavior while pregnant [14]. Smoking
rates during pregnancy in the United Kingdom were
17% in the same year and the rates in Canada were 24%
[15]. The role of secondhand smoke exposure in the
pathophysiology of SIDS is less clear, as is the role of
lower socioeconomic status which has been reported as
an independent risk factor for SIDS [13]. The majority
of studies performed to date are not based on a sample
of infants’ representative of the U.S. population.
A modifiable intervention, sleep position, has been
studied more closely. As a result of the Back to Sleep
Campaign that was implemented in 1994, SIDS mortal-
ity rates in the U.S. have declined for all ethnic groups
[16]. However, even after the implementation of the
Back to Sleep Campaign, non-Hispanic Black infants
were 2.14 times more likely to die from SIDS than
Non-Hispanic White infants [17]. This finding is parti-
cularly problematic considering White women are much
more likely to smoke during pregnancy than Black
women [14,18]. This suggests that there may be an in-
teraction between race and smoking status although this
has not been reported to date.
This study utilized the 1988 National Maternal and
Infant Health Survey (NMIHS) to examine the effect of
maternal smoking status during pregnancy and second-
hand smoke exposure on the prevalence of death due to
sudden infant death syndrome (SIDS) using a nationally
representative sample of infant deaths from the U.S.
population. Previous studies utilizing the NMIHS to exa-
mine the relationship between maternal smoking and
SIDS were incomplete, failing to account for breast
feeding or secondhand smoke exposure [19,20]. The
analysis presented in this manuscript accounts for both
maternal smoking, as well as secondhand smoke expo-
sure during pregnancy. Additionally, the analysis pre-
sented in this paper considers whether or not the infant
was breast fed, a variable previously identified as an
important protective factor against postneonatal death
[21].
2. Methods
The 1988 National Maternal and Infant Health Survey
(NMIHS) were conducted by the National Center for
Health Statistics (NCHS) to assist researchers in study-
ing factors related to poor pregnancy outcomes. The
survey is a nationally representative sample of 9953
(74.2% response rate) women who had live births, 3309
(69.3%) that had late fetal deaths and 5332 (65.3%) who
had infant deaths in 1988. The mother’s questionnaire
included information on prenatal care and health habits,
including tobacco smoke exposure, previous and subse-
quent pregnancies, characteristics of the parents and the
baby’s health through 6 months of age. Only the infant
death group from the total study population was ana-
lyzed for this report. This study was approved by the
University of Miami Institutional Review Board.
Data were abstracted from the birth certificate for the
following information: birthweight, gender, gestational
age, maternal age, infant race and parity. Data were ab-
stracted from the maternal questionnaire for maternal
education level, history of cigarette smoking during
pregnancy, and history of alcohol use during pregnancy.
The NMIHS and Longitudinal Follow-Up drew strati-
fied systematic samples from live births and infant
deaths that were registered in 48 states, the District of
Columbia, and New York City in 1988; and from fetal
deaths that were registered in 46 states, the District of
Columbia, and New York City in 1988. To increase the
reliability of the data, Black infants were oversampled in
the live-birth, fetal-death, and infant-death components
of the NMIHS because Black infants have rates of low
birthweight and infant mortality about twice that of
White infants [22]. Very low-birthweight (<1.500 g) and
moderately low-birthweight (1500 - 2499 g) infants were
oversampled in the live-birth component to obtain a suf-
ficient number of high-risk births for special studies.
The live-birth and fetal-death components were re-
stricted to women 15 years of age or over, and the in-
fant-death component included women under 15 years
of age. Mothers included in this study were women who
had infant deaths and responded to the questionnaire
from the NMIHS (N = 5332). The two study groups in
the analysis consisted of (1) SIDS deaths, as defined by
an ICD-9-CM code recorded as 798.0; and (2) other
deaths consisting of all other infant deaths [23]. This
data was obtained from linkage of the birth certificates
and death certificates.
The following study independent variables were ex-

Maternal Smoking during Pregnancy and Sudden Infant Death Using the National Maternal and
320
Infant Health Survey: A Case-Case Study
pressed both continuously and categorically: smoking
exposure, level of education, infant and maternal age,
infant and maternal birthweight, maternal BMI, gender,
secondhand smoke exposure, breast feeding, prenatal
vitamins, WIC, multiple gestation, sleep apnea monitor
prescription, sleep apnea incidents, and maternal alcohol
use.
Continuous measures are reported as means with 95%
confidence intervals and comparisons between the two
study groups (SIDS deaths vs. other deaths) used the
t-test for independent samples. Categorical measures are
reported as percents with 95% confidence intervals and
were compared using a Chi-square test. Multivariate lo-
gistic regression was performed to identify variables
significantly associated with the odds of mortality from
SIDS. Interactions between maternal smoking status and
all other covariates were tested. All analyses utilized
weighted estimates and were performed using SUDAAN
9.0.0 (Research Triangle Institute, Cary, NC) to adjust
for design effects. A p-value < 0.01 was considered sta-
tistically significant, given the very large sample size.
3. Results
Mothers of White infants were significantly more likely
to smoke during pregnancy (39.1%) than mothers of
Black infants (34.0%) or mothers of Other Race infants
(23.2%) (p < 0.001). Mothers who smoked during preg-
nancy were significantly less likely to take prenatal vi-
tamins, breast-feed the infant, and have more than a high
school education. Mothers who smoked during preg-
nancy were also significantly more likely to give birth to
multiples, use the government sponsored nutrition pro-
gram, Women , In fants and Children (WIC) program, and
drink alcohol during pregnancy. (Table 1)
3.1. Univariate Analysis
A comparison of study categorical and continuous vari-
ables, stratified by maternal race, between the SIDS
deaths and other deaths are presented in Table s 2 and 3.
All covariates included in the weighted univariate
analyses, with the exception of breast feeding and use of
prenatal vitamins, were significantly different between
the two infant death groups (p < 0.001). A significantly
greater proportion of mothers smoked during the 12
months preceding delivery in the SIDS mortality group
(53.3%) compared to the other mortality group (34.2%).
This finding held for both Black and White races, al-
though a higher proportion of White smokers were rep-
resented in the SIDS mortality group. In addition to the
significant variables reported in the unweighted analysis,
maternal prepregnancy BMI was significantly lower in
the SIDS mortality group (22.93) compared to the other
Table 1. Weighted baseline characteristics of mothers who
smoked and did not smoke during pregnanc y (N = 38,917).
Characteristic Non-smokers (%) Smokers (%)p-value
Used Vitamins 76.9 73.4 <0.001
Ever Breast fed 43.5 30.0 <0.001
Born Preterm(<37 weeks)57.5 55.0 <0.001
Multiple Gestation
Pregnancy 10.4 11.4 0.002
Used WIC 27.7 36.3 <0.001
LBW (<2.500 g) 58.9 58.7 0.756
Used Alcohol 34.3 58.1 <0.001
Maternal Education
<High school 21.8 36.1 <0.001
High school Graduate35.8 42.4
Some college
42.3 21.4
Race
White 60.9 39.1 <0.001
Black 66.0 34.0
Other race 76.8 23.2
mortality group (23.27). The lower maternal prepreg-
nancy BMI among the SIDS mortality group was not
observed for Black infants. Maternal birth weight was
significantly greater for the SIDS mortality group (3184
g) compared to the other mortality group (3122 g). This
relationship was observed among White infants, but
there was no significant difference in maternal birth-
weight between SIDS mortality groups for Black or
Other Race infants.
3.2. Multivariate Logistic Regression Analysis,
Weighted and Adjusted for Design Effects
After adjustment of standard errors using SUDAAN, the
overall model was significant (p < 0.001). Smoking
during pregnancy, use of a sleep apnea monitor, an in-
fant that stopped breathing, infant age at death, maternal
age, and infant birthweight were significant predictors of
SIDS mortality. (Table 4) Women who smoked during
pregnancy were 1.83 times more likely to give birth to
an infant that died from SIDS versus some other cause
of death, OR (95% CI) = 1.83 (1.33, 2.51). Secondhand
smoke exposure through the presence of household
smokers was not significantly associated with infant
death from SIDS in the final model. Use of a sleep apnea
monitor and an infant that was previously noted to have
stopped breathing were significantly protective factors
against SIDS death in the model. Infants older at the
time of death and infants born to older mothers were
slightly less likely to die from SIDS. The odds ratio and
95% CI for a 1 gram increase in infant birthweight were
nearly indistinguishable from 1.00, but the birthweight

Maternal Smoking during Pregnancy and Sudden Infant Death Using the National Maternal and 321
Infant Health Survey: A Case-Case Study
Table 2. Baseline categorical characteristics (weighted) of
1988 NMIHS cohort for infant mortality groups (N =
38,917).
Characteristic SIDS Death
(%) Other Death (%) p-value
N
5215 33,702
Gender
Female 40.8 44.1 <0.001
Male 59.2 55.9
Mother Smoked 12 Months
Preceding Delivery
Black 47.1 32.0 <0.001
White 59.3 35.9 <0.001
Other Race 32.4 20.8 <0.001
Smokers in Home
During Pregnancy
Black 47.1 45.3 0.230
White 56.7 38.3 <0.001
Other Race 39.3 33.7 0.090
Mother Drank Alcohol
12 Months Preceding
Delivery
Black 35.8 31.1 0.001
White 54.5 48.7 <0.001
Other 32.4 26.0 0.038
Mother Ever Breastfed
Black 15.5 16.8 0.367
White 43.0 43.8 0.481
Other Race 61.9 46.4 <0.001
Mother’s Education
Level
Total
High School Graduate 38.3 38.3 <0.001
Some College 26.7 35.9
Black
High School Graduate 39.2 40.1 <0.001
Some College 19.6 29.5
White
High School Graduate 39.9 38.4 <0.001
Some College 27.3 38.8
Other
High School Graduate 26.0 27.3 <0.001
Some College 29.8 46.2
Prenatal Vitamin Use
Black 63.6 65.6 0.140
White 78.0 80.9 <0.001
Other Race 82.5 75.4 0.017
WIC Use During
Pregnancy
Black 60.8 43.6 <0.001
White 39.4 21.7 <0.001
Other 52.8 25.8 <0.001
Multiple Gestation Birth
Black 6.4 11.4 <0.001
White 5.5 12.5 <0.001
Other Race 2.3 11.7 <0.001
Premature Bir th < 37 weeks
Total 21.6 62.6 <0.001
Black 32.6 73.4 <0.001
White 17.4 58.6 <0.001
Other 17.2 52.3 <0.001
Apnea Monitor Prescribed
Total 2.4 2.5 <0.001
Black 3.3 1.6 <0.001
White 1.9 2.8 <0.001
Other 0.0 1.2 <0.001*
Infant Ever Stopped
Breathing
Total 17.2 5.2 <0.001
Black 16.3 4.4 <0.001
White 17.8 5.4 <0.001
Other 4.2 5.2 <0.001*
*indicates insufficient sample size (at least 1 cell had a count of 0).
Table 3. Baseline continuous measures (weighted) of 1988
nmihs cohort for infant mortality groups (N = 38,917).
Characteristic SIDS Death (Mean;
95% CI) Other Death (Mean; 95% CI)
p
-value
N
5215 33,702
Number of Ciga-
rettes/Day After
Knowledge of
Pregnancy
Black 7.9 (7.3, 8.6) 8.1 (7.8, 8.4) 0.680
White 12.3 (11.9, 12.7) 9.1 (8.9, 9.3) <0.001
Other 13.9 (10.8, 17.0) 6.8 (5.9, 7.7) <0.001
Maternal Level of
Education (Years)
Black 11.6 (11.5, 11.7) 12.1 (12.0, 12.2)<0.001
White 11.9 (11.8, 12.0) 12.5 (12.4, 12.5)<0.001
Other Race 10.9 (10.4, 11.4) 12.1 (11.8, 12.3)<0.001
Maternal Age
(Years)
Black 22.8 (22.5, 23.1) 24.1 (24.0, 24.2)<0.001
White 24.5 (24.3, 24.7) 26.2 (26.1, 26.3)<0.001
Other 23.8 (23.1, 24.5) 29.1 (28.8, 29.5)<0.001
Maternal BMI
Black 23.8 (23.5, 24.1) 24.0 (23.9, 24.2)0.207
White 22.6 (22.4, 22.7) 23.0 (22.9, 23.1)<0.001
Other 23.2 (22.8, 23.7) 21.7 (21.5, 22.0)<0.001
Maternal Birth
weight (Kg)
Black 3.00 (2.95, 3.05) 3.00 (2.98, 3.02)0.993
White 3.25 (3.23, 3.27) 3.17 (3.16, 3.18)<0.001
Other 3.16 (3.06, 3.25) 3.16 (3.12, 3.21)0.876
Child’s Birth
weight (Kg)
Black 2.86 (2.82, 2.90) 1.47 (1.44, 1.49)<0.001
White 3.19 (3.16, 3.21) 1.90 (1.88, 1.92)<0.001
Other 3.17 (3.10, 3.24) 2.03 (1.95, 2.10)<0.001
Gestational Age
(Weeks) 38.4 (38.3, 38.5) 31.6 (31.5, 31.7)<0.001
Infant Age at
Death (Days) 96.6 (94.5, 98.6) 54.9 (53.7, 56.1)<0.001

Maternal Smoking during Pregnancy and Sudden Infant Death Using the National Maternal and
322
Infant Health Survey: A Case-Case Study
Table 4. Results of multiple logistic modeling* to assess the
association between maternal smoking and infant mortality
from sids (weighted), (N = 8,264).
Characteristic OR (95% CI) p-value
Smoked During Pregnancy
No 1.00
Yes 1.83 (1.33, 2.51) <0.001
Alcohol During Pregnancy
No 1.00
Yes 1.10 (0.81, 1.49) 0.527
Household Smokers
Yes 1.00
No 0.90 (0.67, 1.20) 0.476
Maternal Age 0.95 (0.92, 0.98) <0.001
Maternal Level of Education
Some college 1.00
High school Graduate 1.07 (0.75, 1.54) 0.703
< High school 0.93 (0.62, 1.38) 0.707
Multiple Gestation Pregnancy
No 1.00
Yes 1.69 (0.83, 3.48) 0.151
Child Gender
Female 1.00
Male 1.13 (0.85, 1.51) 0.400
Child Race
Black 1.00
White 0.93 (0.69, 1.25) 0.623
Other 1.96 (0.89, 4.30) 0.095
Gestational Age 0.96 (0.92, 1.01) 0.120
Infant Birthweight
1.04 (1.01, 1.06)† <0.001
Infant Breastfed
No 1.00
Yes 1.02 (0.74, 1.42) 0.885
Used Sleep Apnea Monitor
No 1.00
Yes 0.21 (0.10, 0.44) <0.001
Infant Stopped Breathing
No 1.00
Yes 0.68 (0.47, 0.97) 0.034
Infant Age at Death 0.99 (0.99, 1.00) <0.001
effect was significant in the model (p < 0.001). For a
100 gram increase in infant birthweight, there was a 4%
increased risk of death from SIDS as opposed to another
cause. Other Race infants and Black infants were more
likely to suffer SIDS mortality than White infants, but
the result was not significant in the final model. An in-
fant born as a twin or higher order gestation pregnancy
was 1.69 times more likely to suffer SIDS mortality, but
again, this result was not significant (p < 0.151). The
analysis checked for interaction between smoking status
and all other variables included in the final model and
found no interaction present.
4. Discussion
The NMIHS provides a unique opportunity to gather
information on a representative sample of live births and
infant deaths occurring in the United States. The NM-
IHS continues to be the only nationally representative
survey that provides information on such a wide range
of health behaviors and pregnancy outcomes in conjunc-
tion with birth and demographic information. For these
reasons, it remains an extremely valuable research tool
despite the fact that the data reflect population charac-
teristics that are now over 15 years old.
In this case-case study, the major finding is that
smoking during pregnancy increased the risk of infant
death from SIDS versus some other cause by 83%.
Both in this study cohort and many other populations,
women who smoke in pregnancy tend to have different
sociodemographic characteristics than non-smokers [18,
24]. Because these factors may also be associated with
risk for SIDS, concern exists that associations may re-
flect sociodemographic confounding rather than a causal
relationship. However, in this study, adjustment for fac-
tors such as maternal education and race only minimally
influenced effect sizes. We found no evidence of an in-
teraction between race and maternal smoking status.
Several researchers have corroborated the maternal
smoking and SIDS relationship. Taylor and Sanderson
used the NMIHS to conduct a study of risk factors for
SIDS and found maternal smoking during pregnancy
was significantly more common among infants that died
from SIDS than in infants dying from other causes, OR
(95% CI) = 1.97 (1.59, 2.45) [20]. They did not control
for breast feeding or other smokers in the home during
pregnancy. Schoendorf and Kiely, who also analyzed the
NMIHS data, but stratified the analysis by race, and only
included normal birthweight babies, demonstrated that
maternal smoking is a more significant risk factor for
SIDS than for other postneonatal deaths [19].
It was expected that Black infants would be at an in-
creased risk for SIDS considering the fact that previous
studies have identified Black race as a pertinent risk
factor. The final model of the present study suggested
Black infants are at a slightly increased risk for death
from SIDS compared to White infants; however, this
result was not significant. Infants of “Other Race” were
actually 1.96 times more likely to suffer SIDS mortality
than Black infants, although this result also failed to
reach statistical significance. Only 34 valid “Other
Race” infants were included in the SIDS analysis, most
of which were Asian or American Indian. Hispanic eth-

Maternal Smoking during Pregnancy and Sudden Infant Death Using the National Maternal and 323
Infant Health Survey: A Case-Case Study
nicity was distributed among White, Black, and Other
Race infants and was not readily isolated in this analysis.
Although race was not a significant predictor of SIDS
mortality, future studies should pay special attention to
American Indians, a group in which SIDS accounted for
81.8% of infant deaths in addition to representing a
group where smoke exposure during pregnancy is highly
prevalent. These results further highlight the importance
of counseling women against smoking during pregnancy
and providing access to cessation programs.
Bed-sharing and sleep position have been implicated
as risk factors for SIDS and pacifier use during sleep has
been implicated as a protective factor for SIDS; however,
these variables were not available for analysis in the
NMIHS [25,26]. The NMIHS and Longitudinal Fol-
low-Up were conducted prior to the initiation of the
Back to Sleep campaign. Sleep behaviors are pertinent
variables to consider as risk factors for SIDS mortality
because of their possible impact on the infant’s control
of arousal mechanisms, but due to limitations of the
dataset, only smoke exposure could be assessed [27].
In addition to smoking during pregnancy, use of a
sleep apnea monitor and the infant having ever stopped
breathing were significantly associated with risk of
SIDS mortality. Use of a sleep apnea monitor decreased
the risk of SIDS mortality by 79%, and noticing the in-
fant had stopped breathing reduced the risk of SIDS
mortality by 32%. The univariate analysis indicates
Black infants that died from SIDS were more likely to
have been prescribed a sleep apnea monitor and that
infants that stopped breathing were more likely to have
died from SIDS; however, the final regression model
found the association reversed. There were high per-
centages of missing cases for both of these variables,
possibly reducing the validity of conclusions drawn
about these relationships. An analysis of the NMIHS
using SIDS deaths and live controls also described a
protective effect of home apnea monitoring [28].
It is important to note when interpreting the analysis
that a distinction should be made between statistical
significance and clinical significance. The large sample
size produced by weighting often times results in very
small p-values that reflect small effect estimates of
questionable clinical relevance. In Table 1, for example,
one might expect a significantly higher percentage of
preterm infants to be born to smokers rather than to
non-smokers, rather than vice-versa. Prematurity is a
common characteristic among infants suffering all
causes of mortality; and it is likely that the statistically
significant difference seen between smokers and non-
smokers regarding prematurity is the product of the large
sample size, and not a clinically important observation.
SIDS continues to be a major cause of postneonatal
death. Although lower than in 1988, the prevalence of
smoking during pregnancy continues to be substantial,
both in the U.S. and other Western countries. With
overall smoking prevalence among women increasing in
many countries, maternal smoking is clearly an impor-
tant international health issue. In many developing
countries, the lack of adequate health facilities, and to-
bacco education or control programs can increase the
health risks of maternal smoking on the fetus and infant.
Importantly, this analysis found neither race, nor socio-
economic status, were significantly associated with
SIDS risk, suggesting universal tobacco screening and
education during pregnancy should be undertaken. Fi-
nally, while many health benefits of breast feeding have
been identified, this study failed to find a significant
protective effect of breast feeding in reducing the risk of
SIDS mortality. Further study of this topic is needed
before breast feeding can be recommended or dis-
counted as a modifiable risk factor in the prevention of
SIDS. The relationship between maternal smoking status
and SIDS was clear. The importance of abstaining from
tobacco smoke exposure during and after their preg-
nancy should be stressed when educating all pregnant
women how to maintain a healthy pregnancy and smok-
ing cessation guidance should be made available to all
pregnant women with a positive screen.
REFERENCES
[1] R. Machaalani, K. A. Waters and K. D. Tinworth, “Ef-
fects of Postnatal Nicotine Exposure on Apoptotic Mark-
ers in the Developing Piglet Brain,” Neuroscience, Vol.
132, No. 2, 2005, pp. 325-333.
doi:10.1016/j.neuroscience.2004.12.039
[2] T. A. Slotkin, “Cholinergic Systems in Brain Develop-
ment and Disruption by Neurotoxicants: Nicotine, Envi-
ronmental Tobacco Smoke, Organophosphates,” Toxi-
cology and Applied Pharmacology, Vol. 198, No. 2, 2004,
pp. 132-151. doi:10.1016/j.taap.2003.06.001
[3] L. Sartiani, E. Cerbai, G. Lonardo, et al., “Prenatal Ex-
posure to Carbon Monoxide Affects Postnatal Cellular
Electrophysiological Maturation of the Rat Heart,” Cir-
culation, Vol. 109, 2004, pp. 419-423.
doi:10.1161/01.CIR.0000109497.73223.4D
[4] R. A. Neff, S. J. Simmens, C. Evans and D. Mendelowitz,
“Prenatal Nicotine Exposure Alters Central Cardiorespi-
ratory Responses to Hypoxia in Rats: Implications for
Sudden Infant Death Syndrome,” The Journal of Neuro-
science, Vol. 24, No. 42, 2004, pp. 9261-9268.
doi:10.1523/JNEUROSCI.1918-04.2004
[5] C. A. Browne, P. B. Colditz and K. R. Dunster, “Infant
Autonomic Function Is Altered by Maternal Smoking
during Pregnancy,” Early Human Development, Vol. 59,

Maternal Smoking during Pregnancy and Sudden Infant Death Using the National Maternal and
Infant Health Survey: A Case-Case Study
324
No. 3, 2000, pp. 209-218.
doi:10.1016/S0378-3782(00)00098-0
[6] W. P. Fifer, S. T. Fingers, M. Youngman, E. Go-
mez-Gribben and M. M. Myers, “Effects of Alcohol and
Smoking during Pregnancy on Infant Autonomic Con-
trol,” Developmental Psychobiology, Vol. 51, No. 3, 2009,
pp. 234-242. doi:10.1002/dev.20366
[7] J. R. Duncan, M. Garland, M. M. Myers, et al., “Prenatal
Nicotine-Exposure Alters Fetal Autonomic Activity and
Medullary Neurotransmitter Receptors: Implications for
SIDS,” Journal of Applied Physiology, Vol. 107, No. 5,
2009, pp. 1579-1590.
doi:10.1152/japplphysiol.91629.2008
[8] O. Hafström, J. Milerad, K. L. Sandberg and H. W. Sun-
dell, “Cardiorespiratory Effects of Nicotine Exposure
during Development,” Respiratory Physiology and Neu-
robiology, Vol. 149, No. 1-3, 2005, pp. 325-341.
doi:10.1016/j.resp.2005.05.004
[9] A. Lindblad, K. Marsal and K. E. Andersson, “Effect of
Nicotine on Human Fetal Blood Flow,” Obstetrics and
Gynecology, Vol. 72, No. 3, 1988, pp. 371-382.
[10] G. Thiriez, M. Bouhaddi, L. Mourot, et al., “Heart Rate
Variability in Preterm Infants and Maternal Smoking
during Pregnancy,” Clinical Autonomic Research, Vol. 19,
No. 3, 2009, pp. 149-156.
doi:10.1007/s10286-009-0003-8
[11] R. A. Pierce and N. M. Nguyen, “Prenatal Nicotine Ex-
posure and Abnormal Lung Function,” American Journal
of Respiratory Cell and Molecular Biology, Vol. 26, No.
1, 2002, pp. 10-13.
[12] R. W. Byard and H. F. Krous, “Sudden Infant Death
Syndrome: Problems, Progress and Possibilities,” Arnold,
London, 2001.
[13] N. Spencer and S. Logan, “Sudden Unexpected Death in
Infancy and Socioeconomic Status: A Systemic Review,”
Journal of Epidemiology and Community Health, Vol. 58,
No. 5, 2004, pp. 366-373. doi:10.1136/jech.2003.011551
[14] V. T. Tong, J. R. Jones, P. M. Dietz, D. D’Angelo and J.
M. Bombard, “Trends in Smoking before, during and after
Pregnancy. Pregnancy Risk Assessment Monitoring Sys-
tem (PRAMS), United States, 31 Sites, 2000-2005,” Mor-
bidity and Mortality Weekly Report, Vol. 58, No. SS04,
2009, pp. 1-29.
[15] S. Kafouri, G. Leonard, M. Perron, L. Richer, J. R.
Seguin, S. Veillette, Z. Pausova and T. Paus, “Maternal
Cigarette Smoking during Pregnancy and Cognitive Per-
formance in Adolescence,” International Journal of Epi-
demiology, Vol. 38, No. 1, 2009, pp. 158-172.
doi:10.1093/ije/dyn250
[16] M. Willinger, C.-W. Ko, H. J. Hoffman, R. C. Kessler
and M. J. Corwin, “Factors Associated with Caregivers’
Choice of Infant Sleep Position, 1994-1998: The National
Infant Sleep Position Study,” The Journal of American
Medical Association, Vol. 283, No. 16, 2000, pp. 2135-
2142. doi:10.1001/jama.283.16.2135
[17] P. Muhuri, M. F. MacDorman and T. M. Ezzati-Rice,
“Racial Differences in Leading Causes of Infant Death in
the United States,” Paediatric and Perinatal Epidemiol-
ogy, Vol. 18, No. 1, 2004, pp. 51-60.
doi:10.1111/j.1365-3016.2004.00535.x
[18] T. J. Mathews, “Smoking during Pregnancy in the
1990s,” National Vital Statistics Reports, Vol. 49, No. 7,
2001, pp. 1-14.
[19] K. C. Schoendorf and J. L. Kiely, “Relationship of Sud-
den Infant Death Syndrome to Maternal Smoking during
and after Pregnancy,” Pediatrics, Vol. 90, No. 6, 1992, pp.
905-908.
[20] J. A. Taylor and M. Sanderson, “A Reexamination of the
Risk Factors for the Sudden Infant Death Syndrome,” The
Journal of Pediatrics, Vol. 126, No. 6, 1995, pp. 887-891.
doi:10.1016/S0022-3476(95)70202-4
[21] A. Chen and W. J. Rogan, “Breastfeeding and the Risk of
Postneonatal Death in the United States,” Pediatrics, Vol.
113, No. 5, 2004, pp. 435-439.
doi:10.1542/peds.113.5.e435
[22] Public Health Service, “National Center for Health Statis-
tics. Health, United States, 1991,” Hyattsville, USA, 1992.
[23] Public Health Service and Health Care Financing Ad-
ministration, “International Classification of Diseases,
Ninth Revision, Clinical Modification,” Washington, USA,
1980.
[24] Department of Health and Human Services, “National
Center for Health Statistics. Health, United States, 1998
with Socioeconomic and Health Chartbook,” Hyattsville,
USA, 1998.
[25] M. Mitka, “Policy Targets Lowering SIDS Incidence,”
The Journal of American Medical Association, Vol. 294,
No. 19, 2005, p. 2420. doi:10.1001/jama.294.19.2420
[26] D.-K. Li, M. Willinger, D. B. Petitti, R. Odouli, L. Liu
and H. J. Hoffman, “Use of a Dummy (Pacifier) during
Sleep and Risk of Sudden Infant Death Syndrome (SIDS):
Population Based Case-Control Study,” British Medical
Journal, Vol. 332, No. 7532, 2005, p. 18.
doi:10. 1136/bmj.38671.640475.55
[27] H. L. Richardson, A. M. Walker and R. S. Horne, “Ma-
ternal Smoking Impairs Arousal Patterns in Sleeping In-
fants,” Sleep, Vol. 32, No. 4, 2009, pp. 515-521.
[28] M. H. Malloy and H. J. Hoffman, “Home Apnea Moni-
toring and Sudden Infant Death Syndrome,” Preventive
Medicine, Vol. 25, No. 6, 1996, pp. 645-649.
doi:10.1006/pmed.1996.9998