S.R. Battista et al. / Addictive Behaviors 33 (2008) 1402–1408
Theory also predicts that AS should be unrelated to smoking for positive reinforcement motives (Zvolensky et al., 2003). However, ASI total scores were associated with positive reinforcement motives in bivariate correlational analyses. Although theoretically unexpected, this nding is consistent with prior work showing AS to be correlated with scores on the original pleasurable relaxation scale of the RFS (Brown et al., 2001). Further analyses helped to resolve this issue. When partial correlations were computed, ASI scores remained signicantly associated with negative reinforcement smoking motives when controlling for positive reinforcement motives, and not vice versa, supporting our prediction of specicity in the relation of AS to negative reinforcement smoking motives (Zvolensky et al., 2003). Another goal of the current study was to examine the relationship between each lower-order AS component and the two higher-order smoking motives. First, there was no association between AS social concerns and either of the higher-order smoking motives. This result is consistent with previous ndings (Zvolensky et al., 2006) and provides evidence on specicity in the dimensions of AS which are related to smoking motives. While AS physical and psychological concerns were both positively associated with both negative and positive reinforcement factor scores in bivariate correlational analyses, only AS physical concerns were found to be more strongly related to negative reinforcement than to positive reinforcement motives when comparing correlation coefcients, and only AS psychological concerns remained signicantly associated with positive reinforcement smoking motives when controlling for negative reinforcement motives. Together, the
ndings indicate that smokers who fear physical arousal sensations are specically motivated to smoke as a means of managing
these sensations, whereas those smokers with high AS psychological concerns are more generally motivated to smoke both to relieve aversive states and to achieve pleasurable states.
One possible explanation for the different pattern of relations of AS physical concerns vs. psychological concerns with the two higher-order smoking motives relates to the connection between the lower-order AS components and both panic disorder and depression. AS physical concerns is the AS component most related to panic disorder (McLaughlin, Stewart, & Taylor, 2007), whereas AS psychological concerns is the component most related to depression (e.g., Schmidt, Lerew, & Joiner, 1998). While panic disorder (Zvolensky et al., 2003) and depression (Morrell & Cohen, 2006) are both commonly comorbid with nicotine dependence, the present ndings suggest that future research is needed to explore the processes connecting these two emotional disorders to nicotine dependence. In the case of panic, the reinforcement processes explaining high rates of nicotine dependence might be negative reinforcement specic. However, with depression, the reinforcement processes may be more general, also involving positive reinforcement motives given the prominent anhedonia characteristic of depression.
The current study was not without limitations. First, our sample consisted of individuals completing a smoking cessation program and may not be representative of all smokers. Second, the sample size, although sufcient for PCA of a 23-item measure, was relatively small. We recommend that a future study should make use of an even larger sample of smokers in order to examine the RFS structure using conrmatory factor analysis (CFA). CFA would allow for a direct comparison of several theoretically defensible models of the structure of the RFS: a 6-factor solution (Ikard et al., 1969), a 2-factor solution, a 3-factor solution (Pomerleau et al., 2003), or a hierarchical model with six lower-order factors and two or three higher-order factors. Future research is also necessary to explore these suggested factor structures using other measures of smoking motives (e.g., the Wisconsin Inventory for Smoking Dependence Motives-68 [Piper et al., 2004], which contains a greater number of items and subscales than the RFS), and to also examine their relationship to AS. Additional studies in this area should also include measures of psychopathology to examine whether the relations between AS psychological concerns and smoking for positive reinforcement motives are still maintained after controlling for depression, for example.
The current ndings suggest a number of important clinical implications. In treatment, smokers with high AS physical concerns should have their negative reinforcement smoking motives be a specic target of treatment through such techniques as: (a) expectancy challenge (Darkes & Goldman, 1998); (b) increasing awareness of the link between negative states and smoking behavior; and (c) learning to tolerate negative affect and arousal (Zvolensky, Bernstein, Yartz, McLeish, & Feldner, 2008). On the other hand, smokers with high AS psychological concerns will also need to have positive reinforcement smoking motives be targets for intervention, for example through the use of pleasant event scheduling (Lewinsohn, Sullivan, & Grosscup, 1980) where clients are taught alternative ways of achieving pleasure. The nding that AS was associated with a negative reinforcement smoking motives factor that included strong salient loadings from habitual motives suggests that smoking may be highly automatized for high AS smokers. Recent research has begun exploring how to treat relatively more “implicit” or automatic aspects of addiction, such as through training in implicit avoidance associations (Stacy & Wiers, 2006) or via mindfulness techniques to break automatic behavior patterns (Witkiewitz, Marlatt, & Walker, 2005). These approaches may also prove useful for high AS smokers.
This research was supported by start-up grants from Capital District Health Authority, Addiction Prevention and Treatment Services, and from the Dalhousie University Department of Psychiatry Research Fund awarded to the second author.
Susan Battista is supported by a Doctoral student scholarship from the Social Sciences and Humanities Research Council of Canada. Sherry H. Stewart is supported by an Investigator Award from the Canadian Institutes of Health Research and by a Killam Research Professorship from the Dalhousie University Faculty of Science. Heather Fulton is supported by a Doctoral student scholarship from the Killam Trusts. Christine Darredeau is supported by a Sobey Postdoctoral Fellowship in Psychiatry Research from the Dalhousie University, Department of Psychiatry.
We wish to acknowledge Kelly Evans, MEd, Paul Helwig, MSW, and Tom Payette, MSW from Addiction Prevention and Treatment Services, Capital District Health Authority, Dartmouth, Nova Scotia, Canada for their help in recruiting participants for this project. We also wish to thank Ellen Rhyno, Jennifer Mullane, and Adrienne Girling for their invaluable research assistance.