حساسیت اضطراب: یک پیش بینی کنده منحصر به فرد از ترک اعتیاد در میان مصرف کندگان هروئین و کراک/کوکائین در درمان مصرف مواد ساکنان داخلی شهرها

The present study examined the extent to which anxiety sensitivity (AS) at treatment entry was related to prospective treatment dropout among 182 crack/cocaine and/or heroin-dependent patients in a substance use residential treatment facility in Northeast Washington, DC. Results indicated that AS incrementally and prospectively predicted treatment dropout after controlling for the variance accounted for by demographics and other drug use variables, legal obligation to treatment (i.e., court-ordered vs. self-referred), alcohol use frequency, and depressive symptoms. Findings are discussed in relation to the role of AS in treatment dropout and substance use problems more generally.

Substance use disorders are widespread among the general population and associated with significant economic, societal, and personal costs (Grant et al., 2004; Regier et al., 1990). Though many individuals seek treatment for such problems, a large percentage drop out of treatment prematurely and relapse soon after (Hubbard, Craddock, Flynn, Anderson, & Etheridge, 1997; Ravndal & Vaglum, 2002; Simpson, Joe, & Brown, 1999). A growing number of investigations have empirically explored potential predictors of substance use treatment dropout, including demographics (Maglione, Chao, & Anglin, 2000), psychiatric symptoms (Hattenschwiler, Ruesch, & Modestin, 2001), emotional symptoms (McCusker, Stoddard, & Frost, 1996), drug use severity (Ravndal & Vaglum, 1991), and a variety of social–cognitive variables (e.g., social support, self-efficacy, motivation to quit; Blanchard, Morgenstern, Morgan, Labouvie, & Bux, 2003; Daley, Salloum, Zuckoff, & Kirisci, 1998; Mertens & Weisner, 2000; Messina, Wish, & Nemes, 2000). However, due to little agreement on the consistency or the generalizability of these findings (Agosti, Nunes, Stewart, & Quitkin, 1991; Alterman, McKay, Mulvaney, & McLellan, 1996; Claus, Kindleberger, & Dugan, 2002; McFarlain, Cohen, Yoder, & Guidry, 1977; Nemes, Wish, & Messina, 1999), it has become apparent that it is important to identify and examine other constructs that may contribute to the understanding of the processes involved in a patient's decision to remain in or prematurely leave residential treatment. Building from contemporary models of psychological vulnerability which suggest that the ways in which individuals evaluate and respond to internal events may influence risk for a variety of negative outcomes (e.g., Barlow, 2002), there is good reason to explore theoretically relevant cognitive factors reflecting a hypersensitivity to aversive events in an effort to better understand treatment-dropout processes among drug-using populations (Brown, Lejuez, Kahler, Strong, & Zvolensky, 2005). One such cognitive vulnerability variable that may prove to be especially useful in understanding premature treatment attrition is that of anxiety sensitivity (AS). AS is an individual difference factor reflecting the fear of anxiety-related sensations, which arise from a belief that such sensations have harmful personal consequences (Reiss & McNally, 1985). The global AS construct encompasses fears of the physical, mental, and social consequences of anxiety-related sensations (Zinbarg, Barlow, & Brown, 1997), all of which can function to theoretically amplify pre-existing anxiety (Reiss, 1991). AS has assumed an increasingly important, clinically relevant role as a cognitive-based vulnerability factor for emotional disorders (McNally, 2002; Taylor, 1999). Indeed, recent studies suggest AS is related to maladaptive forms of emotional processing and affect regulation such as catastrophic thinking (Zvolensky, Forsyth, Bernstein, & Leen-Feldner, 2007; Zvolensky, Kotov, Antipova, & Schmidt, 2005), avoidance-based coping (Feldner, Zvolensky, Stickle, Bonn-Miller, & Leen-Feldner, 2006; Tull & Gratz, 2008; Zvolensky & Forsyth, 2002), as well as emotion dysregulation (Kashdan, Zvolensky, & McLeish, 2008; Tull, 2006; Tull & Gratz, 2008). Further, AS has been found to be associated with a general inability to tolerate uncomfortable bodily sensations (Schmidt, Richey, & Fitzpatrick, 2006) which may place an individual at risk for worse anxiety-related outcomes. Specifically, high AS, combined with an intolerance for uncomfortable bodily sensations, has been found to be predictive of greater subjective anxious responding to a CO2 challenge (Schmidt, Richey, Cromer, & Buckner, 2007). The recognition that AS is related to aversive anxiety states and dysfunctional aspects of emotional processing (McNally, 2002) has lead researchers to theorize that this cognitive factor may be related to the maintenance of substance use disorders (Brown et al., 2005; Morissette, Tull, Gulliver, Kamholz, & Zimering, 2007; Norton, 2001; Otto, Safren, & Pollack, 2004; Otto, Powers, & Fischmann, 2005; Stewart & Kushner, 2001; Stewart, Samoluk, & MacDonald, 1999; Tull, Baruch, Duplinsky, & Lejuez, 2007; Zvolensky & Bernstein, 2005; Zvolensky, Schmidt, & Stewart, 2003). To the extent that AS reflects a sensitivity to (and is related to an intolerance of) certain internal states, individuals with greater levels of this cognitive factor may be apt to use drugs and/or alcohol to regulate affective distress. In line with predictions derived from such models, AS has been found to be related to coping-oriented motives for cigarette smoking (Brown, Kahler, Zvolensky, Lejuez, & Ramsey, 2001; Zvolensky, Kotov, Antipova, Leen-Feldner, & Schmidt, 2005), alcohol (Conrod, Pihl, & Vassileva, 1998; Stewart, Karp, Pihl, & Peterson, 1997; Stewart, Zvolensky, & Eifert, 2002), and cannabis (Bonn-Miller, Zvolensky, & Bernstein, 2007; Comeau, Stewart, & Loba, 2001). Additionally, AS is related to heavier alcohol use patterns (Cox, Swinson, Shulman, Kuch, & Reichman, 1993; Stewart, Peterson, & Pihl, 1995; Stewart, Zvolensky, & Eifert, 2001; Zvolensky et al., 2005) and may be elevated among individuals who use substances that primarily function to dampen central arousal such as heroin (Lejuez, Paulson, Daughters, Bornovalova, & Zvolensky, 2006). These data collectively suggest that AS is related to coping-oriented use patterns for numerous substances. Recognizing the explanatory utility of AS to coping-oriented substance use and other problematic aspects of drug use behavior, researchers have theorized that this cognitive factor may increase risk for poor substance use treatment outcomes (Otto et al., 2005; Stewart et al., 1999; Zvolensky et al., 2005). To the extent that individuals with high AS do not have adequate psychological resources to successfully cope with such aversive events, they may be more likely to prematurely terminate substance use treatment. This research is supported by a study suggesting that AS, relative to other predictors, may have utility in identifying patients who drop out of antidepressant trials, perhaps because of a heightened sensitivity to the side effects from these medications (Tedlow et al., 1996). Likewise, data suggest that heightened AS is a risk factor for early lapse during smoking cessation (Brown et al., 2001), and the degree of change in AS has been a significant predictor of relapse among those discontinuing use of benzodiazepines (Bruce, Spiegel, Gregg, & Nuzzarello, 1995). Building from these previous AS findings, we examined the extent to which AS at baseline (i.e., treatment entry) was related to prospective treatment dropout. This research was conducted in an inner-city residential treatment center, a setting which is especially relevant for such work given that it often includes especially difficult to treat patients and evidences especially low rates of treatment completion (SAMHSA, 2002). Further, this work adds to a relative small body of research examining AS among minority individuals and specifically minority inner-city substance users (cf. Lejuez et al., 2006). In doing so, we controlled for other theoretically relevant factors (e.g., demographics and other drug use variables, depressive symptoms, and legal obligation to treatment) in an effort to isolate the unique variance accounted for by AS in regard to treatment dropout and to ensure such effects are not attributable to shared variance with other theoretically relevant characteristics.