Matern Child Health J (2012) 16:1224–1231

DOI 10.1007/s10995-011-0848-z

Characteristics and Factors Associated with the Risk of a Nicotine

Exposed Pregnancy: Expanding the CHOICES Preconception
Counseling Model to Tobacco
Danielle E. Parrish • Kirk von Sternberg •
Mary M. Velasquez • Jerry Cochran •
McClain Sampson • Patricia Dolan Mullen

Published online: 7 July 2011

Ó Springer Science+Business Media, LLC 2011

Abstract The preconception counseling model tested in
the CDC funded Project CHOICES efficacy trial to reduce
the risk of an alcohol-exposed pregnancy (AEP) could be
extended to smokers to prevent a nicotine-exposed preg-
nancy (NEP), when pharmacotherapy can be provided
safely and disclosure of these risk behaviors is more likely.
The CHOICES model, which incorporates motivational
interviewing, encourages reduction of AEP risk by
decreasing risky drinking or using effective contraception;
in the efficacy trial, most women chose both options. We
conducted a secondary analysis of the CHOICES epidemi-
ologic survey data (N = 2,672) (Project CHOICES
Research Group in Am J Prev Med 23(3), 166–173, 2002) to
identify the prevalence of risk of NEP and the factors
associated with this risk using logistic regression modeling
procedures. Conducted in six settings with women at risk for
AEP, the percentage of AEP was 12.5% (333/2,672) among
women of childbearing age (18–44). A total of 464 of the
2,672 (17.4%) were at risk for NEP. Among women at-risk
of an unplanned pregnancy (n = 1,532), the co-occurrence
of AEP and NEP risk was more prevalent (16.3%) than AEP
risk alone (5.5%) or NEP risk alone (14.0%). In the multi-
variable model, statistically significant correlates for NEP
risk included lifetime drug use, prior alcohol/drug treatment,
drug use in the last 6 months, being married or living with a
partner, having multiple sexual partners in the last 6 months,
physical abuse in the last year, and lower levels of educa-
tion. These findings suggest that preconception counseling
for NEP could be combined with a program targeting AEP.
Keywords Nicotine
Pregnancy
Alcohol

Preconception
Prevention
Introduction

Danielle E. Parrish: Work completed as a post-doctoral research
fellow at the University of Texas at Austin, School of Social Work,
Health Behavior Research and Training Institute.
D. E. Parrish (&)
K. von Sternberg
M. M. Velasquez

J. Cochran
M. Sampson
Health Behavior Research and Training Institute,
University of Texas at Austin, School of Social Work,
Austin, TX, USA
e-mail: dparrish@uh.edu
P. D. Mullen
Center for Health Promotion and Prevention Research,
University of Texas School of Public Health (Houston),
Houston, TX, USA
Present Address:
D. E. Parrish
Graduate College of Social Work, University of Houston,
Houston, TX, USA
Women of childbearing age constitute the largest propor-
tion (21.9%) of female smokers, with as many as 22% of
women smoking during pregnancy and 14% throughout the
entire pregnancy [1–4]. Smoking during pregnancy is the
leading preventable cause of low birth weight, chiefly
growth restriction, and is widely recognized as having
deleterious effects during and after pregnancy [3]. Of the
60% of pregnant women who continue to smoke past their
first antenatal visit, non-pharmacological smoking inter-
ventions rarely result in a prenatal quit rate greater than
20% and they make almost no impact on heavier smokers
who are disproportionately poor, undereducated, and whose
social networks are saturated with smokers [5, 6] and whose
pregnancies are most likely to have poor outcomes.
One of the most powerful interventions available in the
general population, pharmacologic aids [7, 8], have been

123
Matern Child Health J (2012) 16:1224–1231
difficult to test in pregnant women [9–11]. Consequently,
evidence for the safety and efficacy of such aids for cessation
is unclear, and medication is not recommended as the first
line of treatment for pregnant smokers. It appears, therefore,
more promising to focus on the preconception period to
prevent smoking during pregnancy, especially with women
who are more strongly addicted to nicotine or are less likely
to present early in pregnancy for prenatal care.
Preconception counseling, an idea first introduced in
1987 [12], has been tested across a broad range of risks
(low rubella titers) and pregnancy/fetal health promoting
behaviors (folic acid supplementation) and more recently,
control of diabetes and hypertension [13, 14]. Since just
under half of pregnancies are unintended [15, 16], and
many women do not realize they are pregnant until several
months after conception, a preconceptional approach to
prevention may help to reduce or eliminate health risk
behaviors or conditions that may lead to a problematic
pregnancy before or in the early stages of pregnancy [13].
Project CHOICES was developed through a program of
research that first identified high-risk populations [1], then
tested the feasibility and impact of the intervention [17],
and, finally, tested the impact of the intervention in a
randomized controlled trial [18]. The CHOICES precon-
ception counseling model, which incorporates motivational
interviewing, was efficacious in reducing the risk of alco-
hol-exposed pregnancy (AEP) among women of child-
bearing age by helping them reduce risky drinking,
improve use of effective contraception, or both; most
women chose both [18]. The current paper is based on a
secondary data analysis from the CHOICES epidemiologic
survey in which high-risk populations were identified.
Recent epidemiologic studies have found that con-
sumption of both alcohol and nicotine result in a multi-
plicative increase in negative health outcomes for mothers
and infants, such as cancer and cardiovascular disease for
the mother and more deleterious prenatal and neurocogni-
tive outcomes for the child [19, 20]. Moreover, a recent
review of the literature has concluded that smoking ces-
sation is unlikely to deter abstinence from alcohol or other
drugs, and may even result in greater abstinence [21].
Although we know that individual risk behaviors can be
effectively targeted and changed (e.g., tobacco use, risky
drinking), there remains a need to develop and test inter-
ventions that target multiple risk behaviors which fre-
quently present together in real practice settings [21, 22].
Integrated screening and treatment models have promise
for improving both the efficiency and effectiveness with
which such services are provided [22]. Given the common
co-occurrence of smoking and drinking among high-risk
women during both the preconceptional and prenatal
periods [23, 24], and the paucity of literature that has
proposed or tested bundled, multi-risk prevention models
1225
for this population, there exists a need to adapt, develop
and test such models.
The Project CHOICES model has promise for future
adaptation to prevent nicotine-exposed pregnancy (NEP), or
perhaps to address both risk factors (AEP and NEP) simul-
taneously. However, research is needed to determine the
characteristics and risk factors associated with the risk of
NEP among women in the preconception period to inform
future targeted prevention efforts [25], and to identify the
prevalence of women at risk for both NEP and AEP. The
current study relies on the original CHOICES cross-sectional
survey data (N = 2,672) to: (1) identify the prevalence of the
risk of NEP, (2) examine the overlap between the risk of AEP
and NEP, (3) identify bivariate correlates of NEP risk, and
(4) use multivariable analysis to identify the most parsimo-
nious set of predictors of NEP risk.
The Project CHOICES epidemiologic dataset is appro-
priate to address these aims as it provides a large sample of
women already identified as high-risk on numerous health
behaviors, including use of illicit drugs, risky alcohol
consumption, and unsafe sexual behaviors, such as trading
sex for drugs or money [1]. In addition, the sample has
been identified as having a large proportion of women who
reported smoking (50%), with as many as 44% of fertile,
sexually active women who were not currently pregnant or
trying to get pregnant reporting ineffective contraception
[1]. Additionally, drinking and smoking are frequently
comorbid [26]. In this study, of those at risk for AEP, 70
percent were also smokers [1].
Methods
Design and Sample
The existing data (n = 2,672) were collected as a cross-
sectional epidemiologic survey (November 1998 to February
2000) of women of childbearing age (18–44 years) to assess
the prevalence of AEP risk [1]. Women were recruited from
six settings that were suspected to contain a large proportion
of women at risk of an alcohol-exposed pregnancy (AEP),
with the primary aim of identifying the prevalence of AEP
risk and characteristics associated with this risk. These set-
tings included a large urban county jail and two publicly-
funded drug and alcohol treatment facilities in Texas; an OB/
GYN clinic in Virginia; a media-recruited sample in Florida;
and two primary care clinics, one in Virginia and one in
Florida. Given the diversity of settings, several different
recruitment methods were used. A probability sample was
obtained for the jail subsample; women from the Texas
residential treatment facilities were recruited from admis-
sions lists during regular visits by the assessor. In the Vir-
ginia clinics, flyers were posted and distributed to recruit
123
1226
women. In the Florida primary care clinics, potential par-
ticipants were identified by reviewing the appointment list
for scheduled and ‘‘walk-in’’ appointments. Finally, the
media sample was recruited from Broward County, Florida
(e.g., newspaper, cable television, and flyers).
Study data were collected in person by trained inter-
viewers. Interviewers received 2 days of local training,
using standard procedures and training manuals, and were
monitored regularly. Participants received compensation
for lost time ($10 cash or deposits in jail commissary
accounts [Texas], $35 vouchers for establishments such as
fast food restaurants [Virginia]), and $25 cash [Florida]).
Across sites, 3,219 women were given information about
the survey and asked to participate and 2,673 agreed to
participate, yielding an overall consent rate of 83%. The
consent rate by site includes: 94% in the jail and 100% in
the treatment centers; 67 and 76% in the Virginia gyne-
cology and primary care clinics, respectively; and 85% in
both of the Florida settings.
Measures
Independent Measures
The survey included closed-ended questions and took about
20 min to complete [1]. Data were collected on socio
demographic characteristics, current and past alcohol use,
relationship status, abuse history, current and past smoking,
obstetric history, current sexual behavior, current and past
contraceptive use, alcohol and drug use during a previous
pregnancy, mental health treatment, and homelessness
(Table 2). All questions that asked participants to recall a
specific time (e.g., ‘‘in the past 6 months’’) were modified
for respondents in jail and treatment settings to say ‘‘in the
6 months before coming here’’ [1]. Most measures were
from nationally accepted and validated instruments such as
the Alcohol Use Disorders Identification Test (AUDIT),
the Mental Health Index (MHI-5), the Addiction Severity
Index (ASI), and the National Health Interview Survey
(NHIS).
Definition of NEP Risk
A variable for NEP risk was created based on the AEP risk
variable from the original CHOICES epidemiological
study where at-risk participants were presumed fertile (no
reported hysterectomy, oophorectomy, or menopause),
sexually active (vaginal intercourse with at least one male
partner in the past 6 months and not using effective con-
traception), and also drinking at risky levels (current
drinking of more than seven drinks per week or consuming
five or more drinks in a single day more than once in the
past 6 months) [1]. A dichotomous NEP risk variable was
123
Matern Child Health J (2012) 16:1224–1231
computed for the present study using the entire study
sample that completed the survey (N = 2,672), where
women were presumed fertile and at risk for preg-
nancy and smoking cigarettes in the last 30 days (1 = at
risk for NEP, 0 = not at risk for NEP). Figure 1 presents a
flow diagram that demonstrates how the NEP risk variable
was formulated. Our reasons for defining the NEP risk
variable using any smoking in the past 30 days is threefold:
(1) social desirability associated with non-smoking status
often leads individuals to underreport the prevalence of
smoking [27], (2) the Surgeon General has concluded,
‘‘There is no safe level of exposure to tobacco smoke’’ and
there is not extant literature that indicates a safe level of
smoking during pregnancy [27], and (3) fewer cigarettes
consumed does not always result in less nicotine con-
sumption or health risk [28].
Data Analysis
Descriptive statistics were utilized to describe the sample
and prevalence of NEP risk using SPSS 17.0. Correlates of
risk for a NEP were based on comparing women at risk of
NEP with sexually active respondents presumed to be
fertile who were not at risk. A model building procedure
recommended by Hosmer and Lemeshow was used to
identify correlates of NEP risk using Stata 10.0 [29]. First,
the linearity of the logit for the one continuous predictor
variable (years of education) assumption was assessed and
was not found to be problematic. Next, univariable logistic
regression analyses were conducted to identify correlates
of risk for the baseline multivariable model. Variables
measuring current smoking and contraceptive behaviors
were not selected as predictor variables, as they were used
to define the at-risk sample. Variables found to be corre-
lated with NEP risk using a value of P \ .25 were entered
as the first set of predictor variables into a baseline mul-
tivariable model with NEP risk as the criterion variable. In
an effort to identify the most parsimonious model, pre-
dictor variables that were not correlated with NEP risk,
using a value of P \ .10, in the multivariable model were
excluded from the subsequent model. To verify the
importance of remaining predictor variables and to help
determine that important variables were not eliminated
within this model, each variable was examined to ensure
that estimated coefficients did not change markedly in
magnitude from the baseline model to the preliminary main
effects model. Practically relevant interaction terms, based
on the extant literature, were also explored. Since the
interaction terms were not statistically significant or
meaningful, the likelihood ratio test was then used to
compare the baseline multivariable model with the new,
more parsimonious model. Because there was no signifi-
cant decrement in fit (LRT X2 = 6.16, df = 9,
Matern Child Health J (2012) 16:1224–1231
Fig. 1 Flow diagram of women at risk for alcohol or nicotine exposed pregnancy, or both
P = .72), the more parsimonious main effects model was
retained.
Results
Sample
Study participants had a mean age of 32 and were primarily
Black, non-Hispanic and between the ages of 25–44 years;
most respondents were single (70%), living below the
poverty level (55%), and employed (51%). Nearly one-
third reported being physically abused in the past year.
With regard to illicit drug use, 75 percent reported lifetime
use, while 48 percent reported using illicit drugs in the last
6 months. Over one-third (35%) of participants were
identified as drinking at risky levels. As shown in Fig. 1,
nearly a third (n = 864) of the sample were infertile, lar-
gely due to tubal ligation (n = 740). Of those who were
sexually active and fertile (n = 1,559), 44% (n = 686)
were using effective contraception.
1227
Prevalence of Risk for a Nicotine-Exposed Pregnancy
Women at-risk of NEP (n = 464, 17.4%), as previously
defined, were fertile, sexually active, reported ineffective
contraception, and smoked at least one cigarette in the last
30 days (Fig. 1). The co-occurrence of risk for NEP and
AEP among women in the sample (N = 2,672) was 9.3%
(n = 249). When considering only those women at risk of
NEP and/or AEP (n = 548), 45% were at risk for both
(n = 249). Table 1 displays a cross-tabulation of AEP and
NEP risk for women who were sexually active and fertile
(n = 1,532). Over 16% of these women were at risk for
both AEP and NEP, compared to 5.5% at risk of AEP alone
and 14.0% at risk of NEP alone. Three-fourths of women at
risk of AEP were also at risk of NEP (249/333), while more
than half of women at risk of NEP were also at risk of AEP
(249/464). Over 30% of sexually active, fertile women
were at risk of NEP, while 22% were at risk of AEP.
Settings differed with regard to NEP risk (X2 = 183.73,
df = 5, P \ .001; Table 2). The risk of NEP, like the risk
of AEP, was highest for women in substance abuse
123
1228 Matern Child Health J (2012) 16:1224–1231

treatment (32%) and jail settings (29.5%) and lowest in the
primary care clinics, gynecology clinic and media sample.
When looking only at the latter non-residential settings
combined, the prevalence of NEP risk was 10.4% (178 of
1,713).
The NEP risk subsample was primarily Black, non-
Hispanic (52%), followed by White, non-Hispanic (28%)
women. Nearly 47% of women at-risk of NEP experienced
physical violence in the last year, while the majority
reported prior illicit drug use (97%), illicit drug use in the
last 6 months (80%), drinking at risky levels (54%), and
were at-risk of AEP (53%).
Correlates of NEP Risk
Multivariable Model
In the final multivariable model (v2 = 485.48, df = 7,
P \ .001), prior use of illicit drugs (OR = 5.0; 95% con-
fidence interval [CI] = 2.7–9.0), use of illicit drugs in the
last 6 months (OR = 2.2; 95% confidence interval
[CI] = 1.6–3.0), and prior alcohol and drug treatment
(OR = 2.6; 95% confidence interval [CI] = 2.0–3.5) were
the strongest predictors of risk for a NEP (Table 4). Other
predictors of risk included (in the following order of
importance): being married or living with a partner, having
multiple partners in the last 6 months, physical abuse in the
last year, and lower levels of education.

Women at-risk of NEP were more likely to be older, White, Discussion

less educated, unemployed, uninsured and married or liv-
ing with a partner when compared to other women who
were presumed to be fertile and not at-risk (Table 3).
Women at-risk were also more likely to begin smoking
after the age of 16 years. Similarly, they were more likely
to engage in several substance abuse behaviors such as
drinking at risky levels, using drugs in the last 6 months,
and smoking and drinking during their most recent preg-
nancy. Also more common among women at-risk of NEP
were prior mental health and alcohol/drug abuse treatment,
and other risk behaviors such as multiple partners and
trading sex for gain. Finally, at-risk women were also more
likely to have experienced sexual and physical abuse in the
past year and to have been homeless for over 48 h. These
results are aggregated for all six settings as further mean-
ingful analysis by site was precluded by the relatively low
number of women at-risk in any one setting.
This study analyzed secondary data from the CHOICES
epidemiologic survey to assess the risk prevalence of nic-
otine-exposed pregnancy (NEP) among women of child-
bearing age (18–44), the factors associated with this risk,
and the proportion of women with overlapping risk of NEP
and alcohol-exposed pregnancy (AEP). Seventeen percent
of the women in the study were at risk of NEP (nearly 26%
of women presumed to be fertile), suggesting the impor-
tance of targeting women in the preconception period, when
they can safely use pharmacotherapy and the disclosure of
such risk is more likely. The findings from this study have
several important implications for the future development
and planning of preconception interventions in real settings.
Among women at-risk of an unplanned pregnancy
(n = 1,532), the co-occurrence of AEP and NEP risk was
more prevalent (16.3%) than AEP risk alone (5.5%) or NEP

Table 1 Frequency of sexually

At-risk for nicotine At-risk for alcohol exposed pregnancy Total
active, fertile women of
exposed pregnancy
childbearing age at-risk for No Yes n %
alcohol- or nicotine exposed
pregnancy, or both n % n %

No 984 64.2 84 5.5 1,068 69.7

Yes 215 14.0 249 16.3 464 30.3
Total 1,199 78.2 333 21.8 1,532 100.0

Table 2 Number and

Setting Total surveyed Number at risk Percent at risk
proportion of women at-risk of
NEP by setting Treatment center, TX 453 149 32.1
Jail, TX 506 137 29.5
Gynecology clinic, VA 425 63 13.6
Media sample, FL 452 35 7.5
Primary care #1, VA 421 53 11.4
Primary care #2, FL 415 27 5.8
*One case with missing All settings 2,672* 464 17.4
responses excluded

123
Matern Child Health J (2012) 16:1224–1231 1229

Table 3 Univariable correlates of NEP risk

Variable At-risk Not at-risk OR 95% CI At risk women more likely to
n % n %

Age
[30 229 48.8 914 41.5 1.3 1.1–1.6 Be older
Race/ethnicity
White, non-Hispanic 129 27.5 426 19.4 1.6 1.3–2.0 Be white, non-Hispanic
Education
\High school 201 42.9 664 30.2 1.7 1.4–2.1 Be less educated
Employment
Unemployed 145 30.9 492 22.4 1.6 1.2–1.9 Be unemployed
Insured (health)
Uninsured 284 60.7 1,202 54.7 1.3 1.0–1.6 Be uninsured
Age first smoked
[16 132 28.1 448 20.4 1.5 1.2–1.9 Begin smoking [age 16
Smoked during last pregnancy 307 75.8 584 31.6 6.8 5.3–8.7 Smoke—most recent pregnancy
Drank during last pregnancy 66 16.4 145 7.8 2.3 1.7–3.1 Drink—most recent pregnancy
Drink at risk levels 252 53.7 684 31.1 2.6 2.1–3.1 Drink at risky levels
Ever use illicit drugs 453 97.0 1,544 70.0 12.0 7.2–19.9 Have used illicit drugs—lifetime
Drug use in last 6 months 375 80.1 917 41.8 5.6 4.4–7.1 Have used drugs in last 6 months
Any alcohol/drug treatment 298 63.5 684 31.1 3.9 3.1–4.8 Have ever had alcohol/drug treatment
Any mental health treatment 163 34.8 604 27.5 1.4 1.1–1.7 Have ever had mental health treatment
Sexual abuse 94 20.0 217 10.0 2.3 1.8–3.0 Have been sexually abused in past year
Multiple partners 244 52.6 297 22.6 3.8 3.0–4.7 Have had more than one partner in last
6 months
Traded sex for gain 143 30.8 97 8.9 4.6 3.4–6.1 Have traded sex for money or drugs or
other gain—in last 6 months
Homeless [48 h 173 36.9 401 18.2 2.6 2.1–3.3 Have been homeless in last year
Physical abuse 219 46.8 565 25.7 2.5 2.1–3.1 Have been physically abused in last year
Married or living together 157 33.5 642 29.2 1.2 .987–1.5 Be married or live with a partner

risk alone (14.0%). Among women at risk for either alcohol
or nicotine exposure during pregnancy, close to half (45.4%)
were at risk for both. This overlap of risk suggests the
potential utility of a bundled approach to prevention in the
preconception period, whereby screening and intervention
would simultaneously target smoking, risky alcohol use, and
effective contraception. The CHOICES preconception
model, which was efficacious in reducing the risk of alcohol-
exposed pregnancy by using motivational interviewing to
assist women in reducing alcohol consumption below risk
levels and/or using effective contraception, may have
promise in expanding to also target a reduction in the risk of
NEP [18]. A bundled approach that targets multiple risk
behaviors can be personalized for each woman’s specific risk
profile and unique preferences and values (e.g., while con-
traception may not be acceptable for some women given
their religious or cultural beliefs, they can successfully
reduce their risk by choosing to reduce risky drinking and
smoking). It may also result in a more efficient service
delivery model, as the same practitioner can be trained to
apply motivational interviewing to target one or more risk
behaviors, as opposed to more traditional uni-risk prevention
models. An expanded CHOICES preconception model is
currently being tested in the CHOICES Plus efficacy trial (PI:
Mary Velasquez, Ph.D; funded by the Centers for Disease
Control and Prevention), where women are screened for the
risk of alcohol-exposed pregnancy and current smoking, and
offered a referral to an evidence-based smoking cessation
program that provides access to smoking cessation phar-
macotherapy (in addition to other behavioral treatments).
The risk of NEP was substantially higher in jail (29%) and
treatment centers (32%), despite very similar rates of tubal
ligations, hysterectomies, oophorectomies, or menopause
across all settings (59–68%, with a median of 65%). Similarly,
the risk of AEP was also considerably higher in these sub-
groups (21% in the jail; and 24% in the substance abuse
treatment centers compared to less than 10% in other settings,
[18]), suggesting that the use of preconception interventions

123
1230 Matern Child Health J (2012) 16:1224–1231

Table 4 Multivariable analysis results: variables predicting NEP risk personalizing each intervention using motivational inter-
Predictors OR 95% Women at-risk for NEP
viewing, which has been found to help individuals change a
CI were more likely to variety health issues, may lead to a more transferable and
cost-effective intervention in real settings when compared to
Any prior use of drugs
single risk interventions. Like the aforementioned CHOI-
Yes/no 5.0 2.7–9.0 Ever use drugs CES Plus project, future research should test the efficacy of
Prior alcohol/drug treatment preconception interventions that target various combinations
Yes/no 2.6 2.0–3.5 Ever have had alcohol/drug of health related risk factors.
treatment
While this study had several strengths, there are also
Recent drug use (last 6 months)
limitations. The sampling methods varied by site, affecting
Yes/no 2.2 1.6–3.0 Use drugs recently
the representativeness of some of the samples, and the
Married or living together
selected settings are not representative of all settings of their
Yes/no 2.0 1.5–2.7 Be married or live with a
type. Also, although this study relied solely on self-report
partner
measures, this concern is somewhat offset by the fact that
Multiple partners last 6 months
50% of the entire sample reported smoking and 75% reported
Yes/no 1.8 1.3–2.3 Have had multiple partners—
in last 6 months prior illicit drug use. These findings should also be inter-
Physical abuse last year preted in light of when this data was collected (1998–2000),
Yes/no 1.5 1.1–2.0 Have been physically as more recent questions for self-disclosure of tobacco use
abused—in past year have improved and trends in smoking have generally
Years of education .86 .81–.91 Be less educated decreased in the US population [32]. However, smoking
(continuous) rates have remained eminent among high risk populations,
such as those reported in this study. Finally, although the
that target multiple risk factors, if effective, may be particu- quantity of smoking was not factored into the definition of
larly important for these higher risk populations. NEP risk, there is no well-defined cutoff point at which
Within the multivariable model, recent or lifetime illicit smoking becomes detrimental during pregnancy, and there is
drug use and prior alcohol drug and alcohol treatment were considerable variation regarding how much nicotine enters
most strongly associated with NEP risk. This strong asso- the body based on the brand of cigarette and how it is smoked.
ciation, and the finding that recent illicit drug use and prior
alcohol and drug treatment were also two of the top three
predictors for AEP risk [18], suggests potential utility in Conclusions
expanding the CHOICES model to target illicit drug use.
The risk of NEP was also associated with various life Seventeen percent of women were at risk of NEP, with the
stressors, such as physical abuse. This is consistent with largest proportion of women at risk in the jails and substance
other findings, which have reported a high prevalence of abuse centers. There was also a substantial overlap between
smoking and unwanted pregnancy among women experi- the risk of alcohol-exposed pregnancy (AEP) and nicotine-
encing partner abuse [29, 30]. exposed pregnancy (NEP), suggesting the potential utility
The variation between the two primary clinics from dis- and promise of the expansion of the CHOICES preconcep-
parate geographic regions within this study may suggest the tion model to target both risks in an efficient manner in real
importance of assessing the unique risk profile of each dis- settings. Preconception counseling is important to prevent
tinct target population, as well as the possible advantage of NEP, as it offers an opportunity to initiate smoking cessation
developing interventions that can be adapted to target mul- prior to pregnancy, when it is safer to use pharmacotherapy
tiple risk factors based on a given risk profile. In doing this, and women may be more likely to disclose risky health
preconception counseling could extend beyond substance behaviors. Future research should examine the utility, effi-
use and abuse in the preconception period to encompass cacy and effectiveness of the CHOICES model to target
other chronic health and preconceptional issues, such as multiple health behaviors and assist women in planning for
elevated blood pressure or the consumption of folic acid pregnancy during preconception period.
supplements. Preconceptional interventions that rely on
motivational interviewing, such as the CHOICES interven-
tion, could then be personalized for each woman’s unique
health issues and habits, providing her the choice to reduce References
her risk by changing specific health behaviors and/or using
1. Project CHOICES Research Group. (2002). Alcohol-exposed
effective contraception to prevent an unplanned pregnancy. pregnancy: Characteristics associated with risk. American Jour-
The targeting of multiple risks and the flexibility in nal of Preventive Medicine, 23(3), 166–173.

123
Matern Child Health J (2012) 16:1224–1231
2. CDC. (2007). Preventing smoking and exposure to secondhand
smoke before, during, and after pregnancy. Atlanta, GA: Centers
for Disease Control and Prevention.
3. CDC. (2008). Early release of selected estimates based on data
from the January–June 2008 national health interview survey.
Atlanta, GA: CDC.
4. SAMHSA. (2010). Assessment and prevention programs: FASD
prevention programs. Retrieved September 27, 2010, from http://
www.fasdcenter.samhsa.gov/assessmentprevention/fasdpreven
tion.cfm.
5. Ershoff, D. H., et al. (2000). Predictors of intentions to stop
smoking early in prenatal care. Tobacco control, 9(Supp 3),
iii41–iii45.
6. Yu, S. M., et al. (2002). Factors associated with smoking cessa-
tion among US pregnant women. Maternal and Child Health
Journal, 6(2), 89–97.
7. Treating Tobacco Use and Dependence Clinical Practice Guide-
line Panel, Staff, & Consortium Representatives. (2008). Treating
tobacco use and dependence: A clinical practice guideline
update. A Public Health Service Report. Rockville MD: US
Department of Health & Human Services.
8. Lumley, J., et al. (2008). Interventions for promoting smoking
cessation during pregnancy. Cochrane Database of Systematic
Reviews, (4), Art No. CD001055. doi:10.1002/14651858.CD00
1055.pub3.
9. Coleman, T. (2007). Recommendations for the use of pharma-
cological smoking cessation strategies in pregnant women. CNS
Drugs, 21(12), 983–993.
10. Forest, S. (2010). Contraversy and evidence about nicotine
replacement therapy in pregnancy. MCN: American Journal of
Maternal Child Nursing, 35(2), 89–95.
11. Osadchy, A., et al. (2009). Nicotine replacement therapy during
pregnancy: Recommended or not recommended? Journal of
Obstetrics and Gynaecology Canada, 31(8), 744–747.
12. National Institutes of Health (DHHS). (1989). Caring for our
future: The content of prenatal care. A report of the public health
service expert panel on the content of prenatal care. Bethesda,
MD: National Institutes of Health (DHHS). http://www.eric.ed.
gov/PDFS/ED334018.pdf.
13. Johnson, K., et al. (2006). CDC/ATSDR Preconception Care
Work Group; Select Panel on Preconception Care. Recommen-
dations to improve preconception health and health care—United
States. A report of the CDC/ATSDR Preconception Care Work
Group and the Select Panel on Preconception Care. MMWR.
Recommendations and Reports, 55(RR-6), 1–23.
14. Korenbrot, C. C., et al. (2002). Preconception care: A systematic
review. Maternal and Child Health Journal, 6(2), 75–88.
15. Finer, L. B., & Henshaw, S. K. (2006). Disparities in rates of
unintended pregnancy in the United States, 1994 and 2001.
Perspectives on Sexual and Reproductive Health, 38(2), 90–96.
16. Misra, D. P., et al. (2003). Integrated perinatal health framework:
A multiple determinants model with a lifespan approach. Amer-
ican Journal of Preventive Medicine, 25, 65–75.
17. The Project CHOICES Intervention Research Group. (2003).
Reducing the risk of alcohol-exposed pregnancies: A study of a
motivational intervention in community settings. Pediatrics, 111,
1131–1135.
1231
18. Floyd, R. L., et al. (2007). Preventing alcohol-exposed pregnan-
cies: A randomized controlled trial. American Journal of Pre-
ventive Medicine, 32(1), 1–10.
19. De Leon, J., et al. (2007). Association between smoking and
alcohol use in the general population: Stable and unstable odds
rations across two years in two different countries. Alcohol and
Alcoholism, 42, 252–257.
20. Drobes, D. J. (2002). Concurrent alcohol and tobacco depen-
dence: Mechanisms and treatment. Alcohol Research and Health,
26, 136–142.
21. Kalman, D., Kim, S., DiGirolamo, G., Smelson, D., & Ziedonis,
D. (2010). Addressing tobacco use disorder in smokers in early
remission from alcohol dependence: The case for integrating
smoking cessation services in substance use disorder treatment
programs. Clin Psychol Rev, 30, 12–24.
22. Goldstein, M. G., et al. (2004). Multiple behavioral risk factor
interventions in primary care. American Journal of Preventive
Medicine, 27(2S), 61–79.
23. Tsai, J., et al. (2007). Patterns of average volume of alcohol use
among women of childbearing age. Maternal Child Health
Journal, 11(5), 437–445.
24. Tsai, J., et al. (2010). Concurrent alcohol use or heavier use of
alcohol and cigarette smoking among women of childbearing age
with accessible health care. Prevention Science, 11, 197–206.
25. Floyd, R. L., et al. (2008). The clinical content of preconception
care: Alcohol, tobacco, and illicit drug exposures. American
Journal of Obstetrics and Gynecology, 199(6 Suppl 2), S333–
S339.
26. Martin, L. T., et al. (2008). Correlates of smoking before, during,
and after pregnancy. American Journal of Health Behavior,
32(3), 272–282.
27. US Department of Health and Human Services. (2010). How
tobacco smoke causes disease: The biology and behavioral basis
for smoking–attributable disease: A report of the surgeon gen-
eral. Atlanta, GA: US Department of Health and Human Ser-
vices, Centers for Disease Control and Prevention, National
Center for Chronic Disease Prevention and Health Promotion,
Office on Smoking and Health.
28. Hurt, R. D., Croghan, G. A., Wolter, T. D., Croghan, I. T., Of-
ford, K. P., Williams, G. M., et al. (2000). Does smoking
reduction result in reduction of biomarkers associated with harm?
A pilot study using a nicotine inhaler. Nicotine & Tobacco, 2(4),
327–336.
29. Hosmer, D. W., & Lemeshow, S. (2000). Applied logistic
regresstion (2nd ed.). New York: Wiley-Interscience.
30. Hathaway, J. E., et al. (2000). Health status and health care use of
Massachusetts women reporting partner abuse. American Journal
of Preventive Medicine, 19(4), 302–307.
31. McNutt, L. A., et al. (2002). Cumulative abuse experiences,
physical health and health behaviors. Annals of Epidemiology,
12(2), 123–130.
32. Garrett, B.E., Dube, S.R., Trosclair, A., Caraball, R.S., & Pech-
acek, T.F. (2011). Cigarette smoking—United States, 1965–2008.
Supplements, 60(01): 109–113. National Center for Chronic
Disease Prevention and Health Promotion, CDC.
123
Copyright of Maternal & Child Health Journal is the property of Springer Science & Business Media B.V. and
its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for individual use.