Pharmacologic and Behavioral Withdrawal From Addictive Drugs

Recent theories suggest that drug withdrawal does not motivate drug use and relapse. However, data now show that withdrawal produces complex changes over time in at least two symptoms (i.e., negative affect and urges) that are highly predictive of relapse. Evidence suggests that falling levels of the drug in the blood and interruption of the drug self-administration ritual both affect these symptoms. Both of these forms of withdrawal motivate renewed drug use in addicted individuals. KEYWORDS—smoking; tobacco dependence; tobacco withdrawal; drug withdrawal Addicts have written powerfully about the ‘‘abstinence agony’’ that occurs when they stop using a drug. For instance, Sigmund Freud described quitting smoking as an ‘‘agony beyond human power to bear.’’ One would assume from such accounts that drug withdrawal produces a powerful motive to resume or continue drug use. Indeed, movies and other popular accounts of addiction typically emphasize the role of withdrawal. However, current theoretical models of addiction downplay the role of drug withdrawal in the maintenance of addictive behaviors (Robinson & Berridge, 1993). Such models hold that withdrawal symptoms do not motivate relapse; for example, measures of withdrawal severity do not predict who is likely to relapse. Also, these models assert that withdrawal is brief and, therefore, cannot account for relapse that occurs long after drug use. Finally, these models assert that effective addiction treatments do not work via the suppression of withdrawal symptoms. These theoretical views of drug motivation emphasize incentive or reward processes rather than withdrawal. In contrast to the claims of recent theories, addicted individuals typically report that withdrawal symptoms motivate them to relapse and that fear of withdrawal causes them to maintain drug use. There is now mounting evidence that the addicted individuals are correct—that withdrawal is a crucial motivator of their drug use. While drug use is no doubt determined by multiple factors, there is compelling evidence that, in the addicted individual, withdrawal potently influences the fluctuating course of drug motivation. We believe that the motivational impact of withdrawal has been obscured by a failure to assess it sensitively and comprehensively. There are two reasons for this failure. One is that withdrawal is multidimensional, and only some elements, such as urges and negative affect, have motivational relevance. Unless studies focus on these symptoms, the motivational impact of withdrawal may be lost. The second reason is that most previous assessments of withdrawal have not adequately captured its dynamic symptom patterns, which may be both highly complex and persistent. These complex symptom patterns provide important clues regarding the nature and determinants of withdrawal. Withdrawal symptoms appear to reflect the effects of two distinct types of deprivation: deprivation of the drug molecule and deprivation of the drug-use instrumental response (such as injecting a drug or lighting and smoking a cigarette). A reduced level of the drug in the body, or pharmacologic withdrawal, results in the escalation of symptoms that has traditionally been labeled withdrawal. However, ceasing drug use also deprives the individual of a behavioral means of regulating or coping with escalating symptoms such as negative affect—in other words, it also causes behavioral withdrawal. At the heart of this model is the notion that the self-administration ritual per se quells withdrawal symptoms and that the absence of the ritual will actually exacerbate symptoms because of a disruption in symptom-regulatory processes. In theory, this disruption leads to very persistent and complex symptom profiles because symptoms may arise in response to cues that occur months after Address correspondence to Timothy B. Baker, Center for Tobacco Research and Intervention, University of Wisconsin School of Medicine and Public Health, 1930 Monroe St. Suite 200, Madison, WI 53711; e-mail: [email protected]. CURRENT DIRECTIONS IN PSYCHOLOGICAL SCIENCE 232 Volume 15—Number 5 Copyright r 2006 Association for Psychological Science discontinuing drug use. This symptom dysregulation will persist until drug cues lose their associative strength (e.g., via extinction) and/or until the individual acquires a coping response that replaces use of the drug. COMPONENTS OF WITHDRAWAL Physical Signs Previous views of withdrawal have been unduly influenced by characteristics of the physical symptoms of withdrawal. Each class of addictive drug produces a withdrawal syndrome that comprises different sorts of physical signs. For instance, ethanol withdrawal produces tremors, exaggerated reflexive behavior, and sometimes convulsions. Opiate withdrawal produces hypothermia, piloerection (gooseflesh), rhinnorhea (nasal discharge), and diarrhea. These signs all tend to follow the same rise-and-fall pattern after the discontinuation of drug use, with symptoms being largely absent within a couple of weeks after cessation. Research has shown that these physical signs are not consistently related to drug motivation (e.g., Baker, Piper, McCarthy, Majeskie, & Fiore, 2004), supporting the idea that withdrawal is motivationally inert. However, the motivational irrelevance of these physical signs should not be surprising as they are so dissimilar across different types of drugs. If withdrawal has a motivational influence that is common to all addictive drugs, it seems sensible to look for this influence among the symptoms that are themselves common across drugs. Negative affect and drug urges are such symptoms. Negative Affect Many of the symptoms used to characterize withdrawal are, in fact, affective terms such as ‘‘irritable,’’ ‘‘stressed,’’ ‘‘anxious,’’ and ‘‘depressed.’’ Robust correlations are observed between measures of withdrawal and mood, and factor-analytic studies have demonstrated that affective items capture much of the reliable variance in withdrawal measures (Piasecki et al., 2000). Experimental manipulations of tobacco withdrawal in the laboratory prompt increases in self-reported and physiological indicators of negative affect (Hogle & Curtin, in press). A listing of negative mood adjectives does not do justice to the affective consequences of withdrawal. Addicted individuals commonly report that giving up a drug seems like losing a dear friend or experiencing a death of a family member. We believe that this reflects a crucial part of the withdrawal syndrome: a feeling akin to social loss or separation distress. Indeed, at the neuropharmacologic and experiential levels, withdrawal produces effects similar to intense social loss (Panksepp, Herman, Connor, Bishop, & Scott, 1978). However, the relationship with the drug, once lost, can be reinstated at any time. There is evidence that the emotional distress of withdrawal differs from other withdrawal elements in terms of both its motivational significance and its physiological substrata. For instance, researchers have shown that brain structures associated with the motivational components of the withdrawal syndrome (e.g., negative affect) show different sensitivity to the opioid antagonist, naloxone, than do brain structures associated with the somatic components (Frenois, Cador, Caille, Stinus, & Le Moine, 2002). In addition, research shows that it is the affective and not the somatic signs of withdrawal that are responsible for its motivational effects (Mucha, 1987; Piasecki et al., 2000). In sum, assessment strategies should target the affective elements of the withdrawal syndrome if the intent is to assess drug motivation or relapse vulnerability. Urge/Craving An urge may be viewed as the conscious recognition of the desire to use a drug. Since a variety of influences may stimulate such desire and make it available to consciousness, urges are not uniquely related to withdrawal (as negative affect is not uniquely related). However, urge measures appear to be sensitive indices of withdrawal and rise precipitously in response to abstinence (Baker et al., 2004). There exist both biological and theoretical reasons to distinguish urges from the emotional components of withdrawal. First, urges and withdrawal-related affectivity appear to be associated with different physiologic substrata (e.g., Curtin, McCarthy, Piper, & Baker, 2006). Moreover, urges show different trajectories in response to drug removal and environmental events (McCarthy, Piasecki, Fiore, & Baker, in press). Finally, as we shall review momentarily, urges appear to exert their own distinct motivational influences. EXTRACTING MEANING FROM COMPLEX WITHDRAWAL PROFILES As noted earlier, most studies of withdrawal have assumed a standard pattern across time (waveform) for all symptoms and signs. This was used, either implicitly or explicitly, to justify simplistic measurement strategies. Researchers often used only a single measure of peak or average withdrawal, collapsing all symptoms together, to reflect the potentially meaningful information. Interviews with addicted individuals, however, indicate that they experience strong urges and negative affect many weeks after discontinuing drug use. This suggests that withdrawal patterns should be assessed in a more comprehensive manner. Therefore, we measured profiles of withdrawal symptoms, especially urges and negative affect, so as to capture their average elevation, trajectories (e.g., whether symptoms are worsening or improving), rise times (how quickly symptoms increase following abstinence), durations, and reactivity to stressors and environmental events. Waveforms of urges and affective symptoms show dramatic differences from one person to the next and possess motivational relevance (see Fig. 1; McCarthy, Piasecki et al., in press). When researchers measure withdrawal in a way that captures this variability, strong relations with smoking relapse are obtained. For instance, relapse to smoking is consistently and powerfully Volume 15—Number 5 233 Timothy B. Baker et al.