; McDowell, Wang, & Kennedy-Stephenson, 2008). Studies have also indicated that women who are married, have greater education, and higher income are also more likely to breast feed (Hendricks, Briefel, Novak, & Ziegler, 2006; Li et al.; McDowell et al.). In contrast with other research (Hendricks et al.; Li et al.), greater age was not significantly associated under with breast feeding in the current sample. Perhaps this is because participants in the current study were somewhat younger (M = 24.8 years) than the women in other nationally representative samples (Hendricks et al.; Li et al.). Several limitations of the current study should be noted.

It is possible that the generalizability of the findings may be limited to treatment-seeking women who quit smoking early in pregnancy, and may not generalize to other groups of women, such as those who quit smoking later in pregnancy or those who do not choose to participate in smoking cessation treatment. Furthermore, the failure of many participants to attend the 8 weeks postpartum visit resulted in some missing breast-feeding status data. Although no demographic or socioeconomic differences were found between those who provided information on breast-feeding status and those with missing breast-feeding status, participants were found to differ in the number of cigarettes smoked per day and the time until first cigarette smoked in the morning (i.e., measures of nicotine dependence). As such, these variables were included as covariates in all models to adjust for differences.

However, it is possible that missing data occurred as a result of other unobserved variables for which we were not able to adjust. The current study also did not evaluate the influence of breast-feeding duration or exclusivity on postpartum smoking abstinence. Some research suggests that breast feeding for ��6 months may have a greater impact on smoking cessation than breast feeding for shorter durations (Kaneko et al., 2008). It is also possible that exclusive breast feeding may be a more effective deterrent of tobacco use than mixed breast/formula feeding because exclusively breast-feeding women who use tobacco would have greater difficulty avoiding or minimizing infant exposure to nicotine. Another limitation is that the current study does not illuminate the mechanisms that may link breast feeding with the prevention of smoking relapse.

One possible explanation is that women may avoid smoking due to concerns about the effects of nicotine on their nursing infants (Edwards & Sims-Jones, 1998). Other research has indicated that breast feeding is associated with reduced negative affect, perhaps due to the release of oxytocin (Mezzacappa & Katkin, 2002). Given that negative affect is commonly found to be associated with smoking relapse in general (e.g., GSK-3 Shiffman et al., 2007) and also during the postpartum period (e.g., Park et al.