Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2019 May 1.
Published in final edited form as: Neurotoxicol Teratol. 2018 Mar 6;67:18–24. doi: 10.1016/j.ntt.2018.02.003

Positive parenting behaviors in women who spontaneously quit smoking during pregnancy: Clues to putative targets for preventive interventions

Suena H Massey a,b,c,✉, Daniel K Mroczek c, James L Burns c, Caron AC Clark d, Kimberly A Espy e,f, Lauren S Wakschlag b,c,g
PMCID: PMC5970967  NIHMSID: NIHMS952006  PMID: 29501649

Abstract

Background

While the majority of pregnant smokers do not respond to intervention, little is known about how a subset of pregnant smokers known as spontaneous quitters achieve sustained biologically-confirmed abstinence through delivery in the absence of intervention. We explore a developmental framework to address this question by viewing spontaneous quitting as an adaptive parenting behavior, facilitated by abilities necessary for sensitive parenting, or responsiveness. Utilizing existing data, we examined responsiveness from parenting assessments in women who exhibited a variety of smoking patterns during pregnancy, including spontaneous quitting.

Methods

Participants were N=305 pregnant women assessed for smoking prospectively and biochemically at 16 weeks, 28 weeks, delivery, and 4 weeks postpartum, then reassessed with their children 5 years later with directly-observed home- and lab-based measures of parenting. We used linear regression analysis to compare spontaneous quitters with women who exhibited other prenatal smoking patterns on parenting responsiveness, controlling for potential confounders.

Results

In home-based observations, spontaneous quitters (n=22) exhibited greater responsiveness with their children relative to intermittent pregnancy smokers (n=70; β=.258, p=.022]; persistent pregnancy smokers [n=66; β=.228, p=.040]; former smokers (quit before pregnancy) [n=78; β=266, p=.028]; and never smokers [n=69; β=.312, p=.009]. Hypothesized differences were not observed in lab-based and self-report measures.

Conclusions

Putative protective characteristics in spontaneous quitters were captured in mother-child interactions at home, but not in lab-based and maternal report measures of responsiveness. Specification of these characteristics using prospective designs that oversample for spontaneous quitters is recommended to enable translation to preventive interventions.

Keywords: Pregnancy smoking, prenatal tobacco exposure, protective factors, maternal sensitivity

1. Introduction

Following over 40 years of intervention research, maternal smoking during pregnancy (MSDP) remains a leading modifiable risk factor for adverse neonatal, neurodevelopmental, and cardiometabolic outcomes across the lifespan (Johnson et al., 2017). Behavioral interventions recommended by the U.S. Preventive Services Task Force and financial incentive-based approaches to this problem are ineffective for the majority of pregnant smokers who have not quit on their own (Chamberlain et al., 2013). Understudied, is how a subset of smokers known as spontaneous quitters, constituting 40–65% of privately-insured pregnant smokers and 11–28% of publicly-insured smokers, successfully suspend smoking in early pregnancy, and remain abstinent through delivery, as confirmed with biomarkers (Solomon and Quinn, 2004). Interestingly, spontaneous quitters achieve far better 6-month abstinence rates than non-pregnant smokers who are highly-motivated to quit for reasons other than pregnancy, for example, men and women who recently suffered a heart attack (Dornelas et al., 2000). Indeed, when viewed as a natural intervention, pregnancy is associated with a greater protective effect on women’s drug use than any intervention, independent of demographic, environmental, and heritable influences (Kendler et al., 2017) including women’s intentions to raise the child (Massey et al., 2011; Massey et al., 2012; Massey et al., 2016). Yet the mechanisms that underlie pregnancy’s apparent protective effect are unclear, and if elucidated, could substantially improve the effectiveness of interventions (Massey and Wisner, In Press).

Towards this end, we have shown that MSDP exists within a broader interpersonal framework that includes women’s psychosocial context (Wakschlag et al., 2003), the maternal fetal relationship (Massey et al., 2015a; Massey et al., 2012), and ways that women tend to relate to others (Bradstreet et al., 2012; Maxson et al., 2012). Moreover, as genetic factors and social cognitive abilities implicated in caregiving behavior have been linked to variability in MSDP (Massey et al., 2015b), the highly evolutionarily-conserved neurobiological processes that subserve caregiving may facilitate attempts to abstain from smoking while pregnant (Massey et al., 2017).

Indeed, the belief that ‘smoking harms [the] baby’ is associated with a 14-fold increase in the odds of quitting smoking during pregnancy—an effect that far exceeds the effect of nicotine dependence (White et al., 2014). Thus, while spontaneous quitters may have lower nicotine dependence and fewer psychosocial risks relative to persistent pregnancy smokers (Solomon and Quinn, 2004), yet-to-be identified protective characteristics (and/or their interaction with the recognition of a pregnancy) beyond a relative absence of risks, likely account for the well-documented positive changes in smoking and other drug use observed in pregnant women. For example, we have provided support, in our earlier work, for a protective developmental trajectory associated with quitting smoking during pregnancy. In a large epidemiologic sample, we found that toddlers of mothers who quit smoking during pregnancy exhibited fewer behavioral problems relative to toddlers of mothers who did not quit, and also toddlers of women who had never smoked (Hutchinson et al., 2010). Thus, parallel to the well-studied risk pathway associated with MSDP, there appears to be a protective pathway associated with quitting smoking during pregnancy (Pickett et al., 2008). We extend this work to elucidate positive characteristics that differentiate spontaneous quitters from other pregnancy smokers and also other pregnant women.

We have previously proposed as a candidate protective factor, the ability to accurately perceive the needs of others and respond by helping, described by the construct, empathy {Decety, 2016 #792}. Empathy is fundamental to maternal caregiving {Eisenberg, 1987 #94} and can be approximated through the degree of responsiveness observed in parents when interacting with their children (Bornstein and Tamis-LeMonda, 1997; Wakschlag and Hans, 2002). Thus, in the absence of direct assessments of empathy in pregnant smokers at the present time, in this study, we aimed to provide initial support for the hypothesized empathy-related model of spontaneous quitting (Massey et al., 2017). Using existing data from a large, well-characterized prenatal smoking cohort with follow-up assessments of observed parenting responsiveness, (used here, as a proxy measure of empathy during pregnancy), we tested the following hypotheses. First, we expected that spontaneous quitters would exhibit greater responsiveness relative to smokers who did not quit. Second, as the relevance of responsiveness to quitting smoking is likely unique to pregnancy, we expected spontaneous quitters would be more responsive than women who had quit smoking earlier in life, presumably for reasons other than the pregnancy (i.e., former smokers). Finally, to replicate our earlier findings about a protective developmental trajectory associated with quitting smoking during pregnancy, we hypothesized that spontaneous quitters would also be more responsive than women who had never smoked.

2. METHODS

2.1 Participants

Women were recruited in early pregnancy (65% prior to 16 weeks, 100% by 28 weeks) over a 4.5 year period from obstetric clinics in Southern Illinois and in a mid-sized city in Nebraska for the Midwest Infant Development Study (MIDS)(Wang et al., 1997). As the MIDS was designed to compare tobacco-exposed and non-exposed infants on developmental outcomes while minimizing the influence of other exposures, cigarette smokers were oversampled, while women who reported illegal drug use or consumption of four or more alcoholic drinks in a single sitting at enrollment were excluded. At 4 weeks postpartum, women reported on infant temperament. When children were approximately 5 years of age (M=5.06, SD=0.26; range=1.37), mothers and children were reassessed in a preschool follow-up study (MIDS-P). Parent-child interactions were assessed during a one-hour home visit, during a 1.5 hour laboratory visit, and by maternal report. Figure 1 shows the derivation of the analytic sample. Since women with multiple births (n=7 sets of twins) were assessed with each child separately, resulting in two sets of parenting measures, we excluded these dyads from the current analysis. All procedures were approved by the Institutional Review Boards of the University of Nebraska–Lincoln (MIDS), University of Illinois—Carbondale (MIDS), and Northwestern University (MIDS-P) and conducted after informed consent procedures.

Figure 1.

Figure 1

Open in a new tab

Derivation of analytic sample (N=305 mother-child dyads).

LMP = Last menstrual period preceding the studied pregnancy

2.2 Maternal smoking

During pregnancy, smoking patterns were characterized prospectively using repeated timeline follow back interviews (Sobell and Sobell, 1996) and cotinine assays at 16 and 28 week visits and at delivery (M = 4.3 visits, SD = 1.1, range 1 – 6). At baseline, we also assessed smoking history, including any lifetime smoking (yes/no), current smoking (yes/no), partner smoking (yes/no), and, for lifetime smokers who were not currently smoking, their quit date. Concentration of cotinine, the major proximate metabolite of nicotine exposure (Benowitz, 1996), was analyzed from maternal urine specimens collected at 16 and 28 weeks of gestation, from infant meconium at delivery, and from maternal urine at 4 weeks postpartum, using gas chromatography-mass spectroscopy with a limit of detection of 4 ng/ml (United States Drug Testing Laboratories, Inc., Des Plaines, IL). At the preschool follow-up, mothers reported on average cigarettes/day smoked in the past week.

2.2.1 Categorization of prenatal smoking groups

While pregnancy-specific cut-offs are typically used to distinguish true smoking from environmental smoke exposure by pregnant women (Dempsey et al., 2002; Kim, 2016), cutoffs do not distinguish between smoking that occurred one or more days prior to specimen collection from environmental tobacco exposure (George et al., 2006). Since certainty about successful abstinence was of utmost importance for the current study, we favored under-identification rather than over-identification of spontaneous quitters, and considered any nonzero cotinine value, a priori, as a positive indication of smoking. Based on all available interview and biological data, we categorized women into the following groups (Table 1):

Table 1.

Descriptive characteristics by prenatal smoking pattern

A - E A B C D E A - E C vs. D+E B vs. C A vs. C
Total (N = 305) Never smokers (n = 69) Former smokers (n = 78) Spontaneous pregnancy quitters (n = 22) Intermittent pregnancy smokers (n = 70) Persistent pregnancy smokers (n = 66) All groups Spontaneous quitters vs. pregnancy smokers Spontaneous quitters vs. former smokers Spontaneous quitters vs. never smokers
Timing of assessments Mean (SD) or % p or χ2
Prenatal (16–28 weeks)
Age, mean (SD) 25.9 (4.9) 27.4 (5.5) 25.7 (3.8) 26.1 (6.2) 26.0 (5.0) 24.2 (4.2) .006 .479 .722 .411
Non- Hispanic White (%) 81.6 73.9 84.6 95.5 94.3 78.8 .153 .129 .287 .035
Unmarried (%) 58.0 43.5 48.7 63.6 70.0 69.7 .002 .621 .237 .141
Education in years 13.3 (2.0) 14.0 (1.9) 13.7 (2.0) 13.2 (2.3) 13.1 (2.0) 12.3 (1.6) <.001 .276 .331 .094
Multiparous (%) 59.0 63.8 61.5 59.1 48.6 62.1 .371 .819 .999 .801
Maternal income in USD 824 (803) 904 (8.3) 838 (785) 988 (1351) 870 (679) 623 (651) .201 .197 .623 .784
Partner smoking (%) 51.3 27.5 39.7 63.6 61.4 75.4 <.001 .806 .055 .004
a Nicotine dependence -- -- -- 1.6 (0.9) 1.7 (0.8) 2.9 (1.9) -- .205 -- --
Cig/day (cotinine-corrected) 1.5 (2.9) -- -- -- 0.7 (1.6) 2.9 (4.0) -- -- -- --
Cotinine, maternal (ng/ml) 255 (390) -- -- -- 107 (231) 495 (455) -- -- -- --
Cotinine, meconi um (ng/g) 114 (653) -- -- -- 55.7 (202) 505 (1386) -- -- -- --
Alcohol use (% within group) 40.0 16.7 34.9 47.4 45.2 59.6 <.001 .806 .420 .013
b Depressive symptoms 2.8 (3.5) 2.1 (2.8) 3.3 (3.9) 3.1 (3.3) 2.7 (3.1) 3.1 (3.9) .299 .729 .859 .210
c Hostility 3.3 (2.9) 2.8 (2.4) 3.2 (3.1) 3.7 (2.6) 3.5 (2.9) 3.6 (3.0) .448 .828 .522 .170
d Infant Temperament (4 weeks postpartum)
Easy (%) 33.7 40.9 43.8 15.8 30.8 26.3 .060 .166 .025 .043
Slow to warm (%) 37.8 30.3 34.2 57.9 38.5 43.9 .190 .334 .060 .028
Difficult (%) 28.5 28.8 21.9 26.3 30.8 29.8 .789 .722 .684 .833
Preschool Follow-Up (Age 5)
e Depressive symptoms 6.3 (5.2) 6.0 (5.4) 5.6 (4.8) 6.0 (4.9) 7.1 (5.5) 6.6 (5.4) .439 .478 .733 .983
f Impulsivity 8.5 (5.4) 8.4 (4.8) 8.5 (5.2) 8.1 (5.3) 8.5 (5.8) 8.6 (6.0) .996 .726 .781 .800
g Antisocial behavior 2.6 (2.2) 1.6 (1.9) 2.2 (1.7) 3.8 (2.2) 2.9 (2.3) 3.6 (2.5) <.001 .353 .001 <.001
h Family conflict 45.0 (9.9) 43.4 (8.7) 44.5 (9.6) 44.4 (9.5) 44.7 (10.3) 48.2 (10.9) .057 .426 .990 .641
Maternal smoking (cig/day) 4.2 (6.6) 0.1 (0.9) 0.9 (3.4) 6.1 (6.2) 4.5 (6.3) 12.0 (6.6) <.001 .277 <.001 <.001
Maternal responsiveness
i EC-HOME 3.6 (1.9) 3.4 (2.0) 3.5 (2.0) 4.5 (1.5) 3.7 (1.8) 3.5 (1.6) .178 .011 .016 .009
j P-COS 12.4 (3.1) 13.3 (3.1) 12.6 (2.5) 11.9 (3.0) 12.2 (33.2) 12.4 (3.1) .079 .977 .334 .110
k PSDQ 4.5 (0.4) 4.5 (0.4) 4.5 (0.4) 4.5 (0.5) 4.5 (0.4) 4.5 (0.4) .976 .975 .997 .886
Open in a new tab
a

Fagerstrom Test of Nicotine Dependence;

b

Brief Symptom Inventory, Depression subscale;

c

Brief Symptom Inventory, Hostility subscale;

d

Early Infant Temperament Questionnaire;

e

Center for Epidemiological Studies-Depression Scale, 9-item version;

f

Connors Adult ADHD Rating Scale, Impulsivity Subscale;

g

Zoccolillo Antisocial Behavior Questionnaire;

h

Family Environment Scale, Conflict subscale;

i

EC-HOME = Early Childhood Home Observation for Measurement of the Environment, Responsivity Subscale;

j

P-COS = Parent-Child Observation Schedule, Responsive Involvement Domain;

k

PSDQ = Parenting Styles and Dimensions Questionnaire, Warmth and Support dimension

  1. Never smokers - reported never having smoked in their lifetime or during pregnancy and all cotinine values were zero;

  2. Former smokers - reported lifetime smoking, quit prior to month of LMP (mean time = 5.5 years prior to LMP; range = 0.3 to 16.9 years), reported no smoking during pregnancy and all cotinine values were zero;

  3. Spontaneous quitters - reported smoking in month of the LMP, quit before the baseline visit, reported no smoking during pregnancy and all cotinine values were zero;

  4. Intermittent pregnancy smokers - evidence of smoking either from self report or cotinine at some but not all visits; and

  5. Persistent pregnancy smokers - evidence of smoking from either self report or cotinine at every visit.

2.3 Responsiveness

2.3.1 In-home observation

Parenting was assessed by direct observation and structured interviews in participants’ homes using the Early Childhood Home Observation for Measurement of the Environment (EC-HOME) (Totsika and Sylva, 2004). This descriptive profile, designed for children 3 – 6 years of age, measures the quality and quantity of support and stimulation available to a child in his/her home environment in a naturalistic context (Bradley, 1993) and has been validated in both community and high-risk samples (Totsika and Sylva, 2004). The EC-HOME was coded independently by two trained research assistants (weighted Kappa = .85 – 1 for all subscales). For hypothesis testing, we used the 7-item Responsivity Subscale (range 0 – 7) which quantifies warm and responsive interactions towards the child (Cronbach’s α = .619). An example interview question from this subscale is: “As your child has gotten older, does he/she still want to be held by you or sit close to you?” Sample observations assessed are: “Parent answers child’s questions or requests verbally” and “Parent caresses, kisses, or cuddles child twice during visit.”

2.3.2 Direct-observation in the lab

Responsiveness was also assessed with the Responsive Involvement Domain (range 0 – 20) of the Parent Clinical Observation Schedule (PCOS) (Hill et al., 2008) based on parenting behaviors during a 20-minute interaction involving a series of standardized laboratory activities specifically designed to elicit variations in parent engagement across different interactional contexts (working on a challenging puzzle together, putting crayons back into a box together, parent asking child to work independently while he/she is busy filling out a questionnaire). Specific parenting behaviors viewed in video recordings were coded for responsiveness on a 4-point scale (0=none to 3=high) by two trained independent raters (weighted Kappa=0.77).

2.3.3 Maternal report and correlations among responsiveness measures

During the laboratory visit, mothers completed the Parenting Styles and Dimensions Questionnaire (PSDQ) (Robinson et al., 1995). In the PSDQ, parents rate the degree to which statements are true about them on a 1 to 5 scale (i.e., I give my child comfort and understanding when he/she is upset). We used the 11-item Warmth and Involvement Dimension (Cronbach’s α=.72 in analyses (range 3 – 20).

2.4 Covariates

At baseline, we measured several potential covariates for inclusion in our regression models based on their previous associations with smoking during pregnancy and parenting responsiveness. These included maternal age, marital status, education, race and ethnicity, partner smoking, and hostility (Homish et al., 2012). As covariates of parenting behavior, we considered infant temperament (Clark et al., 2016), concurrent smoking, impulsivity (White et al., 2014), depressive symptoms (Lovejoy et al., 2000), antisocial behaviors (Ehrensaft et al., 2003), and family conflict (Cooklin et al., 2015). Specific measures used are shown in the legend under Table 1.

2.5 Data analysis

We assessed between-group differences among covariates using one-way ANOVA and two-sided Pearson χ2 tests for continuous and binary variables, respectively (Table 1). To discern the degree to which responsiveness differed as a function of successful smoking cessation versus continued smoking (hypothesis 1), we compared spontaneous quitters with intermittent and persistent smokers. To examine the association of responsiveness with smoking cessation in and outside the context of being pregnant (hypothesis 2), we compared spontaneous quitters with former smokers. Finally, to determine whether spontaneous quitters exhibited a level of responsiveness beyond the absence of risks associated with lifetime smoking (hypothesis 3), we compared spontaneous quitters with never smokers.

Covariates showing significant between-group differences (α < .05) were selected for inclusion in subsequent regression models. No evidence of multicollinearity, defined a priori, as r ≥ .60, was found among variables from bivariate correlations. Finally we conducted 3 separate linear regression models using the EC-HOME, PCOS, and PSDQ as dependent variables, respectively, using spontaneous quitters as the reference group, and the 4 other smoking groups entered as dummy variables.

3. Results

3.1 Maternal demographics

At study enrollment, mothers were primarily non-Hispanic Caucasian women (81.6%) in their mid-twenties (M = 25.9 ± 4.9 years) with one year of post-high-school education (M = 13.3 ± 2.0 years). Over half of women (58.0%) were unmarried, had other children at the time of the studied pregnancy (59.0%), and reported having a romantic partner who smoked (51.3%).

3.2 Maternal tobacco and other drug use

Cotinine-corrected MSDP, and cotinine concentrations from maternal and infant specimens (ranges = 1905.67 ng/ml and 81.67.0 ng/g, respectively) are shown in Table 1. One third (33.4%) of mothers (n=102) reported using alcohol during pregnancy and 22 (8.6%) out of the 281 infants with meconium data were positive for alcohol metabolites. Metabolites of marijuana and opioids were additionally detected from meconium in 11 (3.9%) and 7 infants (2.5%), respectively.

Among the 70 women categorized as intermittent pregnancy smokers, 36 (51.4%) had quit at enrollment but showed evidence of relapse at a later study visit. Three relapsed before 16 weeks (8.3%); 20 relapsed between 16 and 28 weeks (55.6%); and 13 relapsed close to delivery (36.1%). The remaining 34 intermittent smokers (48.6%) were smoking at enrollment. Of these, 31 (91.2%) were abstinent just before delivery (91.2%); 2 were abstinent just prior to the16 week visit (5.9%); and 1 was abstinent before the 28 week visit (2.9%). Among the 22 spontaneous quitters, 10 had resumed smoking (45.5%) by 4 weeks postpartum, 8 were still abstinent (36.4%), and the remaining 4 reported abstinence, but did not have cotinine data for confirmation.

3.3 Responsiveness

The PCOS showed small but significant correlations with the EC-HOME (r = .148, p = .019) and the PSDQ (r = .215, p = .001). The EC-HOME and PSDQ were not significantly correlated with one another (r = .093, p = .108). Concurrent smoking at the preschool follow-up was unrelated to scores of responsiveness from the EC-HOME (r=.035, p=.535); PCOS (r=.030, p=.647); and PSDQ (r=.045, p=.461).

In between-group comparisons of responsiveness, spontaneous quitters had higher EC-HOME scores relative to women who did not quit (i.e. intermittent and persistent pregnancy smokers) (4.5 ± 1.5 versus 3.6 ± 1.7; p = .011); former smokers (4.5 ± 1.5 versus 3.5 ± 2.0; p = .016); and never smokers (4.5 ± 1.5 versus 3.4 ± 2.0; p = .009). Responsiveness measured by the PCOS and PSDQ, maternal depressive symptoms, maternal impulsivity, and family conflict did not show hypothesized between-group differences (Table 1).

3.4 Covariates

Among covariates of prenatal smoking, maternal age (F(4, 299 =3.664; p=.006), unmarried status (F(4, 300)=4.390; p=.002), education (F(4, 300)=7.735; p< .001) and partner smoking (F(4, 299)=11.025; p<.001) showed between-group differences. At 4 weeks postpartum, spontaneous quitters were less likely to rate their infants’ temperament as ‘easy’ compared to former smokers (p=.025) and never smokers (p=.043); spontaneous quitters were also more likely to rate their infants as ‘slow to warm’ relative to never smokers (p=.028). At the preschool follow-up, we observed between-group differences in concurrent smoking (F(4, 270)=54.723; p<.001) and antisocial behavior (F(4,296)=10.450; p<.001).

3.5 Regression models of responsiveness

Table 2 shows estimated marginal means of responsiveness (EC-HOME) by prenatal smoking pattern (left), and contrast estimates (right) relative to spontaneous quitters (reference group) derived from linear regression. Spontaneous quitters were more responsive than pregnancy smokers, former smokers, and never smokers, controlling for maternal age, unmarried status, education, partner smoking, infant temperament, concurrent smoking, and maternal antisocial behavior. We did not observe differences in responsiveness measured by the PCOS or the PSDQ. Responsiveness measured by the PCOS was predicted, instead, by maternal educational attainment (β = .167, p = .025) while responsiveness measured by the PSDQ was predicted by easier infant temperament at 4 weeks postpartum (β = .142; p = .009).

Table 2.

Estimates of maternal responsiveness a by prenatal smoking pattern from linear regression

Responsiveness estimated marginal means (SE) 95% C.I. Difference from spontaneous quitters (SE) 95% C.I. for difference
Never smokers 3.390 (.257) 2.885 – 3.895 −1.351 (.513)** −2.362 – −.341
Former smokers 3.644 (.229) 3.193 – 4.095 −1.097 (.497)* −2.077 – −.118
Spontaneous quitters 4.741 (.427) 3.900 – 5.583 - -
Intermittent pregnancy smokers 3.640 (.228) 3.192 – 4.088 −1.101 (.479)* −2.044 – −.159
Persistent pregnancy smokers 3.688 (.316) 3.066 – 4.311 −1.053 (.511)* −2.059 – −.047
Open in a new tab

Covariates:

Prenatal- maternal age, unmarried status, maternal education, partner smoking

Postpartum- infant temperament (difficult/slow to warm vs. easy);

Preschool- maternal smoking and antisocial behavior

a

EC-HOME = Early Childhood Home Observation for Measurement of the Environment, Responsivity Subscale (range 0 – 7)

*

p < .05;

**

p<.01

4. Discussion

While pregnancy is associated with an unparalleled protective effect on women’s smoking (Solomon and Quinn, 2004) and other drug use (Kendler et al., 2017), the underlying mechanisms are unclear, but potentially valuable for identifying novel intervention targets. Building on our prior work aimed at characterizing a pregnancy-specific mechanism of smoking behavior change (Massey et al., 2017; Massey et al., 2015b; Massey and Wisner, In Press), we examined responsive parenting behavior, as a proxy measure of empathic capacity (Eisenberg and Miller, 1987), hypothesized to differentiate smokers who successfully quit in early pregnancy (spontaneous quitters) from women who did not engage in this protective action during pregnancy (either because they continued smoking, had quit earlier in life for reasons other than pregnancy, or had never smoked){Massey, 2017 #732}.

We found partial support for our hypotheses. When observed with their children at home (EC-HOME), spontaneous quitters compared favorably on responsiveness relative to other women including never smokers, which broadly echoes our previous findings about a protective developmental trajectory associated with quitting (Hutchinson et al., 2010; Pickett et al., 2008). However, exactly how quitters differ from other women may not be wholly captured by the construct of responsiveness, as operationalized here, since the PCOS and PSDQ did not show these differences. This was unexpected due to overlap among these measures. However, one aspect of parenting assessed by the PCOS and PSDQ, but not the EC-HOME, is how parents react to their child’s frustration. Thus, it is plausible that while spontaneous quitters displayed greater warmth with their children at home relative to other mothers, they did not differ in their ability to respond to their children’s distress, by their own reports, or elicited in the lab-setting.

Nonetheless, we carefully considered a number of alternative explanations for these results in light of the discrepancy. For example, spontaneous quitters could have represented outliers with unusually favorable characteristics. But this possibility was not supported by the data. Spontaneous quitters were hardly free from risk - they engaged in more antisocial behaviors (i.e. skipping school, legal problems, physical fights, being fired from a job for performance) relative to former smokers and never smokers. This heterogeneity in subsamples categorized based on prenatal smoking patterns was also evident in our previous work (in a different sample); pregnancy quitters were more likely to be in a stable romantic relationship compared to other groups, but were less likely to obtain adequate preventive health care (Wakschlag et al., 2003).

Another potential explanation for superior responsiveness observed in spontaneous quitters is an increase in self-efficacy resulting from quitting smoking (Mathieu et al., 1993) that was later observable as responsive parenting (Coleman and Karraker, 2003). However, if this were the case, former smokers would be expected to demonstrate comparable responsiveness to spontaneous quitters, but they did not. Summarily, responsive mother-child interactions at home differentiated mothers who had abruptly suspended smoking in early pregnancy and remained abstinent through delivery from other mothers in this sample, including mothers who never smoked. This finding, together with our earlier work regarding children of women who quit smoking during pregnancy (Hutchinson et al., 2010; Pickett et al., 2008), provide compelling support for the presence of protective factors, partially described by responsiveness, that could plausibly facilitate smoking cessation in the context of pregnancy, independent of nicotine dependence and psychosocial stressors. Confirmation of these initial findings using a prospective design, and specification of key malleable component(s) that facilitate or sustain behavior change could provide the empirical foundation for a pregnancy-specific conceptual model from which novel prenatal intervention strategies could be developed.

4.1 Limitations

Findings should be interpreted in the context of study limitations, especially the small sample of 22 spontaneous quitters, who represented 13.9% of the 158 women who were smoking at the time of their LMP. While this cessation rate is well-within ranges previously reported in pregnant women recruited from public obstetric clinics when biochemical verification is used (i.e. 11 – 28%) (Higgins and Solomon, 2016; Kendrick et al., 1995; Secker-Walker et al., 1998; Windsor et al., 1998), replication is recommended. Relatedly, our approach to characterizing spontaneous quitters, (i.e. zero cotinine values rather than a cutoff) could raise concern about miscategorization of ‘true’ spontaneous quitters who were exposed to partner smoking as intermittent smokers. We considered the possibility, for example, that the 22 spontaneous quitters all had non-smoking partners, and avoided this common source of exposure to secondhand smoke. But the data do not support this. The prevalence of partner smoking was comparable between spontaneous quitters and pregnancy smokers (p=.806), and higher in spontaneous quitters compared to former smokers (p=.055), and never smokers (p=.004), as would be expected (Table 1).

Furthermore, as a post hoc test of validity of our categorization scheme, we recategorized the intermittent smokers who had abstinent study visits later in pregnancy (n=34) into the quitters group, then re-examined between group-differences in responsiveness. Re-categorization had the effect of reducing the quitters’ EC-HOME score, and rendered previously-found differences between spontaneous quitters and other groups non-significant (p=.362). The PCOS and PSDQ still did not differentiate quitters from other groups, as seen before. This provided a measure of validity of the found relationship between quitting before 16 weeks and remaining abstinent through delivery and responsive mother-child interactions 5 years later. Finally, the use of other substances by pregnant smokers is documented in nationally-representative survey studies (Coleman-Cowger et al., 2017). Women who endorsed binge drinking or illicit drug use at enrollment were excluded from this study which could threaten generalizability. However, available meconium data showed that alcohol, cannabis, and opioid use in late pregnancy was not uncommon (Section 3.3).

5. Conclusions

Findings from this study add to the limited knowledge about mechanisms associated with pregnancy that facilitate the phenomenon of abrupt and successful suspension of smoking in the absence of treatment (Solomon and Quinn, 2004). Replication and elaboration of the proposed developmental framework via secondary analysis of pooled developmental cohorts to increase power is recommended. Ultimately, confirming exactly how pregnancy exerts a protective effect on smoking and substance use will require new prospective investigation(s) that oversample for and accurately identify spontaneous quitters, and assess putative protective mechanisms directly (Massey et al., 2017). This research could yield novel targets for interventions aimed at initiating and sustaining changes in smoking and other drug use (Massey and Wisner, In Press).

Highlights.

  • Most pregnant smokers who have not quit on their own do not quit with intervention.

  • Smokers who quit on their own early in pregnancy are understudied.

  • Spontaneous quitters may possess unmeasured protective factors beyond lower risks.

  • Quitters were more responsive to their children at home compared to other women.

  • Quitters were more responsive even compared to women who never smoked.

Acknowledgments

Role of Funding Sources

This work was supported by the National Institute on Drug Abuse (R01DA014661 to Dr. Espy, R01DA023653 to Drs. Espy and Wakschlag, and K23DA037913 to Dr. Massey). NIDA had no role in the study design, data collection, analysis or interpretation of the data, or the

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Benowitz NL. Cotinine as a biomarker of environmental tobacco smoke exposure. Epidemiologic reviews. 1996;18(2):188–204. doi: 10.1093/oxfordjournals.epirev.a017925. [DOI] [PubMed] [Google Scholar]
  2. Bornstein M, Tamis-LeMonda C. Mothers’ responsiveness in infancy and their toddlers’ attention span, symbolic play, and language comprehension: Specific predictive relations. Infant Behavior and Development. 1997;20:283–296. [Google Scholar]
  3. Bradley RH. Children’s home environments, health, behavior, and intervention efforts: a review using the HOME inventory as a marker measure. Genetic, social, and general psychology monographs. 1993 [PubMed] [Google Scholar]
  4. Bradstreet MP, Higgins ST, Heil SH, Badger GJ, Skelly JM, Lynch ME, Trayah MC. Social discounting and cigarette smoking during pregnancy. Journal of behavioral decision making. 2012;25(5):502–511. doi: 10.1002/bdm.750. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Chamberlain C, O’Mara-Eves A, Oliver S, Caird JR, Perlen SM, Eades SJ, Thomas J. The Cochrane Library. 2013. Psychosocial interventions for supporting women to stop smoking in pregnancy. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Clark C, Espy KA, Wakschlag LS. Developmental pathways from prenatal tobacco and stress exposure to behavioral disinhibition. Neurotoxicology and teratology. 2016;53:64–74. doi: 10.1016/j.ntt.2015.11.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Clark CA, Massey SH, Wiebe SA, Espy KA, Wakschlag LS. Does early maternal responsiveness buffer prenatal tobacco exposure effects on young children’s behavioral disinhibition? doi: 10.1017/S0954579418000706. In review. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Coleman-Cowger VH, Schauer GL, Peters EN. Marijuana and tobacco co-use among a nationally representative sample of US pregnant and non-pregnant women: 2005–2014 National Survey on Drug Use and Health findings. Drug and Alcohol Dependence. 2017 doi: 10.1016/j.drugalcdep.2017.03.025. [DOI] [PubMed] [Google Scholar]
  9. Coleman PK, Karraker KH. Maternal self-efficacy beliefs, competence in parenting, and toddlers’ behavior and developmental status. Infant Mental Health Journal. 2003;24(2):126–148. [Google Scholar]
  10. Cooklin A, Westrupp E, Strazdins L, Giallo R, Martin A, Nicholson J. Mothers’ work–family conflict and enrichment: associations with parenting quality and couple relationship. Child: care, health and development. 2015;41(2):266–277. doi: 10.1111/cch.12137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dempsey D, Jacob P, Benowitz NL. Accelerated metabolism of nicotine and cotinine in pregnant smokers. Journal of Pharmacology and Experimental Therapeutics. 2002;301(2):594–598. doi: 10.1124/jpet.301.2.594. [DOI] [PubMed] [Google Scholar]
  12. Dornelas EA, Sampson RA, Gray JF, Waters D, Thompson PD. A randomized controlled trial of smoking cessation counseling after myocardial infarction. Preventive medicine. 2000;30(4):261–268. doi: 10.1006/pmed.2000.0644. [DOI] [PubMed] [Google Scholar]
  13. Ehrensaft MK, Wasserman GA, Verdelli L, Greenwald S, Miller LS, Davies M. Maternal antisocial behavior, parenting practices, and behavior problems in boys at risk for antisocial behavior. Journal of Child and Family Studies. 2003;12(1):27–40. [Google Scholar]
  14. Eisenberg N, Miller PA. The relation of empathy to prosocial and related behaviors. Psychological bulletin. 1987;101(1):91. [PubMed] [Google Scholar]
  15. Flemming K, Graham H, Heirs M, Fox D, Sowden A. Smoking in pregnancy: a systematic review of qualitative research of women who commence pregnancy as smokers. Journal of advanced nursing. 2013;69(5):1023–1036. doi: 10.1111/jan.12066. [DOI] [PubMed] [Google Scholar]
  16. George L, Granath F, Johansson AL, Cnattingius S. Self-reported nicotine exposure and plasma levels of cotinine in early and late pregnancy. Acta obstetricia et gynecologica Scandinavica. 2006;85(11):1331–1337. doi: 10.1080/00016340600935433. [DOI] [PubMed] [Google Scholar]
  17. Higgins ST, Solomon LJ. Some Recent Developments on Financial Incentives for Smoking Cessation Among Pregnant and Newly Postpartum Women. Current addiction reports. 2016;3(1):9–18. doi: 10.1007/s40429-016-0092-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hill C, Maskowitz K, Danis B, Wakschlag L. Validation of a clinically sensitive, observational coding system for parenting behaviors: The Parenting Clinical Observation Schedule. Parenting: Science and Practice. 2008;8(2):153–185. doi: 10.1080/15295190802045469. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Homish GG, Eiden RD, Leonard KE, Kozlowski LT. Social–environmental factors related to prenatal smoking. Addictive behaviors. 2012;37(1):73–77. doi: 10.1016/j.addbeh.2011.09.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hutchinson J, Pickett KE, Green J, Wakschlag LS. Smoking in pregnancy and disruptive behaviour in 3-year-old boys and girls: an analysis of the UK Millennium Cohort Study. Journal of epidemiology and community health. 2010;64(01):82–88. doi: 10.1136/jech.2009.089334. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Johnson CD, Jones S, Paranjothy S. Reducing low birth weight: prioritizing action to address modifiable risk factors. Journal of Public Health. 2017;39(1):122–131. doi: 10.1093/pubmed/fdv212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kendler KS, Ohlsson H, Svikis DS, Sundquist K, Sundquist J. The Protective Effect of Pregnancy on Risk for Drug Abuse: A Population, Co-Relative, Co-Spouse, and Within-Individual Analysis. The American journal of psychiatry. 2017;174(10):954–962. doi: 10.1176/appi.ajp.2017.16091006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kendrick JS, Zahniser SC, Miller N, Salas N, Stine J, Gargiullo PM, Floyd RL, Spierto FW, Sexton M, Metzger RW. Integrating smoking cessation into routine public prenatal care: the Smoking Cessation in Pregnancy project. American Journal of Public Health. 1995;85(2):217–222. doi: 10.2105/ajph.85.2.217. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kim S. Overview of Cotinine Cutoff Values for Smoking Status Classification. Int J Environ Res Public Health. 2016;13(12) doi: 10.3390/ijerph13121236. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lovejoy MC, Graczyk PA, O’Hare E, Neuman G. Maternal depression and parenting behavior: A meta-analytic review. Clinical psychology review. 2000;20(5):561–592. doi: 10.1016/s0272-7358(98)00100-7. [DOI] [PubMed] [Google Scholar]
  26. Massey SH, Bublitz MH, Magee SR, Salisbury A, Niaura RS, Wakschlag LS, Stroud LR. Maternal–fetal attachment differentiates patterns of prenatal smoking and exposure. Addictive behaviors. 2015a;45:51–56. doi: 10.1016/j.addbeh.2015.01.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Massey SH, Decety J, Wisner KL, Wakschlag LS. Specification of change mechanisms in pregnant smokers for malleable target identification: A novel approach to a tenacious public health problem. Frontiers in Public Health. 2017;5:239. doi: 10.3389/fpubh.2017.00239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Massey SH, Estabrook R, O’Brien TC, Pine DS, Burns JL, Jacob S, Cook EH, Wakschlag LS. Preliminary evidence for the interaction of the oxytocin receptor gene (oxtr) and face processing in differentiating prenatal smoking patterns. Neuroscience letters. 2015b;584:259–264. doi: 10.1016/j.neulet.2014.10.049. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Massey SH, Lieberman DZ, Reiss D, Leve LD, Shaw DS, Neiderhiser JM. Association of clinical characteristics and cessation of tobacco, alcohol, and illicit drug use during pregnancy. The American journal on addictions. 2011;20(2):143–150. doi: 10.1111/j.1521-0391.2010.00110.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Massey SH, Neiderhiser JM, Shaw DS, Leve LD, Ganiban JM, Reiss D. Maternal self concept as a provider and cessation of substance use during pregnancy. Addictive behaviors. 2012;37(8):956–961. doi: 10.1016/j.addbeh.2012.04.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Massey SH, Reiss D, Neiderhiser JM, Leve LD, Shaw DS, Ganiban JM. Maternal personality traits associated with patterns of prenatal smoking and exposure: Implications for etiologic and prevention research. Neurotoxicology and teratology. 2016;53:48–54. doi: 10.1016/j.ntt.2015.11.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Massey SH, Wisner KL. Understanding Pregnancy’s Protective Effect on Drug Use Within a Developmental Framework. The American journal of psychiatry. doi: 10.1176/appi.ajp.2017.17101099. In Press. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Mathieu JE, Martineau JW, Tannenbaum SI. Individual and situational influences on the development of self-efficacy: implications for training effectiveness. Personnel Psychology. 1993;46(1):125–147. [Google Scholar]
  34. Maxson PJ, Edwards SE, Ingram A, Miranda ML. Psychosocial differences between smokers and non-smokers during pregnancy. Addictive behaviors. 2012;37(2):153–159. doi: 10.1016/j.addbeh.2011.08.011. [DOI] [PubMed] [Google Scholar]
  35. Pickett K, Wood C, Adamson J, DeSouza L, Wakschlag L. Meaningful differences in maternal smoking behaviour during pregnancy: implications for infant behavioural vulnerability. Journal of Epidemiology and Community Health. 2008;62(4):318–324. doi: 10.1136/jech.2006.058768. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Robinson CC, Mandleco B, Olsen SF, Hart CH. Authoritative, authoritarian, and permissive parenting practices: Development of a new measure. Psychological Reports. 1995;77(3):819–830. [Google Scholar]
  37. Secker-Walker RH, Solomon LJ, Flynn BS, Skelly JM, Mead PB. Smoking relapse prevention during pregnancy: A trial of coordinated advice from physicians and individual counseling. American journal of preventive medicine. 1998;15(1):25–31. doi: 10.1016/s0749-3797(98)00029-4. [DOI] [PubMed] [Google Scholar]
  38. Sobell LC, Sobell MB. Timeline Followback user’s guide: A calendar method for assessing alcohol and drug use. Toronto: Addiction Research Foundation; 1996. [Google Scholar]
  39. Solomon LJ, Quinn VP. Spontaneous quitting: self-initiated smoking cessation in early pregnancy. Nicotine & Tobacco Research. 2004;6(Suppl_2):S203–S216. doi: 10.1080/14622200410001669132. [DOI] [PubMed] [Google Scholar]
  40. Swain JE, Konrath S, Brown SL, Finegood ED, Akce LB, Dayton CJ, Ho SS. Parenting and beyond: Common neurocircuits underlying parental and altruistic caregiving. Parenting. 2012;12(2–3):115–123. doi: 10.1080/15295192.2012.680409. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Totsika V, Sylva K. The home observation for measurement of the environment revisited. Child and Adolescent Mental Health. 2004;9(1):25–35. doi: 10.1046/j.1475-357X.2003.00073.x. [DOI] [PubMed] [Google Scholar]
  42. Wakschlag LS, Hans SL. Maternal smoking during pregnancy and conduct problems in high-risk youth: a developmental framework. Development and psychopathology. 2002;14(2):351–369. doi: 10.1017/s0954579402002092. [DOI] [PubMed] [Google Scholar]
  43. Wakschlag LS, Henry DB, Blair RJR, Dukic V, Burns J, Pickett KE. Unpacking the association: Individual differences in the relation of prenatal exposure to cigarettes and disruptive behavior phenotypes. Neurotoxicology and Teratology. 2011;33(1):145–154. doi: 10.1016/j.ntt.2010.07.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Wakschlag LS, Pickett KE, Middlecamp MK, Walton LL, Tenzer P, Leventhal BL. Pregnant smokers who quit, pregnant smokers who don’t: Does history of problem behavior make a difference? Social science & medicine. 2003;56(12):2449–2460. doi: 10.1016/s0277-9536(02)00248-4. [DOI] [PubMed] [Google Scholar]
  45. Wang X, Tager IB, Van Vunakis H, Speizer FE, Hanrahan JP. Maternal smoking during pregnancy, urine cotinine concentrations, and birth outcomes. A prospective cohort study. International Journal of Epidemiology. 1997;26(5):978–988. doi: 10.1093/ije/26.5.978. [DOI] [PubMed] [Google Scholar]
  46. White TJ, Redner R, Skelly JM, Higgins ST. Examining educational attainment, pre-pregnancy smoking rate, and delay discounting as predictors of spontaneous quitting among pregnant smokers. Experimental and clinical psychopharmacology. 2014;22(5):384. doi: 10.1037/a0037492. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Windsor RA, Boyd NR, Orleans CT. A meta-evaluation of smoking cessation intervention research among pregnant women: improving the science and art. Health Education Research. 1998;13(3):419–438. doi: 10.1093/her/13.3.419. [DOI] [PubMed] [Google Scholar]

RESOURCES