Maternal-Fetal Attachment Differentiates Patterns of Prenatal Smoking and Exposure

Abstract

Objective

Smoking cessation during pregnancy may reflect altruistic motives on behalf of the unborn baby. We test the hypothesis that pregnancy quitters have higher maternal-fetal attachment than persistent smokers, and secondarily explore how maternal-fetal attachment differs among non-smokers, pregnancy quitters, and persistent smokers.

Methods

Participants were 156 women in the Behavior and Mood in Babies and Mothers study who provided report of smoking throughout pregnancy via timeline follow back interviews, with salivary cotinine confirmation of reported cessation at 30 and 35 weeks gestation, and postpartum day one. Maternal Fetal Attachment Scale total and subscale scores (role-taking, differentiation of self from fetus, interaction with fetus, attributing characteristics to fetus, giving of self) were examined among non-smokers, pregnancy quitters, and persistent smokers.

Results

At 30 weeks, pregnancy quitters scored higher on the ‘giving of self’ subscale compared to persistent smokers (21.6 ± 2.4 versus 19.9 ± 2.9; p = .004). Maternal ‘giving of self’ also differentiated pregnancies exposed to cigarette smoking from those without exposure from 30 weeks through delivery (19.9 ± 2.9 versus 21.2 ± 2.2; p = .002). Controlling for age, income, unemployment, gravida, and father’s smoking status, ‘giving of self’ differentiated pregnancy quitters from persistent smokers [OR = 5.144; 95% C.I. 1.509 – 17.538; B (SE) = 1.638 (.626); p = .009].

Conclusions

Women who reported a greater desire to maintain their personal health for the health of their fetus were more likely to quit smoking during pregnancy. Implications of findings for interventions and understanding mechanisms of risk are discussed.

1. Introduction

Maternal smoking during pregnancy remains one of the most common preventable risk factors for poor pregnancy and birth outcomes in the United States and other industrialized nations (Cnattingius 2004; Floyd et al., 2007) and has been associated with an estimated $366 million per year in neonatal healthcare costs in the United States (Adams, Miller, Ernst, Nishimura, Melvin, & Merritt, 2002). Children born to mothers who smoked during pregnancy are at an increased risk of developing a broad range of problems including sudden infant death syndrome (Schoendorf & Kiely, 1992), asthma (Gilliland, Li, & Peters, 2001), neurological handicaps (Paula & Koiranen, 1987), disruptive behavior disorders (Wakschlag et al., 1997; Wakschlag, Pickett, Cook, Benowitz, & Leventhal, 2002), and substance abuse and criminality in adulthood (Brennan, Grekin, Mortensen, & Mednick, 2002). Despite the increasing evidence and public awareness of the short and long-term consequences associated with smoking during pregnancy, the majority of women who have not quit smoking by their first prenatal visit will continue smoking throughout pregnancy (Lumley, 2009).

A significant barrier to improving prenatal smoking cessation rates may be an inadequate understanding of why women make health behavior changes during pregnancy (Stotts et al., 1996). To address prenatal smoking, the American College of Obstetrics and Gynecology recommends the 5 A’s: (Ask, Assess, Advise, Assist and Arrange), derived from the Transtheoretical Stages of Change Model (Prochaska & DiClemente, 1983; Prochaska, Velicer, 1997) based on how non-pregnant adults (largely men) initiate and maintain health behavior changes (Melvin, Dolan-Mullen, Windsor, Whiteside & Goldenberg, 2000). This standard intervention relies on pregnant smokers to weigh pros and cons of continued smoking, then utilize behavioral strategies to remain abstinent from cigarette smoking through gestation. Unfortunately, less than 10% of smokers who receive these interventions achieve and maintain abstinence from cigarette smoking during pregnancy (Ershoff, Quinn, Boyd, Stern, Gregory & Wirtschafter, 2000; Lumley, 2009). Other more intensive interventions, including the use of fetal ultrasound visualization, motivational interviewing techniques (also based on the transtheorhetical model), and financial incentives for confirmed abstinence, are also ineffective for the majority of pregnant smokers (Stotts et al., 2009; Marteau, Thorne, Aveyard, Hirst & Sokal, 2013; Hayes, et al., 2013). To substantively reduce prenatal nicotine exposure, there is a great need for a more accurate understanding of the psychological processes involved in smoking behavior change during pregnancy (Massey & Compton, 2013).

Several key observations suggest that psychological and behavioral processes involved in smoking cessation during pregnancy are fundamentally different from processes involved in smoking cessation at other times in life, and thus are not accurately described by the Stages of Change Model. First, women who successfully quit smoking during pregnancy report greater confidence about sustaining abstinence during gestation compared to non-pregnant smokers, yet do not use the types of behavioral strategies (i.e. coping skills, distractions) used by their non-pregnant counterparts (Ruggiero, Tsoh, Everett, Fava & Guise 2000). Second, while non-pregnant ex-smokers are most vulnerable to relapse in the first weeks after quitting, when symptoms of nicotine dependence are most prominent (Killen & Fortmann, 1994), women who successfully quit smoking during pregnancy appear to abstain over many months of gestation, only to relapse shortly after delivery, long after symptoms of nicotine withdrawal have subsided (Stotts, DiClemente, Carbonari, & Mullen, 1996; Buja et al., 2011).

What, then, drives women’s motivation to suspend this addictive behavior during pregnancy? Women who view themselves as adequate providers for others may be more likely to suspend the use of tobacco, alcohol and illicit drugs during pregnancy; the importance of this self concept may operate independent of socioeconomic differences, family history of substance abuse, and depressive symptoms (Massey, Neiderhiser, Shaw, Leve, Ganiban, & Reiss, 2012). Building on this early evidence, to develop a more accurate understanding of how and why women may be motivated to change smoking behavior during pregnancy, we aimed to provide empirical evidence for the conceptualization of prenatal smoking cessation as a maternal action taken on behalf of the fetus, rather than a personal health decision.

Maternal-fetal attachment constitutes a mother-to-be seeking to know, be with, avoid separation and loss of, protect, and identify and gratify the needs of the fetus (Condon, 1993). Within this broader construct, cognitive attachment has been described as the desire to know the fetus; affective attachment as the pleasure associated with thoughts of, or interaction with the fetus; and altruistic attachment as a desire to protect the unborn child (Shieh, Kravitz, and Wang, 2001). Low maternal-fetal attachment has been linked to poorer health practices during pregnancy (Lindgren 2001) and we have previously shown lower maternal-fetal attachment to be associated with higher levels of smoking among women who smoke persistently throughout pregnancy (Magee et al., 2014). Maternal-fetal attachment also appears to be higher among pregnant women who are planning to quit smoking in the future (Slade, Laxton-Kane & Spiby, 2006). However, whether greater attachment is related to successful smoking cessation during pregnancy is not known.

In this study, we tested the hypothesis that women who quit smoking during pregnancy (pregnancy quitters) would report higher maternal-fetal attachment relative to women who do not quit (persistent smokers). If confirmed, increasing maternal-fetal attachment during pregnancy may improve chances of smoking cessation, thereby optimizing the short and long term health of both mothers and their children (Akai et al., 2008; Suchman, DeCoste, Castiglioni, Legow & Mayes, 2008). To elucidate how modifiable maternal psychological factors related to prenatal smoking exposure may influence child outcomes, we secondarily explored how attachment may differ between non-smokers and pregnancy quitters, and by intrauterine smoking exposure status (non-smokers and pregnancy quitters versus persistent smokers).

2. Methods

2.1 Sample

Participants were 156 women recruited from prenatal clinics in the Northeastern United States for the Behavior and Mood in Babies and Mothers (BAM BAM) study, a prospective study whose primary aim is to identify the effects of maternal smoking on fetal and infant development. Women were excluded from participation in BAM BAM if they were less than 18 or over 40 years old, were pregnant with more than one fetus, or were at high risk for adverse neonatal outcomes. Women were primarily low-income, unmarried women from diverse racial and ethnic backgrounds (average annual income $20,000 – 29,999; 57% unemployed; 74% unmarried; 19% African-American, 24% Hispanic, 47% non-Hispanic Caucasian, 3% Asian-American, 2% Native American, 4% more than one race or ethnicity, 1% other/not known). The majority of pregnancies (77%) were unplanned.

2.2 Procedures

Women were enrolled in the study at an average of 30 weeks gestation (SD = 2 weeks) and were interviewed again in the third trimester (mean = 35 weeks gestation, SD = 1 week), and following delivery (mean = 1 day postpartum, SD = 1 day). Interviews were conducted in private rooms by trained interviewers. At the initial assessment, women reported on demographic and pregnancy-related information and were assessed for maternal-fetal attachment (see 2.3.1) and nicotine dependence (Fagerstrom & Schneider, 1989). At every session (baseline, third trimester, delivery) participants reported on smoking and provided saliva samples for cotinine measurement. All study procedures were approved by the Institutional Review Boards of the Lifespan and Women & Infants Hospitals.

2.3 Measures

2.3.1 Maternal-Fetal Attachment

Participants completed the Maternal Fetal Attachment Scale, a 24-item self-report measure of emotional connection to the fetus during pregnancy (Cranley, 1981) at the initial visit. The five subscales reflecting different facets of maternal-fetal attachment and sample items, respectively, are: role-taking (I picture myself feeding my baby); differentiation of self from fetus (I enjoyed watching my tummy jiggle as the baby kicked inside); interaction with fetus (I talked to my unborn baby); attributing characteristics to fetus (I worried if my baby felt cramped in there); and giving of self (I did things to try and stay healthy that I would not do if I had not been pregnant). Women rated the degree to which statements were true or false for them on a 5 point scale (5 = definitely yes to 1 = definitely no). The scale has previously demonstrated good reliability (Cranley, 1981).

2.3.2 Cigarette Smoking During Pregnancy

Women reported their cigarette use during each week of pregnancy using the Time Line Follow Back (TLFB) Interview (Sobell & Sobell 1986), a semi-structured interview that utilizes a calendar to aid in participants’ recall of smoking during pregnancy. The TLFB method is the preferred approach for measuring substance use retrospectively (Savage, Wray, Ritchey Sommers, Dyehouse, & Fulmer, 2003), and has shown good reliability in pregnant samples (Dum, Sobell, L., Sobell, M., Heinecke, Voluse, & Johnson, 2009). Participants also provided saliva samples (passive drool) at baseline, third trimester, and delivery to measure cotinine, a stable nicotine metabolite with a half-life of 10 hours produced by the liver (Benowitz, 1999). After collection, saliva samples were frozen at −80°C and shipped to Salimetrics (State College, PA, USA) for biochemical analysis using enzyme-linked immuno-sorbent assay (ELISA) kits, with a sensitivity of 0.15 mg/mL (Salimetrics, 2006).

2.4 Theory/calculation

Participants were categorized into three different smoking groups based on TLFB interviews and confirmed by salivary cotinine concentrations: non-smokers (never smoked); pregnancy quitters (smoked prior to conception, but quit by 30 weeks of gestation through delivery); and persistent smokers (smoked throughout gestation). Individuals whose salivary cotinine level suggested continued smoking despite reports of quitting were re-categorized accordingly. Demographic and pregnancy characteristics, and nicotine dependence were compared among non-smokers, pregnancy quitters and persistent smokers using one-way analysis of variance (ANOVA) and chi square tests for continuous and binary variables, respectively. Total and subscale scores on the Maternal Fetal Attachment Scale were also compared among non-smokers, pregnancy quitters, and persistent smokers using one-way ANOVA.

Secondary comparisons using t-tests were conducted to further explore differences among smoking groups. To specifically examine how maternal fetal attachment was related to smoking cessation during pregnancy, a comparison between pregnancy quitters and persistent smokers (excluding non-smokers) was conducted. Next, to explore how maternal-fetal attachment may differ based on intrauterine exposure to cigarette smoking (from 30 weeks through delivery), we compared non-smokers and pregnancy quitters (no exposure) with persistent smokers (exposure), and also non-smokers and pregnancy quitters. Finally, the independent relationship between maternal-fetal attachment and prenatal smoking cessation was assessed using logistic regression, using pregnancy quitters versus persistent smokers as the dependent variable. Demographic and pregnancy characteristics that differentiated pregnancy quitters and persistent smokers in descriptive analyses were entered into the regression model as covariates.

3. Results

Sixty-one women (39%) of participants were non-smokers; 32 women (21%) quit smoking by 30 weeks (26 quit in the first trimester, 6 quit in the second trimester); and 63 women (40%) smoked persistently throughout pregnancy. Four women who reported quitting smoking (11% of reported quitters) were re-categorized as persistent smokers based on salivary cotinine levels. Maternal characteristics by smoking group (non-smokers, pregnancy quitters, and persistent smokers) are shown in Table 1. Significant differences in annual income, employment status, number of previous pregnancies, breastfeeding, depression symptoms, anxiety symptoms, father smoking, and nicotine dependence scores were observed among groups. Pregnancy quitters in this sample were younger and more likely to be unmarried relative to non-smokers and persistent smokers. Compared to persistent smokers, pregnancy quitters were younger, had higher incomes and fewer previous pregnancies, were more likely to be employed and have a non-smoking partners, and had lower nicotine dependence.

Table 1.

Maternal characteristics by biochemically-verified prenatal smoking patterns^a

	Non-smokers (n = 61)	Pregnancy quitters (n = 32)	Persistent smokers (n = 63)	All groups	Non-smokers excluded
				p
Age in years, mean (SD)	25.8 (5.8)	22.9 (4.3)	25.3 (4.4)	.027	.015
Non-Hispanic Caucasian (%)	42.6	50.0	49.2	.446	.942
Annual income, mean (SD) b	5.5 (2.1)	3.9 (2.1)	3.0 (1.8)	< .001	.046
High school graduate (%)	68.2	41.9	23.8	< .001	.071
Married (%)	44.4	12.5	16.4	< .001	.618
Unemployed status (%)	39.3	53.1	76.2	< .001	.022
Gravida, mean (SD)	2.2 (1.5)	2.3 (1.5)	3.2 (2.1)	.007	.015
Parity, mean (SD)	0.8 (1.0)	0.8 (1.5)	1.4 (1.5)	.053	.095
Unplanned pregnancy (%)	66.7	83.9	82.3	.082	.846
Considered adoption/abortion (%)	13.0	16.1	27.4	.128	.227
Breastfeeding (% yes)	30.4	24.0	21.2	< .001	.134
Alcoholic drinks per week	0.0 (0.1)	0.0 (0.1)	0.3 (0.1)	.357	.126
Depression symptoms c	2.3 (3.0)	5.2 (4.4)	5.5 (5.4)	.001	.804
Anxiety symptoms d	2.4 (3.1)	6.0 (4.3)	5.4 (4.9)	< .001	.505
Father smokes (%)	17.0	45.2	72.1	< .001	.011
Nicotine dependence score e	0.0 (0.0)	1.0 (1.2)	3.2 (2.1)	< .001	.001

^aSelf-reported cigarettes per day across pregnancy, adjusted based on salivary cotinine assessed at 30 and 35 weeks gestation and 1 day postpartum

^b1 = < $5,000; 2 = $5,000 – $9,999; 3 = &10,000 – $19,999; 4 = $20,000 – $29,999; 5 = $30,000 – $39,999; 6 = $40,000 – $49,999; 7 = $50,000 – $99,999; 8 = > $100,000

^cHamilton Depression Scale (HAM-D)

^dHamilton Anxiety Scale (HAM-A)

^eFagerstrom Test of Nicotine Dependence

As shown in Table 2, the ‘giving of self’ subscale differentiated women by prenatal smoking groups (p = .006). Pregnancy quitters had higher scores on the ‘giving of self’ subscale, compared to persistent smokers (21.6 ± 2.4 versus 19.9 ± 2.9; p = .004), but did not differ from non-smokers in this regard (p = .242). The ‘giving of self’ subscale also differentiated women by intrauterine smoking exposure (30 weeks through delivery) status; exposure was associated with a lower ‘giving of self’ score (p = .002). Total maternal-fetal attachment scores and other subscale scores were not significantly different among smoking groups. Regression coefficients of psychosocial factors at 30 weeks associated with prenatal smoking cessation through delivery are shown in Table 3. The ‘giving of self’ subscale score was independently associated with a five-fold increase in the odds of smoking cessation during pregnancy [OR = 5.144; 95% C.I. 1.509 – 17.538; B (SE) = 1.638 (.626); p = .009]. Other predictors of smoking cessation were younger age, employment, and absence of father smoking. Items in the ‘giving of self’ subscale are listed in Table 4.

Table 2.

Maternal fetal attachment (30 weeks) by biochemically–verified smoking patterns ^a (N = 156)

	Non-smokers (n = 61)	Pregnancy quitters (n = 32)	Persistent smokers (n = 63)	Within all groups	Pregnancy quitters vs. persistent smokers	Non-smokers vs. pregnancy quitters	Exposed vs. non-exposed c
	Mean (standard deviation)			p
Total score b	100.9 (11.1)	103.6 (9.3)	102.4 (11.8)	.518	.590	.222	.760
Role-taking	22.8 (2.7)	23.5 (2.1)	23.0 (2.6)	.410	.272	.147	.766
Differentiation	21.1 (3.3)	21.8 (2.4)	21.1 (2.9)	.525	.236	.254	.880
Interaction	17.1 (3.5)	17.5 (2.7)	18.3 (3.6)	.143	.255	.493	.662
Attribute characteristics	19.6 (3.3)	19.2 (3.4)	20.2 (3.5)	.432	.222	.582	.060
Giving of self	21.0 (2.2)	21.6 (2.4)	19.9 (2.9)	.006	.004	.242	.002

^aSelf-reported cigarettes per day across pregnancy, adjusted based on salivary cotinine assessed at approximately 30 and 35 weeks gestation and 1 day postpartum

^bMaternal-Fetal Attachment Scale (subscales listed below) assessed at approximately 30 weeks gestation

^cNon-exposed to intrauterine cigarette smoke from 30 weeks through delivery; non-smokers and pregnancy quitters versus persistent smokers

Table 3.

Regression model of psychosocial factors at 30 weeks associated with prenatal smoking cessation^a

	B	SE	Wald	df	sig	Exp(B)	95% C.I. for Exp(B)
Age	−.231	.089	6.741	1	.009	.794	1.058 – 1.500
Income	.178	.145	1.510	1	.219	1.194	.631 – 1.112
Umemployed	−1.725	.680	6.436	1	.011	.178	.047 – .676
Gravida	.034	.203	.028	1	.867	1.035	.696 – 1.539
Father smokes	−1.203	.596	4.069	1	.044	.300	.093 – .967
Giving of self b	1.638	.626	6.850	1	.009	5.144	1.509 – 17.538

^aSelf-reported cigarettes per day across pregnancy, adjusted based on repeated salivary cotinine levels assessed at 30 and 35 weeks gestation and 1 day postpartum

^bGiving of Self subscale of the Maternal Fetal Attachment Scale

Table 4.

‘Giving of self’ subscale from the Maternal Fetal Attachment Scale (Cranley, 1981).

I felt all the trouble of being pregnant was worth it. I did things to try to stay healthy that I would not do if I had not been pregnant. I ate meat and vegetables to be sure my baby got a good diet. I felt my body was ugly (reverse scored). I gave up doing certain things because I wanted to help my baby.

4. Discussion

We recently reported that among persistent pregnancy smokers, lower maternal-fetal attachment was associated with a greater number of cigarettes smoked per day across pregnancy (Magee, 2014). In this study, we extend findings by demonstrating significant differences in attachment among nonsmokers, smokers who successfully quit smoking during pregnancy (pregnancy quitters) and smokers who did not quit (persistent smokers). Pregnancy quitters reported making health related changes because they “wanted to help the baby” to a greater extent relative to persistent smokers. In fact, these attributes, as embodied by the ‘giving of self’ subscale, were independently associated with biochemically-confirmed abstinence from cigarette smoking from 30 weeks of gestation through delivery. Other facets of maternal-fetal attachment, namely role taking, differentiation of self from fetus, interaction with the fetus, and attribution of characteristics to the fetus, did not differ significantly between pregnancy quitters and persistent smokers. Thus, altruistic attachment, but not affective or cognitive attachment, may be central to motivation for smoking cessation during pregnancy. This provides the first empirical data to our knowledge to support smoking cessation during pregnancy as an action taken on behalf of fetal well-being, rather than for personal health reasons. There are two important implications of findings, if confirmed through replication in larger samples.

4.1 Enhancing empathy in pregnant smokers may improve chances of cessation

First, the tendency to view health related behavior changes as worthwhile sacrifices on behalf of the unborn child, which may be broadly conceptualized as a tendency toward empathic behavior, may represent a viable target for interventions aimed at reducing prenatal nicotine and other substance exposures. Substantial research has documented the efficacy of brief behavioral interventions for increasing empathic behaviors in diverse samples including among romantic partners (Block-Lerner, Adair, Plumb, Rhatigan & Orsillo, 2007), health care providers (Heuberger, 2010), child welfare workers (Mullins, 2011), male parolees (Marlow, Nyamanthi, Grajeda, Bailey, Weber, & Younger, 2012), and prison inmates (Palusci, Crum, Bliss & Bavolek, 2008). Furthermore, brief training interventions aimed at increasing parents’ empathy towards their children have been shown to reduce harmful parenting practices and improve parent-child relationships (Akai et al., 2008; Stolk, et al., 2008; Suchman et al., 2008; Edwards, Sullivan, Meany-Walen & Kantor, 2010; Kendrick, Mulvaney, Stevens, Mytton & Stewart-Brown, 2013). Thus, the application of targeted interventions that increase empathy for fetal well-being among smokers during pregnancy may increase their chances of quitting.

The difference in the ‘giving of self’ subscale score between pregnancy quitters and persistent smokers was statistically significant, but small (21.4 ± 2.4 versus 19.9 ± 2.5, p = .016). Thus, even a slight change in the capacity to act on behalf of the fetus may effect a meaningful change in smoking behavior during pregnancy. It is also notable that in this sample, the ‘giving of self’ subscale differentiated pregnancy quitters from persistent smokers, while symptoms of depression and anxiety did not (Table 1). This suggests that altruistic tendencies embodied by the ‘giving of self’ subscale may promote smoking cessation during pregnancy even in women experiencing depression and anxiety symptoms. We considered that an aversion to the taste and smell of cigarettes during gestation may also drive motivation to quit smoking during pregnancy (Pletsch & Thornton, 2004). However, in this sample, women who successfully quit smoking reported giving up “doing certain things because I wanted to help my baby” to a greater extent than women who did not quit (Table 4). Thus, this altruistic sentiment is likely to be an independent motivating factor in quitting smoking during pregnancy.

4.2 Differences in maternal attachment may extend to parenting behavior

Our findings are also meaningful in the context of understanding the complex mechanisms linking prenatal smoking with child psychopathology (Knopik, 2009). In this study, compared to women who smoked throughout pregnancy, women who quit reported to a greater extent that “the trouble of pregnancy was worth it,” and reported engaging in behaviors they “would not have, had they not been pregnant.” Moreover, the ‘giving of self’ subscale differentiated pregnancies with intrauterine smoking exposure (persistent smokers) from pregnancies without exposure from 30 weeks through delivery (non-smokers and pregnancy quitters) (Table 2). There was also no difference in ‘giving of self’ between non-smokers and pregnancy quitters. This suggests that compared to non-smokers and pregnancy quitters, women who continue smoking throughout gestation may feel more ambivalent about their pregnancy and the changes it brings to their bodies (“I felt my body was ugly”), and may be less ready to respond to the transition to motherhood by making behavioral changes (Table 4). Furthermore, as maternal attachment to the fetus during pregnancy is predictive of maternal-child attachment postpartum (Dubber, Reck, Müller, & Gawlik, 2014), children of persistent pregnancy smokers are not only exposed to cigarette smoking during gestation, but may also be raised by mothers who may feel less emotionally attached to them, constituting an important mechanism through which prenatal smoking may confer risk to exposed children beyond nicotine exposure (Alhusen, Hayat, & Gross, 2013). If our findings are confirmed in larger samples with more comprehensive measures of attachment, prenatal interventions that enhance maternal fetal attachment may not only reduce prenatal nicotine exposure, but may also improve important parenting factors that affect child development postpartum.

Strengths of this study include the examination of a potentially modifiable, yet understudied construct in relation to prenatal smoking patterns using repeated biochemical confirmation of self-reported smoking, known to be affected by the stigma of continued smoking during pregnancy (Dietz et al., 2011). It is important to note that above interpretations and potential implications discussed rely on the assumption that greater maternal fetal attachment during pregnancy influenced women’s motivation to quit smoking and remain abstinent through delivery. The observational study design and timing of attachment assessments do not support a model of causality. Other limitations include the relatively small sample size and lack of biochemical verification of smoking prior to study enrollment. The ideal study would include the measurement of attachment prior to conception and an observational measure of maternal-infant attachment after delivery.

5. Conclusions

We found that compared to women who smoked throughout pregnancy, women who quit smoking reported making health behavior changes they would not have made during other times in life to help their unborn babies, providing empirical support for the conceptualization of prenatal smoking cessation as an empathic action taken on behalf of fetal well-being. Secondary analyses suggest that maternal-fetal attachment may constitute an important mechanism through which intrauterine smoking exposure influences child behavioral outcomes. More research is needed to understand how enhancing maternal empathy and attachment during pregnancy may positively affect change in addictive behaviors.

Highlights.

• Smoking cessation during pregnancy may reflect maternal altruism towards the fetus.

• We examined maternal-fetal attachment and biochemically verified smoking cessation.

• Pregnancy quitters had higher altruistic attachment compared to persistent smokers.

• Attachment, empathy and prenatal smoking behavior change should be further studied.