تعدیل اثر سطح حداکثر مصرف سیگار بر روی اختلال پانیک در طول عمر با روان رنجوری: آزمون استفاده از اپیدمولوجیکی تعریف شده نمونه ملی افراد سیگاری

The present study evaluated a moderational model of neuroticism on the relation between smoking level and panic disorder using data from the National Comorbidity Survey. Participants (n = 924) included current regular smokers, as defined by a report of smoking regularly during the past month. Findings indicated that a generalized tendency to experience negative affect (neuroticism) moderated the effects of maximum smoking frequency (i.e., number of cigarettes smoked per day during the period when smoking the most) on lifetime history of panic disorder even after controlling for drug dependence, alcohol dependence, major depression, dysthymia, and gender. These effects were specific to panic disorder, as no such moderational effects were apparent for other anxiety disorders. Results are discussed in relation to refining recent panic–smoking conceptual models and elucidating different pathways to panic-related problems.

There has been a long-standing recognition that rates of cigarette smoking are greater among persons with certain types of psychopathology than in persons without a history of psychopathology. To date, most of the investigations in this domain have focused on individuals with schizophrenia, alcohol and drug dependencies, attention deficit hyperactivity disorder, and depressive disorders (e.g., Black et al., 1999 and Merikangas et al., 1998). There has been a recent focus on addressing associations between smoking and panic disorder (Zvolensky and Schmidt, 2004). This attention is due, in part, to a number of studies documenting a relation between panic psychopathology and smoking status (Amering et al., 1999, Degenhardt et al., 2001, Hayward et al., 1989, Himle et al., 1988, Kandel et al., 1997, McCabe et al., 2004 and Pohl et al., 1992). In a recent epidemiological study, for instance, approximately 43% of respondents diagnosed with panic disorder in the past month were current smokers compared with approximately 22% observed among individuals with no mental illness (Lasser et al., 2000). Moreover, the observed associations between smoking and panic problems are not attributable to sociodemographic characteristics, other psychiatric conditions, or symptom overlap in diagnostic criteria for panic attacks and nicotine dependence (Zvolensky et al., 2003c). Aside from high rates of smoking among those with panic disorder, a number of studies suggest that cigarette smoking increases the chance of developing panic attacks and panic disorder. Using data from the Epidemiologic Study of Young Adults and the National Comorbidity Survey Tobacco Supplement, Breslau and Klein (1999) found that self-reported daily smoking was associated with retrospective report of the first occurrence of a panic attack as well as panic disorder. On the other hand, panic attacks or panic disorder did not increase the risk for subsequent smoking behavior. In another study, Johnson et al. (2000) investigated the longitudinal association between cigarette smoking and anxiety disorders among adolescents and young adults using a community-based sample (n = 688). Heavy smoking (≥ 20 cigarettes per day) during adolescence (MAge = 16) was associated with higher risk of developing panic disorder and agoraphobia during early adulthood (MAge = 22) even after controlling for a variety of theoretically relevant factors (e.g., alcohol and other drug use). More specifically, adolescents who were heavy smokers were 15.5 times more likely to develop panic disorder in early adulthood than non-smokers, suggesting heavier smoking levels impart a substantial panic-related risk. Similar results recently were found in a longitudinal study conducted over a 4-year period in Germany with over 2500 participants (aged 14–24 years at baseline; Isensee et al., 2003). In this study, nicotine-dependent smokers were at greater risk for later onset of panic attacks. Although initial work suggests that smoking, particularly heavy amounts, is associated with increased risk for developing panic-related problems, there has been little focus on potential moderators of this relation. In a moderational model, the effects of smoking level on panic vary as a function of another vulnerability factor (Baron and Kenny, 1986). In a recent study, Zvolensky et al. (2003a) examined whether anxiety sensitivity, a cognitive-based risk factor for anxiety psychopathology defined as the fear of anxiety and anxiety-related sensations (McNally, 1990 and Reiss and McNally, 1985), moderated the effects of level of smoking in regard to prototypical panic outcome variables (assessed via self-report) in an epidemiologically defined sample of regular smokers from Moscow, Russia. According to theory and research on anxiety sensitivity, individuals with fears of anxiety and anxiety sensations may be more apt to respond with anxiety to smoking-related interoceptive cues, and catastrophize the potential negative consequences of such stimuli. As expected, results indicated that anxiety sensitivity moderated the effects of smoking (as indexed by cigarettes smoked per day) on self-reported agoraphobic avoidance; this effect was observed above and beyond variance attributed to problematic alcohol use and negative affectivity. No moderating effect of anxiety sensitivity was found for panic attacks, however, perhaps due to the restriction of the assessment of panic attacks to the most recent week. Overall, these initial findings suggest that smokers are not a homogeneous group in regard to panic problems, and that anxiety sensitivity is one of perhaps several individual difference factors that may be important in accounting for such differences. Although the results of Zvolensky et al. (2003a) are promising, it is noteworthy that only the cognitive-based risk factor of anxiety sensitivity has been examined as a moderator of the smoking level–panic relation. Thus, it is currently unclear whether more general psychological vulnerability factors also may function as moderators of smoking level in terms of panic-related problems. A large body of empirical work has identified personality variables (e.g., neuroticism) that reflect a generalized disposition to experience negative affect (i.e., negative affectivity), which provides a common and relatively stable diathesis for anxiety- and mood-related disorders (Brown et al., 1998, Clark et al., 1994, Fowles, 1995, Trull and Sher, 1994 and Zinbarg and Barlow, 1996).1 Additionally, research suggests such personality variables are largely genetically based, arise early in the developmental lifespan, and may confer risk for cigarette smoking (Kendler et al., 1993 and Lerman et al., 1999). Negative affectivity is relevant to panic disorder, specifically in the sense that contemporary models of the etiology of the disorder posit that a generalized tendency to experience negative affect may enhance the tendency to develop conditioned emotional responses to bodily cues (Bouton et al., 2001). Consistent with such accounts, Hayward et al. (2000) found that negative affectivity prospectively predicted the onset of panic attacks among a nonclinical sample (n = 2365); moreover, negative affectivity, relative to anxiety sensitivity (as well as gender and presence of childhood separation anxiety disorder), was the most robust predictor of panic attack onset. These data suggest negative affectivity may be a fundamental vulnerability variable relevant to panic-related problems. One study has addressed the role of neuroticism in the relation between panic and smoking (Goodwin and Hamilton, 2002). In this investigation, participants were drawn from the Midlife Development study (n = 3032). Results indicated that neuroticism was a significant predictor of the co-occurrence of panic disorder and smoking, suggesting this factor may be a shared vulnerability for the co-occurrence of these problems. The results of this investigation underscore the general utility in better understanding the role of neuroticism in the smoking–panic association. Yet, it remains unclear whether a disposition to experience negative affect moderates the effect of smoking level on panic disorder among regular smokers. The identification of such moderating effects among smokers is clinically and theoretically important, as it helps to refine our understanding of the association between cigarette smoking and panic psychopathology, and to identify subpopulations of smokers with potentially different causal mechanisms or course of illness ( Kraemer et al., 2002). A recent integrative model of panic and smoking (Zvolensky et al., 2003b and Zvolensky et al., 2003c) notes that regular smoking produces a range of aversive internal cues in the form of physical health impairment (e.g., lung impairment) or physical disease (e.g., asthma) and nicotine withdrawal symptoms. Descriptive research further suggests that these aversive interoceptive cues represent potent stimuli to which vulnerable persons respond with heightened negative emotional responses (Last et al., 1984, Rapee et al., 1990 and Roy-Byrne et al., 1986). Moreover, various smoking-related cardiopulmonary medical problems (e.g., asthma, pneumonia, bronchitis) are commonly reported as preceding the development of panic attacks and panic disorder (Verburg et al., 1995). In this context, persons high compared with low in negative affectivity are more likely to respond to smoking-related internal cues with elevations in anxiety and fear. It is noteworthy that heavier amounts of smoking increase the level and types of aversive internal cues (American Cancer Society, 1999), thereby promoting risk of an anxiety response among emotionally vulnerable individuals. In the absence of an opportunity to smoke to cope with these distressing experiences, a further intensification of anxiety symptoms may occur, perhaps culminating in a panic attack. From this perspective, a generalized tendency to experience negative affect would enhance the likelihood of panic-relevant emotional learning to internal cues (enhanced intensity of emotional learning) and thereby increase vulnerability for developing panic-related problems via interoceptive conditioning (Barlow, 2002 and Bouton et al., 2001). The purpose of the present investigation was to empirically evaluate the potential moderating role of negative affectivity on the relation between smoking level and panic disorder. Participants were drawn from the National Comorbidity Survey (NCS; Kessler et al., 1997). The present study focused on maximal smoking level (lifetime) given empirical evidence documenting that heavier smoking amounts impart greater risk for panic-related problems (Johnson et al., 2000 and Zvolensky et al., 2003a) and, theoretically, smoking more cigarettes per day would provide an index of “greater exposure” to panic-relevant aversive interoceptive cues (e.g., bodily sensation and withdrawal symptoms; Zvolensky et al., 2003b and Zvolensky et al., 2003c). Based upon the moderational model, it was expected that a generalized tendency to experience negative affect (neuroticism) would moderate the effects of maximum smoking frequency (i.e., cigarettes smoked per day during heaviest smoking period) on lifetime history of panic disorder even after controlling for drug dependence, alcohol dependence, major depression, dysthymia, and gender. If negative affectivity enhances emotional learning to smoking-related internal cues, the moderational effects of this factor should be specific to panic disorder due to the important role of internal cues in panic problems (Bouton et al., 2001), rather than being applicable to all types of anxiety psychopathology. Thus, no moderational effects of neuroticism on maximum smoking level were hypothesized for agoraphobia without panic disorder, as research has demonstrated that such a clinical condition is not necessarily accompanied by panic attacks (Fava et al., 1988 and Marks, 1987) and presumably interoceptive conditioning of anxiety to bodily sensations. Furthermore, whereas negative affectivity was expected to be associated with an increased risk for anxiety disorder diagnoses in general, it was not expected to moderate the relation between maximum smoking level and other anxiety disorders (i.e., demonstrate specificity for panic disorder).