Postpartum smoking relapse and becoming a mother

Originally Published:20060101.

The incidence of smoking during pregnancy in the US has decreased precipitously in the past decade. Smoking cessation interventions for pregnant women have been successful in reducing tobacco-related health risks, including preterm birth, premature rupture of membranes, abruptio placentae, placenta previa, stillbirth, and neonatal deaths (U.S. Department of Health and Human Services, 2001). However, despite favorable outcomes in smoking cessation during pregnancy, recent evidence indicates that most women who smoked before pregnancy and quit during pregnancy returned to smoking during the 1st year postpartum (Colman, &c Joyce, 2003; Mullen, Richardson, Quinn, & Ershoff, 1997).

The significance of a return to smoking by this subpopulation of women is its effect on both maternal and infant health. The risk to women's health posed by smoking, including chronic cardiovascular and respiratory problems, is well-established. Evidence of the untoward effects of maternal tobacco use on infant health is also increasing. Recent reviews indicate that the health risks for infants posed by having a mother who smokes are many, including otitis media, exacerbations of asthma, respiratory infections, and gastrointestinal dysregulation such as colic and acid reflux (Gaffney, 2000, 2001; Shenassa & Brown, 2004).

To reduce the untoward health effects of tobacco use on mothers and their infants, tailored interventions are needed to support women who quit smoking during pregnancy and wish to remain smoke-free after delivery. The purpose of this paper is to propose a new, theoretically-based conceptual framework for postpartum smoking research that blends the context-specific factors associated with becoming a mother with existing designs for examining smoking relapse prevention in the general population. These context-specific factors consist of the developmental experiences in a woman's transformation of self as she incorporates a new identity and assumes responsibility for her infant and her infant's future welfare (Mercer, 2004).

Preventing Smoking Relapse

The theoretical work of Marlatt and his colleagues has provided a useful foundation for studies to focus on events associated with smoking behavior after an initial period of smoking cessation (Marlatt, 1985, 1996; Marlatt & Gordon, 1985). Three underlying principles from this theoretical work have application to all subpopulations of smokers, including postpartum women. First, high-risk situations serve as triggers to a return to smoking. Examples of situations that have been found to be triggers for the general population of smokers include living with a smoker, being with friends at a party, and feeling anger toward someone or something.

Second, an individual's capacity to cope with high-risk situations directly relates to his or her smoking abstinence self-efficacy. People with effective coping responses to high-risk situations have confidence that they can maintain abstinence. This confidence is referred to as smoking abstinence self-efficacy. Conversely, people with ineffective coping responses to high-risk situations experience decreased self-efficacy that can contribute to a single episode of smoking (a lapse). The initial lapse, in turn, leads to emotional responses that increase the probability of extended smoking, including both further lapses and eventually a relapse (a complete return to smoking on a regular basis).

A third principle from Marlatt's smoking relapse prevention research is that effective interventions are those that are tailored to individual smokers' specific high-risk situations. Identification of a person's high-risk situations is necessary for understanding the abstaining smoker's response behaviors and planning interventions to promote maintenance behaviors. Failing to identify context-specific risk factors results in less targeted and less effective interventions (Gwaltney et al, 2001).

Effect of Current Interventions

To date, postpartum smoking relapse prevention interventions for mothers who were smokers before pregnancy have had limited success. For example, Mullen and colleagues (1997) tested the effectiveness of a prenatal self-help smoking cessation program on postpartum tobacco use and found that 63% (83/133) of the mothers in their sample had returned to smoking by the time their infants were 6 months old. The average time to return to smoking was 111 days after delivery.

Similarly, a study of an intervention protocol comprised of brief advice, counseling, informational booklets, and a video administered at the time of well-child visits up to 6 months infant age (N = 2,901) showed no significant differences in maternal smoking between treatment and usual-care groups when the infants were 12 months old. In discussing their findings, the research team advanced the notion that stress associated with the transition to parenthood may have contributed to the observed incidence of smoking relapse (Severson, Andrews, Lichtenstein, Wall, & Akers, 1997). However, this variable was not measured.

A randomized trial (N = 399) also showed no significant difference in rates of not smoking at 1-year postpartum between a group of mothers who received brief advice to stop smoking and a quit-smoking booklet when compared to a group who received a structured intervention comprised of physician's advice and individual relapse prevention counseling during pregnancy (seeker-Walker, Solomon, Flynn, Skelly, & Mead, 1998).

McBride and colleagues (1999) evaluated the effectiveness of smoking relapse interventions (N = 900) during pregnancy and postpartum. Three levels of interventions were tested: (a) a self-help smoking-cessation booklet provided during pregnancy only; (b) the same booklet plus a personalized letter tailored to the woman's readiness to quit, a relapse prevention kit with tips for handling high-risk situations, and three prepartum counseling telephone calls; and (c) the prepartum interventions, plus three postpartum counseling telephone calls within the first 4 months after delivery. At 8 weeks and 6 months postpartum, the difference in relapse rates was not significant. By 12 months postpartum the proportion of women who were smoking was virtually the same in the three intervention groups. The researchers concluded that future interventions might need to be increased in intensity or duration if lasting outcomes are to be achieved.

Valanis and colleagues (2001) tested a low-intensity, smoking-cessation and relapse-prevention program based on brief motivational interviewing techniques delivered during prenatal visits and pediatrie well-child visits in a large health maintenance organization. A small but significant difference (11% control group vs. 14% intervention group) in self-reported smoking between 6 and 12 months after delivery was found in a study (N - 824). By contrast, Ratner, Johnson, Bottorff, Dahinten, and Hall (2000) tested a highintensity intervention that consisted of a nurse-delivered home visitation relapse prevention program (N = 238). This intervention also showed only small differences between treatment and control groups (21% vs. 19%) in 12-month continuous smoking abstinence.

These interdisciplinary studies have shown clinically nonsignificant sustainability for current smoking relapse prevention interventions for postpartum women. The approaches for relapse prevention that are effective for the general population have not been successful with postpartum women. To advance the science and, subsequently, the practice of nurses and all health care providers related to postpartum smoking behavior, an innovative approach is needed that blends existing knowledge of risk and protective factors for smoking relapse prevention for the general population with the substantial and evolving science in the area of transition to motherhood.

Postpartum Smoking

Smoking relapse in the first few months after the delivery of a baby might be a different process than that experienced by other smokers who have quit. In addition to the challenges that all smokers experience during attempts to quit, women who are making the transition to motherhood deal with the developmental transition from a known reality with respect to self-identity and important relationships to an unknown reality that requires adaptation.

The premise that postpartum smoking relapse might be different than that experienced by other smokers who quit was supported in McBride and colleagues' (1999) observation that studies of nonpregnant women who are not mothers of infants showed steep relapse curves in the initial days and weeks after cessation. Conversely, relapse after pregnancy was more gradual, with most relapses occurring over a period of months, rather than a few days or weeks, after delivery.

Postpartum smoking relapse might also be different than that experienced during pregnancy. In a qualitative study of women's smoking experience during pregnancy, Edwards and Sims-Jones (1998) found that concern about the baby's health was often described as the central reason for stopping or cutting down on smoking during pregnancy. Later, Johnson and colleagues (2000) found that 93.6% of the women in their study (N = 254) who had quit smoking during pregnancy reported having done so for the health of their unborn babies. However, they also found that, although almost all of these women (90.8%) reported that they intended to remain nonsmokers after pregnancy, most had relapsed by the time their babies were 6 months old. The observations made by these research teams indicate that smoking during pregnancy might be influenced by the mother's working model of caregiving in the earliest stage of becoming a mother. Protecting the unborn baby's health was a context-specific protective factor for preventing smoking relapse during pregnancy. In most cases, however, this protective factor was not sustained during the postpartum months. Knowledge of context-specific protective and risk factors for postpartum smoking relapse is needed.

Factors Linked with Postpartum Smoking

In an integrative review of studies of postpartum smoking relapse, the two dominant contributing factors identified were: (a) living with smokers; and (b) choosing bottle feeding instead of breast feeding (Zimmer, 2000). However, a study focused on early weaning (Ratner, Johnson, & Bortorff, 1999) showed that smoking relapse preceded weaning in the temporal chain for most women, thus, raising a question about the nature of the relationship between these two factors.

Additional factors that have been linked with postpartum smoking include quitting late in pregnancy, smoking more than one pack per day before pregnancy, minimal support for cessation efforts, low income, younger age, lower educational level, and alcohol use (Kahn, Certain, & Whitaker, 2002; Ratner et al., 2000; Severson, Andrews, Lichtenstein, Wall, & Zoref, 1995). Using population-based data from the Pregnancy Risk Assessment Monitoring System, Carmichael and Ahluwalia (2000) found several additional independent correlates of postpartum smoking relapse, including African American race and ethnicity, multiparity, and weight gain during pregnancy greater than 35 pounds.

Despite this relatively wide-range of variables associated with postpartum relapse, no studies were found on the association between relapse and context-specific variables associated with being the mother of an infant. In discussing their findings, however, several researchers cited here confirmed that the current identification of risk and protective factors for relapse during the first few months after birth is incomplete.

Collectively, these current studies show that some variables associated with smoking relapse among postpartum women are comparable to those for the general public, including low income, low educational attainment, and number of cigarettes per day smoked before the attempt to quit. However, they also indicated other possible protective or risk factors yet to be identified. Evidence that mothers quit smoking during pregnancy in order to protect the health of the baby and have the intention of remaining smoke-free after delivery for the same reason indicates that variables associated with the transition to motherhood might influence smoking behavior.

Becoming A Mother

New ways of thinking about being the mother of a baby could potentially advance current knowledge of postpartum smoking relapse and associated context-specific risk and protective factors. Mercer (2004) recommended that maternal-child researchers adopt the term "becoming a mother" to describe the unique experiences of women who undergo the process of establishing a maternal identity. This term connotes the dynamic process that has been confirmed in decades of research on the transition to motherhood.

Pridham and colleagues (Pridham, Shroeder, & Brown, 1999; Pridham, Lin, & Brown, 2001; Pridham, Saxe, & Limbo, 2004) posited that a mother's internal working model for infant caregiving influences her interpretation of information and subsequent evaluation of self and infant. The working model is comprised of motivations, feelings, expectations, beliefs, and intentions in the process of caregiving. It encompasses the mother herself, her infant, their relationship, and her caregiving activity and is characterized by changes as the mother, infant, and animate and inanimate environments change (K.F. Pridham, personal communication, March 29, 2005).

Likewise, the process of becoming a mother is both dynamic and evolutionary with stages that are overlapping and progressive, but not predictably linear. The stages of becoming a mother, according to Mercer, have relevance to advancing current knowledge of the achievement of maternal identity. An understanding of the processes of becoming a mother, with application of the internal working-model concept, might also contribute to an understanding of postpartum smoking relapse.

The first stage in becoming a mother occurs during pregnancy and is described as the commitment, attachment, and preparation stage. Mercer said that a woman's work during this stage has long-term implications for the quality of later adaptation to motherhood. Positive processing of the commitment, attachment, and preparations stage might explain the findings of two separate postpartum smoking relapse studies (Edwards & Sims-Jones, 1998; Johnson et al., 2000) that mothers who quit smoking during pregnancy reported that the central motivation for doing so was to promote the health of their babies. This finding is consistent with the theoretical premise of Pridham et al. (1999) that a mother's working model of caregiving influences goal-directed activities, such as making behavior changes to keep her baby safe.

During the first 2 to 6 weeks after birth the process of establishing maternal identify in becoming a mother is one of acquaintance, learning, and physical restoration. Mercer said that this is a stage in which a mother studies her infant's responses to her and learns through trial and error how to comfort and care for her infant. Pridham and colleagues (1999) pointed out that a mother's working model of infant caregiving at this stage is influenced by her own childhood and adolescent experiences as well as her current personal resources. The adaptiveness of a mother's working model for caregiving at this stage might also reflect infant characteristics such as health state, responsiveness, and consolability. The stage acquaintance, learning, and physical restoration in becoming a mother coincides with a physiologic stage of infants in which they normally have increased levels of crying. This potential stressor is particularly operative when mothers have experienced a smoking lapse or relapse early in postpartum; the presence of colic with associated inconsolable crying episodes has been found to be higher among infants of smoking mothers than in those of nonsmoking mothers (Reijneveld, Brugman, & Hirasing, 2000). As a mother's expectations and evaluations of herself and her infant change during this stage, her smoking abstinence self-efficacy and subsequent risk for lapses and relapse might also shift.

The third stage in becoming a mother is "moving toward a new normal." It generally occurs between 2 weeks and 4 months of infant age. Mercer (2004) explained that women in this stage structure their mothering to fit themselves and their families based on past experiences and future goals. They make adjustments to a new reality based on what they have learned from their infant's cues and increasing knowledge about what is best for their infants. During this stage, the adaptiveness of a mother's working model of infant caregiving might vary as experience with her infant evolves, her life changes in relation to the developing skills of her infant, and as she encounters new opportunities and demands from important people in her social network (Pridham et al., 1999; 2001, 2004).

Achievement of the maternal identity, according to Mercer, occurs about 4 months after delivery. At this stage of becoming a mother, a woman has established an intimate knowledge of her infant, feels competent and confident in her mothering, and feels love for her infant. She has reached a new normal with the significant people in her life. Further, she experiences a transformation of herself as she incorporates an expanded identity that includes assuming responsibility for the welfare of her infant. George and Solomon (1999) proposed that the mother's evaluation of herself as effective in providing protection for her infant, a hallmark of the working model during this stage, is a powerful influence on their mutual attachment. For many mothers, the return to smoking along with an increased likelihood of nicotine dependence has already occurred before this relatively calm and satisfying time in maternal development. To date, the potential conflict between a mother's intentions to be her infant's protector and the challenge to maintain her own smoking abstinence self-efficacy has not been explored.

Recommended Research

The next necessary step in advancing postpartum smoking relapse research is to design studies based on a conceptual framework comprised of the two dynamic, evolving and overlapping processes, smoking abstinence self-efficacy and becoming a mother. Designing studies of postpartum tobacco use based on this blended conceptual framework will enhance the likelihood that new knowledge will be gained about interrelationships between the two processes. Findings about these interrelationships might indicate previously unidentified factors that distinguish postpartum smoking relapse from smoking in the general population.

Using this conceptual framework as a foundation for future studies might also assure that the variables selected to test questions about protective and risk factors for postpartum smoking relapse will be uniquely tailored to the experiences of mothers of infants. For example, a researcher might test the contribution of infant irritability to postpartum smoking relapse. Infant irritability is a common and challenging problem that generally begins during the 1st month of life and subsides by 4 months of age (Keefe, Froese-Fretz, & Kolzer, 1998). This time overlaps with Mercer's stage of becoming a mother, entitled moving toward a new normal. In a qualitative study of infant irritability, mothers described feelings of frustration, anger, depression, isolation, guilt, and self-doubt when they could not console their babies. They questioned their abilities as mothers (Keefe & Froese-Fritz, 1991). Coincidently, research on postpartum smoking has shown a high incidence of relapse during the same time that infant irritability peaks, indicating that an untested relationship might exist. Using the proposed conceptual framework as a foundation for study, investigators might test whether perceived difficulty in consoling a crying infant contributes to delays in achieving maternal identity and concomitantly serves as a trigger for a return to smoking as comforting and familiar, a return to a past feeling of normalcy.

In addition to the examination of a mother's capacity to cope with inconsolable infant crying within the context of the dynamic and interactive processes of developing postpartum smoking abstinence self-efficacy and becoming a mother, many other potential factors might be tested. A few examples include postpartum depression, postpartum fatigue, and having a preterm baby. These variables have been associated with interruptions in the process of becoming a mother, and they influence the goals, motivations, expectations, and intentions that shape her working model of parenting and caregiving. Concomitantly, they might also be context-specific factors that influence a return to smoking. Using a conceptual framework to examine the interactions of the two overlapping developmental processes might yield new knowledge about this recalcitrant clinical problem that cannot be achieved when either process is examined separately.

Conclusions

Interventions to prevent smoking relapse in postpartum women have been substantially ineffective. Future clinical intervention research should be focused on postpartum smoking relapse within a conceptual framework comprised of two overlapping developmental processes: smoking abstinence self-efficacy and becoming a mother. The underpinnings of this framework are theoretically sound and highly familiar to both new mothers who try to remain smoke-free and the healthcare providers who support them. Evidencebased interventions are needed to help women cope and thrive as they progress through these processes. The use of the proposed conceptual framework is likely to advance scientific knowledge of tobacco use by this special subpopulation of women, consistent with recommendations of the recent National Cancer Institute (2004) on Women, Tobacco, and Cancer.

(C) 2006 Journal of Nursing Scholarship. via ProQuest Information and Learning Company; All Rights Reserved