An approach that has gained widespread application in the treatment of substance abuse is cognitive-behavioral therapy (CBT). Its origins are in behavioral theory, focusing on both classical conditioning and operant learning; cognitive social learning theory, from which are taken ideas concerning observational learning, the influence of modeling, and the role of cognitive expectancies in determining behavior; and cognitive theory and therapy, which focus on the thoughts, cognitive schema, beliefs, attitudes, and attributions that influence one's feelings and mediate the relationship between antecedents and behavior. Although there are a number of similarities across these three seminal perspectives (see Carroll, 1998), each has contributed unique ideas consistent with its theoretical underpinnings. However, in most substance abuse treatment settings, the prominent features of these three theoretical approaches are merged into a cognitive-behavioral model.
Before focusing more specifically on the cognitive-behavioral model, this chapter examines the behavioral and cognitive theories and therapies that serve as the foundations of and have contributed significantly to the cognitive-behavioral approach to substance abuse treatment. Both behavioral and cognitive theories have led to interventions that individually have been proven effective in treating substance abuse. Several of these are reviewed, as they have been successfully incorporated into an integrated cognitive-behavioral model of addictive behaviors and their treatment.
In contrast to many other methods, behavioral approaches to the treatment of substance abuse have substantial research evidence in support of their effectiveness. Two recent comprehensive reviews of the treatment research literature offer strong evidence for their effectiveness (Holder et al., 1991; Miller et al., 1995). However, some critics argue that this is because behavioral approaches have been developed under controlled conditions and that in "real" therapy there are many more variables at work than can be measured in controlled experiments. Providers should take advantage of the wide range of behavioral therapy techniques that are available. These techniques can be conducted successfully in individual, group, and family settings, among others, to help clients change their substance abuse behaviors.
Behavioral approaches assume that substance abuse disorders are developed and maintained through the general principles of learning and reinforcement. The early behavioral models of substance abuse were influenced primarily by the principles of both Pavlovian classical conditioning and Skinnerian operant learning (O'Brien and Childress, 1992; Stasiewicz and Maisto, 1993). (See Figure 4-1 for definitions of classical conditioning and operant learning.)
Today, behavioral therapy for the treatment of substance abuse disorders is based primarily, though not exclusively, on methods derived from both operant and classical theories of learning. A major tenet of behavioral therapy is that because substance abuse is a learned behavior pattern, changing the reinforcement contingencies that govern this behavior can modify it. This goal can be achieved by focusing on either the classically conditioned craving responses or on the operant reinforcement patterns that are assessed as maintaining the substance abuse. More specifically, the classically conditioned response can be addressed either through extinction or counterconditioning procedures; the operant responses can be targeted through contingency management or coping skills training. (More information about the basic assumptions of behavioral theories concerning substance abuse disorders is contained in Figure 4-2 .)
According to behavioral theory, changes in behavior come about through learning new behaviors. Because substance abuse behavior is learned, it can be changed by teaching the client more adaptive, alternative behaviors aimed at achieving the same rewards. Figure 4-3 provides an overview of some of the advantages of behavioral theories of substance abuse and dependence and their treatment.
By its very design, most behavioral therapy is brief. The aim is not to remake personality, but rather to help the client address specific, identifiable problems in such a way that the client is able to apply the basic techniques and skills learned in therapy to the real world, without the assistance of the therapist. Behavioral therapy focuses more on identifying and changing observable, measurable behaviors than other therapeutic approaches and hence lends itself to brief work. Treatment is linked to altering the behavior, and success is the change, elimination, or enhancement of particular behaviors.
Regular assessment and measurement of progress are integral to effective behavioral therapy. Decisions about the length of treatment are made on the basis of these assessments, rather than according to a formula or theoretical assumption about how long therapy should take. Each individual is approached as a unique case, albeit one to which broad principles can be applied.
A principal of classical conditioning is that if a behavior occurs repeatedly across time but is not reinforced, the strength of both the cue for the behavior and the behavior itself will diminish and the behavior will extinguish. This principal has been the foundation of behavioral treatments known as "cue exposure" (O'Brien et al., 1990; Rohsenow et al., 1991; Rohsenow and Monti, 1995). Even after relatively long periods of abstinence from substances, being placed in situations that have physical-environmental, social, or emotional cues associated with past substance abuse will elicit strong physiological arousal reactions and reports of strong sensations of craving. In cue exposure, a client is purposefully presented with such cues physically (e.g., by showing his personal drug paraphernalia or by accompanying him into a well-frequented bar), or visually through video depiction of a drug-using scenario or through visualization of such a scenario. However, the client is prevented from drinking or taking drugs. This extinction process, over time, leads to a decreased reactivity to such cues.
O'Brien and colleagues found that cocaine-dependent clients showed the prototypical arousal and craving responses when first presented drug-related cues that reminded them of their drug use (O'Brien et al., 1990). Clients then began the cue-extinction protocol. By the sixth 1-hour treatment session, they no longer reported either subjective highs or physiological withdrawal. By the 15th session, all clients reported that they no longer experienced craving when presented with the drug-related cues. Clients who received the cue exposure as part of their standard outpatient treatment for cocaine use were also less likely to drop out of treatment and had more cocaine-free weeks than did clients attending the same outpatient program but who did not receive cue exposure.
Another method used to modify behavior according to classical conditioning principles is to make behaviors that had been associated with positive outcomes less appealing by more closely associating them with negative consequences. By repeatedly pairing those cues that previously elicited a particular behavior with negative rather than positive outcomes, the cues lose their ability to elicit the original classically conditioned response; instead, they elicit a negative outcome. This has led to the development of what have been described as aversive conditioning or counterconditioning treatment approaches (Howard et al., 1991; Rimmele et al., 1995). These procedures repeatedly pair negative outcomes with the substance-related cues previously associated with the positive consequences of substance use.
For example, the Shick-Shadel Hospital in Seattle uses aversive conditioning techniques with alcohol-dependent clients (Lemere, 1987). Before a treatment session, the client is asked to drink a warm saline solution and is given an emetic medication that will ultimately lead the client to become nauseated and to vomit. The client is then brought into "Duffy's Bar," a room filled with vivid alcohol- and drinking-related posters, a bar with bottles of a large number and wide range of alcoholic beverages, and other drinking-related cues. The room is meant to highlight and make more salient the cues associated with drinking. The client is asked to identify her favorite type and brand of alcohol. After pouring a drink, she is asked to swirl the alcohol around in the glass, to smell the alcohol, to place the glass to her lips and taste, and then to begin to take a sip of the drink. At that point, as she is about to take a drink, the effects of the emetic drug "kick in" and the client becomes nauseated and vomits. Over repeated sessions, which occur every other day for a 10-day period, the alcohol-related sight, smell, and taste cues not only do not elicit craving and positive feelings about drinking, but rather they now elicit conditioned nausea.
Therapies based on counterconditioning theory typically use chemically induced aversion or electric shock as negative consequences to be paired with the substance-related cues. Visual imagery can also be used in a technique called covert sensitization. In this procedure, the client is asked to imagine as vividly as possible a sequence of events that begin by seeing his favorite bar; this is typically accompanied by increased craving. As the person proceeds further in imagining entering the bar, sitting down, ordering a drink, and so on, the initial sense of craving shifts to mild discomfort. As he visualizes beginning to take a drink and tastes the alcohol, he is then asked to imagine becoming violently sick and vomiting (Rimmele et al., 1995).
While aversive conditioning procedures have most often been used in the treatment of alcohol dependence, they have also been applied to the treatment of marijuana and cocaine use (Frawley and Smith, 1990; Smith et al., 1988). It should be noted that these aversive conditioning techniques, as well as cue exposure approaches, are best viewed as components of a more comprehensive treatment program rather than as independent, free-standing treatments (O'Brien, et al., 1990; Smith and Frawley, 1993). In this context, Smith and colleagues reported positive outcomes for dependent users of both alcohol and cocaine who received chemical aversion procedures as part of their treatment in comparison to those who did not receive similar treatment (Frawley and Smith, 1990; Smith et al., 1997). Rimmele and colleagues also recommended covert sensitization as a highly effective and portable treatment component which, unlike chemical or electric aversion therapies, can be used at any time and in any setting as a self-control strategy (Rimmele et al., 1995).
A number of substance abuse treatment strategies have derived from operant learning principles. While they are often incorporated into broad-spectrum cognitive-behavioral approaches, they have also been used as independent forms of treatment. Common elements of behavioral treatments based on theories of operant learning include contingency management, behavior contracting, community reinforcement, and behavioral self-control training. The following sections describe some of the elements used in brief behavioral therapies based on the operant learning model.
In contingency management approaches, an active attempt is made to change those environmental contingencies that can influence substance abuse behavior (Higgins et al., 1998). The goal is to decrease or stop substance use and to increase behaviors that are incompatible with use. In particular, those contingencies that are found through a functional analysis (see Figure 4-4 ) to prompt as well as reinforce substance abuse are weakened by associating evidence of substance use (e.g., a drug-positive urine screen) with some form of negative consequence or punishment. Contingencies that prompt and reinforce behaviors that are incompatible with substance abuse and that promote abstinence are strengthened by associating them with positive reinforcers.
One recent study evaluated the effects of a voucher program in the treatment of methadone-maintained opiate addicts with a history of cocaine use (Silverman et al., 1998). Clients who provided cocaine-free urine samples received vouchers that had monetary value. The value of the vouchers increased as the number of consecutive cocaine-free urine samples increased. Clients in the contingent voucher condition, compared to those who received vouchers on a noncontingent basis, reported decreased craving for cocaine and significantly increased cocaine abstinence. A more general positive treatment effect was also noted, with clients in the contingent voucher condition also demonstrating an increased abstinence from opiates.
Chutuape and colleagues have also shown that providing methadone take-home privileges contingent on drug-free urine samples among methadone clients with persistent multiple drug abuse resulted in marked reductions in drug use (Chutuape et al., 1999). Nearly 25 percent of clients in the take-home incentive program met the criterion for marked reduction in drug use and also were significantly more likely to achieve the criterion of having 4 consecutive weeks of drug-free urine samples. None of the clients in a control condition (no take-home privileges) met these criteria. Whereas only 2 percent of the control group evidenced a decrease in the frequency of drug-positive urines, clients in the incentive program decreased use between 14 and 18 percent.
In addition to increasing drug abstinence, similar voucher systems have been effective in maintaining attendance of methadone clients at a job-skills training program (Silverman et al., 1996). However, in contrast to drug treatment, less evidence is available concerning the effectiveness of such contingency management approaches in the treatment of alcohol problems (Higgins et al., 1998).
Attempts to incorporate real-world contingencies into treatment programs are increasing (Higgins, 1999). Clearly, programs can build contingencies such as take-home medication privileges into the structure of their programs. Milby and colleagues provide an example of a contingency management system incorporated into treatment that is more relevant to real-life situations of users (Milby et al., 1996). In this study, homeless substance abusers were enrolled in an intensive day treatment program. A subgroup of these clients was also involved in a contingent work therapy and housing program. As long as the clients remained substance free, they were able to remain in the work program and remain in the therapeutic housing; if they were found to be drinking or using drugs, they became ineligible for both the job training/work program and housing. Clients involved in the abstinence-contingent program had fewer cocaine-positive urine samples, fewer days of drinking, fewer days of homelessness, and more days of employment during the followup period than those in the standard treatment.
Naturalistic contingencies may also be useful in treatment. These contingencies include threatened loss of job, spouse, or driver's license and were positively related to treatment outcome among alcohol users (Krampen, 1989). However, the prognosis was less favorable in those patients who had already experienced a loss in one of those areas because the contingency no longer existed for them.
Higgins and colleagues noted that written contracts may be used to help implement a contingency management program (Higgins et al., 1998). The contract should specify clearly, using the client's own words, the target behavior to be changed, the contingencies surrounding either changing behavior or not, and the timeframe in which the desired behavior change is to occur. The act of composing and signing a contract is a small but potentially important ritual signifying the client's commitment to the proposed change. In the contract, the client may include contingencies, especially rewards or positive incentives that will reinforce target behaviors (e.g., attending treatment sessions, getting to 12-Step meetings, avoiding stimuli associated with substance use). Goals should be clearly defined, broken into small steps that occur frequently, and revised as treatment progresses; contingencies should occur quickly after success or failure.
Most often, behavioral contracts and contingency management procedures are embedded in a more comprehensive treatment program. Contracts targeting goals supportive of recovery (e.g., improving vocational behavior, saving money, being prompt for counseling, regularly taking medication) are generally more likely to be achieved and lead to better outcomes than those more directly related to substance use (e.g., clean urine samples) (Anker and Crowley, 1982; Iguchi et al., 1997; Magura et al., 1987, 1988). For instance, research found that receiving vouchers contingent on completing objective, individually tailored goals related to one's overall treatment plan was more effective in reducing substance abuse than either a voucher system specifically targeting drug-free urine samples or a standard treatment without either of these contingency contracts added (Iguchi et al., 1997). The effectiveness of such contracts also appears to be linked to the severity of the consequences that might result from a broken contract (Magura et al., 1987).
Behavioral contracting and contingency management are often found as elements in a number of more comprehensive approaches such as community reinforcement and behavioral self-control training.
The community reinforcement approach (CRA) was developed as a treatment for alcohol abuse disorders (Azrin, 1976; Hunt and Azrin, 1973). After a period during which it appears to have been little used, it has received increased interest as a behavioral approach to substance abuse (Higgins et al., 1998; Meyers and Smith, 1995; Smith and Meyers, 1995). CRA is a broad-spectrum approach based on the principles of operant learning, the goal of which is to increase the likelihood of continued abstinence from alcohol or drugs by reorganizing the client's environment. In particular, CRA attempts to weaken the influence of reinforcement received by substance abuse and its related activities by increasing the availability and frequency of reinforcement derived from alternative activities, particularly those vocational, family, social, and recreational activities that are incompatible with substance abuse (Higgins et al., 1998).
A goal of CRA is to make these alternative interpersonal and social sources of reinforcement available when the person is sober or drug-free, but to make them unavailable if the person drinks or uses. The program consists of a number of components, and it can be tailored to the specific circumstances of a client. Vocational counseling and job clubs can improve clients' basic skills as well as job-seeking skills (e.g., résumé development, application completion, job interview skills). Social and recreational counseling is provided to help clients learn about and sample a number of substance-free recreational pursuits and social activities. In some cases, social clubs have been established to provide clients with a substance-free environment where they can gather and have fun.
For those clients who are married or in a relationship, marital counseling and communication skills training are provided to enhance the quality of the relationship and reduce the stress of substance-related arguments. Couples are trained to give each other positive attention through compliments, appreciation, affection, and offers to help. A focus is placed on clarifying expectations that each partner has about the behavior of the other. For those with a problem with alcohol, medication (e.g., disulfiram [Antabuse]) monitored by the spouse may be used. The client also receives training in problemsolving and in ways to refuse requests to drink or use drugs.
CRA has been described as a promising but underutilized treatment for alcohol abuse (McCrady, 1991). A review of the alcohol treatment outcome literature identifies CRA among those interventions having the greatest empirical support (Miller et al., 1995). CRA's application to substances other than alcohol also appears to have been successful (Higgins et al., 1998). This extension is exemplified by the recent publication of a detailed CRA therapy manual for the treatment of cocaine dependence by the National Institute on Drug Abuse (Budney and Higgins, 1998). This manual relies heavily on the early work of Higgins and colleagues in evaluating the effectiveness of combining CRA with contingency management approaches (e.g., use of vouchers for drug-free urine samples) in the treatment of cocaine dependence (Higgins et al., 1991, 1993). In comparison to standard outpatient treatment, clients in the CRA-plus-vouchers condition remained in treatment longer, had more continuous weeks of drug-free urine samples, and had greater amounts of cocaine abstinence even at a 12-month followup. A similar pattern of findings has been obtained with methadone-maintained opiate addicts (Abbott et al., 1998).
The CRA model has been modified into the Community Reinforcement and Family Training procedure (CRAFT) (Meyers et al., 1996). The client's significant others and family members, who are an integral part of this approach, receive training in behavior modification and enhancing motivation. CRAFT seeks to reduce or stop substance abuse by working through nonusing family and friends. While CRA involves family or significant others in treatment, CRAFT is more of a form of family therapy (rather than individual therapy) and therefore is discussed in Chapter 8 of this TIP.
In contrast to CRA, which incorporates a wide array of individuals in the treatment process, the behavioral self-control training approach focuses on the substance abuser and his attempts to reduce or stop substance abuse either on his own or with the aid of a therapist (Hester, 1995; Hester and Miller, 1989). The goal of this approach is either moderation and harm reduction or abstinence. As applied to alcohol problems, the approach consists of the eight sequential steps listed below (Hester, 1995):
Although a therapist may guide the individual in a behavioral self-control model, the substance abuser maintains primary responsibility for changing his behavior. During the course of therapy, the client and therapist meet in brief sessions to go over homework and ensure that the client is following through. Rather than involvement with a therapist, the person may be guided instead by self-help manuals (Miller and Munoz, 1982; Sanchez-Craig, 1995), intervention via correspondence (Sitharthan et al., 1996), or even a computer program (Hester and Delaney, 1997).
McCrady also included behavioral self-control training as another promising but underutilized treatment approach (McCrady, 1991). Hester indicated that there is good empirical support for behavioral self-control training in achieving the goal of moderate, nonproblematic drinking (Hester, 1995). In randomized clinical trials, problem drinkers assigned to behavioral self-control with a goal of either moderation or abstinence typically have comparable long-term outcomes. Although behavioral self-control approaches have been used primarily with alcohol problems, they have also been used with other substances such as opiates (van Bilsen and Whitehead, 1994).
Behavioral therapies are often delivered using a specific manual, but they are also adaptable to the individual client. A number of the behavioral techniques described here are also used by therapists using cognitive-behavioral therapy. The following sections describe how brief behavioral therapy might be applied at different stages of treatment. Some of the techniques developed for brief behavioral therapy are also presented.
The initial session in brief behavioral therapy involves an exploration of the reasons the client is seeking treatment at this particular time; the extent to which this motivation for treatment is intrinsic, rather than influenced by external sources; the areas of concern that the client and significant others may have about his substance abuse; the situations in which she drinks or uses excessively; and the consequences she experiences (both positive and negative, as well as proximal and removed from the actual substance abuse). This involves an abbreviated functional analysis. (See the section with that name later in this chapter.)
The information gained in the session will assist the counselor in determining the antecedents that prompt substance abuse and the reinforcers that appear to maintain it. Based on the information obtained, the counselor can begin to formulate a treatment plan with respect to the specific target behaviors to address, the behavioral interventions that address these target behaviors most effectively, and behaviors incompatible with heavy drinking that should be reinforced and targeted for an increase in frequency.
During the initial session, the therapist should note the most salient problems identified by the client and intervene with them first. The therapist also should assess the client's readiness to change and then develop initial behavioral goals in collaboration with the client. For substance abuse disorders, these goals will, of course, involve a reduction in or cessation of substance use. In addition to targeting substance abuse as the primary focus, other goals will be developed to assist the client in improving daily functioning (e.g., by reducing stress, as described in Figure 4-5 ). The focus of the therapy might be to negotiate with the client to accomplish these other goals by reducing use. The therapist will continue to engage the client in a collaborative process in which they determine those problems to target, their relative priority, and ways to resolve them.
Near the end of the initial session the therapist reviews with the client the procedure for filling out the self-monitoring records. In addition, the therapist might provide the client with self-help manuals that outline the specific steps in the behavioral self-control process. Self-monitoring of substance abuse behavior is one form of written homework common in behavioral approaches; other types of homework might also be used. Homework assignments can include such things as keeping a journal of behaviors, activities, and feelings when using substances or at risk of doing so. In the brief behavioral model designed by Phillips and Weiner, techniques such as programmed therapy and writing therapy (see Figure 4-6 ) make what is typically thought of as "homework" the central concern of the therapy session (Phillips and Weiner, 1966).
Based on a review of the information collected through self-monitoring, subsequent sessions involve negotiation about treatment goals. While many problem drinkers, for example, choose a moderation goal, across time those with more severe problems shift to a goal of abstinence (Hodgins et al., 1997). Later sessions might also consider the introduction of cue exposure training or relapse prevention targeted at substance abuse above a particular level. These behavioral techniques have been incorporated into more comprehensive behavioral self-control approaches, even those with an abstinence goal (Larimer and Marlatt, 1990; Sitharthan et al., 1997). The decision to implement such interventions will be guided by the client's continued self-monitoring, which the client and counselor review at each session.
Brief behavioral therapy might also involve the client's spouse or significant others, who may attend several of the therapy sessions. In addition to serving as a corroborator of the client's self-reported substance use, a significant other may be involved in behavioral contracting and community reinforcement interventions. The significant other could be taught to positively reinforce a client's reduced drinking or abstinence and not to argue with her drinking when she is intoxicated, but rather to approach her when she is sober and provide positive feedback. The client and the significant other may develop a contingency contract that will encourage reinforcement of her positive behaviors.
Cognitive theory assumes that most psychological problems derive from faulty thinking processes (Beck and Wright, 1992; Beck et al., 1993; Beck and Liese, 1998; Ellis, 1982; Ellis et al., 1988). The diagram in Figure 4-7 illustrates the three bidirectional components of this theory: (1) cognitions or thoughts, (2) affect or feelings, and (3) behavior. While cognitive theory owes a debt to the behavioral model, the differences are apparent. Unlike behavioral models that focus primarily on observable behaviors, cognitive theory views antecedent events, cognitions, and behavior as interactive and dynamic, as indicated by the double-headed arrows (depicted in Figure 4-7). Each of these components is capable of affecting the others, but the primary emphasis is placed on cognition. The way we act and feel is most often affected by our beliefs, attitudes, perceptions, cognitive schema, and attributions. These cognitive factors serve as a template through which events are filtered and appraised. To the extent that our thinking processes are faulty and biased, our emotional and behavioral responses to what goes on in our life will be problematic. According to this theory, changing the way a client thinks can change the way he feels and behaves.
Cognitive theory was developed by A.T. Beck as a way of understanding and treating depression but has since been applied to numerous other mental health issues including substance abuse disorders. Beck believed that depressed clients held negative views of themselves, the world, and their future, and that these negative views were the real causes of their depression. He found that their psychological difficulties were due to automatic thoughts, dysfunctional assumptions, and negative self-statements. Automatic thoughts often precede emotions but occur quite rapidly with little awareness; consequently, individuals do not value them highly. For example, depressed people address themselves in highly critical tones, blaming themselves for everything that happens. Figure 4-8 is a list of 15 common cognitive errors found in the thinking processes of individuals with emotional and behavioral problems, including substance abuse disorders. An overview of the nature and content of distorted thinking more specifically associated with substance abuse is provided in Figure 4-9 (Ellis et al., 1988). These thoughts are presumably automatic, overlearned, rigid and inflexible, overgeneralized and illogical, dichotomous, and not based on fact. They also tend to reflect reliance on substances as a means of coping with boredom and negative emotions, a negative view of the self as a person with a substance abuse problem, and a tendency to facilitate continued substance use.
Such negative thoughts and irrational beliefs have been found to be associated with substance abuse disorders. Problem avoidance, dwelling on negative events, holding a negative outlook on the world and on one's future, and avoidance of responsibility have been associated with the development of patterns of substance abuse and urges to drink among individuals with alcohol problems (Butterfield and Leclair, 1988; Denoff, 1988; Rohsenow et al., 1989). Rohsenow and associates found that irrational beliefs--particularly feeling doomed about the past--were predictive of both the frequency of drinking and the average quantity of alcohol consumed following substance abuse treatment (Rohsenow et al., 1989).
Given the view that dysfunctional behavior, including substance abuse, is determined in large part by faulty cognitions, the role of therapy is to modify the negative or self-defeating automatic thought processes or perceptions that seem to perpetuate the symptoms of emotional disorders. Clients can be taught to notice these thoughts and to change them, but this is difficult at first. Cognitive therapy techniques challenge the clients' understanding of themselves and their situation. The therapist helps clients become more objective about their thinking and distance themselves from it when recognizing cognitive errors or faulty logic brought about by automatic thinking.
Treatment, therefore, is directed primarily at changing distorted or maladaptive thoughts and related behavioral dysfunction. Cognitive restructuring is the general term given to the process of changing the client's thought patterns. Figure 4-10 shows a number of distorted addictive thoughts and more rational alternatives that the therapist might help develop and practice over the course of cognitive restructuring.
Once a specific faulty thought is identified, the therapist will challenge a client to look at alternative ways of seeing the same event. Whenever a client has difficulty changing a perception, the therapist can give him homework to test the truth of his cognitions. If, for example, a client insists that his boss hates him, the therapist can ask him to verify this with an assignment: "Ask your coworkers if your boss treats them the same way he treats you." Figure 4-11 gives an example of how a thought leads to a feeling and then to a behavior.
Once the maladaptive thoughts are discovered in a person's habitual, automatic thinking, it becomes possible to modify them by substituting rational, realistic ideas for the distorted ones to create a happier and healthier life without substance abuse.
The approach developed by Beck and colleagues to achieve the goal of a substance-free life is referred to as cognitive therapy (Beck et al., 1993; Beck and Liese, 1998), while Ellis' approach is known as rational-emotive therapy (Ellis et al., 1988). Generally, the therapist takes a more active role in cognitive therapy than in other types of therapy, depending on the stage of treatment, severity of the substance abuse, and degree of the client's cognitive capability.
While Ellis and Beck have similar views about the prominent role that cognitions play in the development and maintenance of substance abuse disorders, their theories differ in considering how the therapist should treat irrational or maladaptive cognitions. Rational-emotive therapy is often more challenging and confrontative, with the therapist informing the client of the irrationality of certain types of beliefs that all people are prone to. Beck, on the other hand, believes that the cognitive therapist, using a supportive Socratic method, should enlist the client in carefully examining the accuracy of her beliefs. Thus, Beck places more importance on the client's own discovery of faulty and unproductive thinking, while Ellis believes that the client should simply be told that these exist and what they are. Nevertheless, there is substantial overlap in both the theory and practice of these two therapies. Clearly, different clients will have different responses to these qualitatively different approaches to modifying their thoughts and beliefs.
Therapeutic work in cognitive therapy is devoted primarily, although not exclusively, to addressing specific problems or issues in the client's present life, rather than global themes or long-standing issues. At times, however, it is important to understand the connection between the origins of a set of cognitions and the client's current behavior. Such an understanding of how the individual got to the present emotional and behavioral state is often essential to understanding the mechanism of change. The client's attention to current problems is intended to promote her development of a plan of action that can reverse dysfunctional thought processes, emotions, and behavior--such as avoidance of problems or feelings of helplessness. Clients are enlisted as coinvestigators or scientists who study their own thought patterns and associated consequences.
Cognitive therapy can be useful in the treatment of substance abuse disorders in several ways. When distorted or unproductive ways of thinking about daily life events lead to negative emotional states that then promote substance use, cognitive therapy can be used to alter the sequence by targeting and modifying the client's thoughts. When clients limit their options for coping with stress by rigid or all-or-nothing thinking (e.g., "nothing will help me deal with this problem but a drink"), cognitive therapy can help them explore alternative behaviors and attitudes that do not involve the use of substances. In addition, cognitive therapy can help the client develop healthier ways of viewing both his history of substance abuse and the meaning of a recent "slip" or relapse so that it does not inevitably lead to more substance abuse.
Cognitive therapy works under the assumption that a client can be educated to approach his problems rationally. Because of this emphasis on rational understanding, the cognitive therapist will typically begin therapy by explaining the nature of her approach (see Figure 4-12 for a sample opening script).
In the opening session of cognitive therapy, the therapist will assess the client's view of his problems and their causes. The therapist pays careful attention to the meaning the client assigns to significant events and how that meaning is related to subsequent feelings and unwanted behavior. In the middle to late phases of the first session, the therapist will emphasize the collaborative aspect of the therapy process and introduces the cognitive model to the client. There are three major steps in this process:
Cognitive therapy tends to follow a standard within-session structure to make the maximum use of time, to focus on the most important current problems, to set the tone for a working atmosphere, and to maintain continuity between sessions. Beck structures sessions into eight elements, listed below, which he describes in greater detail (Beck et al., 1993):
Cognitive therapy adheres to the basic goals of planned brief therapy, but treatment times can vary. It typically lasts from 12 to 20 weeks, with the client and therapist meeting once per week. (Freeman et al., 1990). However, it can be conducted in less time--for instance, once per week for six to eight sessions. The number of sessions will depend on the nature of the problem.
Because cognitive therapy is usually planned for comparatively short treatment times, there has not been much research to study the relative effectiveness of longer term cognitive therapy. However, Lyons and Woods in their meta-analysis of 70 different rational-emotive therapy studies found that increased effects correlated with longer treatment times (Lyons and Wood, 1991). More research needs to be conducted looking at the effect of treatment duration on the efficacy of these therapies.
In a brief version of this therapy, there is less time to understand and restructure all of the cognitions that may be influencing substance abuse. The therapist must use the early sessions to determine the most productive focus of the therapy, given the short timeframe. If the client used substances primarily to cope with negative mood states, then therapy may focus on understanding how the client's interpretation of events led to the negative moods. Restructuring these thought processes may help decrease reasons for substance abuse. Alternatively, if the client drinks largely to party and have a good time with friends, a focus on expected effects may lead to the client's gaining greater awareness of negative consequences and, perhaps, a reduced association of the substance with positive experiences. If the client is returning to therapy after a period of sobriety that ended in relapse, a focus on the circumstances leading to relapse and other resulting consequences may shape the therapy.
A number of specific cognitive therapy techniques may be appropriate for use, depending on the phase of treatment and the issues raised by the client. Cognitive interventions can be introduced at any point throughout the treatment process, whenever the therapist feels it is important to examine a client's inaccurate or unproductive thinking that may lead to the risk of substance abuse. They also can be used episodically with clients who leave and then return to treatment or during aftercare or continuing care following a more intensive treatment episode.
Periods without therapy sessions allow clients time to practice the new skills of identifying and challenging unproductive thinking on their own. However, it is easy to fall back into old, automatic ways of thinking that may require a return to therapy. The therapist can productively build on what was learned in previous sessions, help the client see how she slipped into old patterns, and further reinforce the process of catching oneself in the process of thinking negative automatic thoughts. The therapist must be prepared to move from topic to topic while always adhering to the major theme--that how the client thinks determines how the client feels and acts, including whether the client abuses substances.
Cognitive therapy can be quite successful as an option for brief therapy for several other reasons (Carroll, 1996a):
Early behavioral theories of substance abuse were nonmediational in nature (Donovan and Marlatt, 1993). They focused almost exclusively on overt, observable behaviors, and it was believed that understanding the antecedents and reinforcement contingencies was sufficient to explain behavior and to modify it. Over time, however, these behavioral theories began to incorporate cognitive factors into their conceptualizations of substance abuse disorders. These more recent models are mediational in nature; that is, a greater role is attributed to the interaction among a variety of individual difference variables such as beliefs, values, perceptions, expectations, and attributional processes in mediating the development and continuation of substance abuse disorders (Abrams and Niaura, 1987; Mackay and Donovan, 1991; Marlatt et al., 1988; Marlatt and Donovan, 1981). This expanded, mediational model has been described as cognitive social learning or cognitive-behavioral theory. This theory postulates that cognitive factors mediate all interactions between the individual, situational demands, and the person's attempts to cope effectively.
Cognitive-behavioral theory represents the integration of principles derived from both behavioral and cognitive theories, and it provides the basis for a more inclusive and comprehensive approach to treating substance abuse disorders. However, a broader range of cognitions is included in cognitive-behavioral theory than had been involved in earlier versions of cognitive theory. These include attributions, appraisals, self-efficacy expectancies, and substance-related effect expectancies. Each of these will be reviewed briefly below. Common elements of brief cognitive-behavioral therapy are listed in Figure 4-13 .
An attribution is an individual's explanation of why an event occurred. Abramson and colleagues proposed that individuals develop attributional styles (i.e., individual ways of explaining events in their lives that can play a role in the development of emotional problems and dysfunctional behaviors) (Abramson et al., 1978). The basic attributional dimensions are internal/external, stable/unstable, and global/specific. For instance, clinically depressed persons tend to blame themselves for adverse life events (internal), believe that the causes of negative situations will last indefinitely (stable), and overgeneralize the causes of discrete occurrences (global). Healthier individuals, on the other hand, view negative events as due to external forces (fate, luck, environment), as having isolated meaning (limited only to specific events), and as being transient or changeable (lasting only a short time). Figure 4-14 lists and further defines the three dimensions of attribution that make up an "attributional style."
Attributional styles play a major role in the cognitive-behavioral theory of substance abuse disorders (Davies, 1992; Marlatt and Gordon, 1985). The nature of substance abusers' attributional styles is thought to have considerable bearing on their perception of their substance abuse problem and their approach to recovery. An alcohol-dependent client, for instance, may believe that he drank because he was weak (an internal attribution) or because he was surrounded by people encouraging him to have a beer (an external attribution). He may believe that his failure to maintain abstinence shows that he is a weak person who can never succeed at anything (a global attribution) or that a drinking episode does not represent a general weakness, but was instead due to the specific circumstances of the moment (a specific attribution). He may believe that the cause of his slip is something he cannot change (a stable attribution) or that the next time, he will catch himself and exert better coping responses (an unstable attribution). Whereas the internal, global, and stable attribution for the use of alcohol is likely to lead to feelings of hopelessness and a return to drinking, the external, specific, unstable attribution is likely to lead to greater efforts to cope with similar situations in the future.
Marlatt and Gordon described a negative attributional process that can occur after a slip (the first use of a substance after a period of abstinence) and that may lead to continued use in a full-blown relapse (Marlatt and Gordon, 1985). This process, known as the abstinence violation effect (AVE), involves the attribution of the cause of an initial slip to internal, stable, and global factors. These clients may believe that they are hopeless addicts and failures, that they will never be able to achieve and maintain sobriety, and that there is no use in trying to change because they think that they cannot succeed.
AVE also has an emotional component associated with it. Substance abusers who have slipped and have internal, stable, and generalized attributions will feel depressed, worthless, helpless, and hopeless. This attributional style tends to be associated with a form of "learned helplessness" that is perpetuated by the substance users' distorted perceptions. Together, the sense of helplessness and the negative emotional state increase the likelihood that the initial lapse will develop into a full-blown relapse. Research with individuals dependent on alcohol, marijuana, opiates, and other illicit drugs, provides empirical support for the attributional style hypothesized to mediate the AVE (Birke et al., 1990; Bradley et al., 1992; Reich and Gutierres, 1987; Stephens et al., 1994; Walton et al., 1994).
For the cognitive-behavioral therapist, an individual's appraisal of stressful situations and his ability to cope with the demands of these situations are important influences on the initiation and maintenance of substance abuse, as well as relapse after cessation of use (Hawkins, 1992; Marlatt and Gordon, 1985; Shiffman, 1987, 1989; Wills and Hirky, 1996).
Folkman and Lazarus described two different levels of cognitive appraisal (Folkman and Lazarus, 1988, 1991). The first level is a primary appraisal. This represents the individual's perception of a situation and an estimation of the potential level of stress, personal challenge, or threat involved with the situation. Secondary appraisal represents the individual's evaluation of her ability to meet the challenges and demands specific to the situation. This secondary appraisal, which will be influenced by the extent, nature, and availability of the individual's coping skills, further mediates the individual's perception of stress and the person's emotional response.
To the extent that the individual senses that she has the necessary behavioral, cognitive, or emotional coping skills to meet the challenges of the situation, it will be appraised as less threatening or stressful. Conversely, if the person judges that the necessary coping skills are lacking, the situation is viewed as more threatening and stressful, and the person is likely to be frightened, anxious, depressed, or helpless. The results of Smith and colleagues suggest that such cognitive appraisals may play a more prominent role than attributions in mediating emotional responses to potentially threatening situations (Smith et al., 1993).
In substance use-related situations, coping "refers to what an individual does or thinks in a relapse crisis situation so as to handle the risk for renewed substance use" (Moser and Annis, 1996, p. 1101). Cognitive-behavioral theory posits that substance users are deficient in their ability to cope with interpersonal, social, emotional, and personal problems. In the absence of these skills, such problems are viewed as threatening, stressful, and potentially unsolvable. Based on the individual's observation of both family members' and peers' responses to similar situations and from their own initial experimental use of alcohol or drugs, the individual uses substances as a means of trying to deal with these problems and the emotional reactions they create. From this perspective, substance abuse is viewed as a learned behavior having functional utility for the individual--the individual uses substances in response to problematic situations as an attempt to cope in the absence of more appropriate behavioral, cognitive, and emotional coping skills.
A number of dimensions are involved in the coping process as it relates to substance abuse (Donovan, 1996; Hawkins, 1992; Lazarus, 1993; Shiffman, 1987; Wills and Hirky, 1996). The first is the general domain in which the coping response occurs. Coping responses can occur within the affective, behavioral, and cognitive domains. Litman identified a number of behavioral and cognitive strategies that are protective against relapse (Litman, 1986). There are two behavioral classes of coping behavior: (1) basic avoidance of situations that have been previously associated with substance abuse and (2) seeking social support when confronted with the temptation to drink or use drugs.
The cognitive domain also includes two general categories of coping: (1) negative thinking, or thinking about all the negative consequences that have resulted from substance abuse and a desire to no longer experience these, and (2) positive thinking, or thinking about all the benefits that are accrued by being clean and sober and not wanting to lose these. Litman suggests that these coping strategies operate in a somewhat sequential manner (Litman, 1986). Initially, when clients are attempting to initiate and stabilize abstinence from substances, they appear to rely more heavily on the behavioral strategies. As the period of abstinence increases, there appears to be a transition from predominantly behavioral strategies toward a greater reliance on cognitive methods of coping.
Coping strategies have a number of other dimensions. They can be emotion focused, problem focused, or avoidant. A distinction is also made between those that are general coping strategies and those that are expressly attempting to cope with urges, craving, and temptation to use in settings associated with past substance abuse. Another important dimension of coping strategies is the stage at which they are used in response to a potentially difficult substance-related situation (Shiffman, 1989). Anticipatory coping is employed as one anticipates and attempts to plan how to deal with upcoming situations. They take the form of "What can I do if..." There are also coping strategies that are employed in the moment that one is having to deal with the difficult substance-related situations. They take the form of "What can I do now..." Finally, there are restorative coping strategies that can be employed if one fails to cope and finds himself using in the situation. These take the form of "What can I do now that I've..." It is these restorative coping strategies that play a role in determining whether an initial drink or use of drugs will escalate into a full-blown relapse.
Research on coping behavior as it relates to substance abuse disorders has generally supported the basic tenet of cognitive-behavioral approaches, namely that these clients are deficient in their coping skills, that these deficiencies contribute to their continued substance abuse, and that those whose deficits are not remedied are at a greater risk of relapse than those who increase their coping through treatment (Wills and Hirky, 1996). Another study found that the number of coping attempts and the type of coping will influence both relapse and the return to abstinence (Moser and Annis, 1996). Attempting to cope with a relapse crisis led to higher rates of abstinence than not trying to cope, and the greater the number of coping strategies employed, the less likely the person was to use. If one coping response was performed, the probability of abstinence was 40 percent; the probability rose to 80 percent if two coping attempts were made. Similarly, the greater the number of coping strategies used by an individual following a relapse, the greater the likelihood of returning to abstinence. Exclusive use of active coping strategies (e.g., engaging in alternative activities that are incompatible with drinking, problemsolving, seeking support from others, thinking of consequences of using, using positive/negative self-talk) was associated with maintaining abstinence in contrast to exclusive reliance on avoidant strategies (e.g., ignoring the situation, dealing with it indirectly by eating, or relying on willpower).
Neidigh and colleagues investigated the strategies employed to cope with stress and the temptation to drink among individuals attempting to control their drinking (Neidigh et al., 1988). They found that both cognitive and behavioral coping strategies were effective in resisting a drink. Two other important findings were obtained. First, there appears to be a considerable degree of situational specificity in the coping process. That is, different types of substance-related situations seem to require different types of coping responses rather than a general coping strategy's being equally effective across situations. Second, strategies used to cope with nonspecific stress appear to be somewhat different from those used to cope with temptation. These findings suggest that treatment not only should rectify deficiencies in coping abilities, but that it may be necessary to focus on skills to deal with both general stress and substance-related temptation. Furthermore, it may be necessary to develop coping skills specific to several possible situations in which the client may use substances.
The apparent lack of coping skills among substance users is an important contributor to another key construct in cognitive-behavioral approaches, namely self-efficacy expectancies (Bandura, 1977). These expectancies refer to an individual's beliefs about his ability to successfully execute an appropriate response in order to cope with a given situation. Self-efficacy expectancies are determined in part by the individual's repertoire of coping skills and an appraisal of their relative effectiveness in relation to the specific demands of the situation. Bandura has hypothesized that expectations of personal efficacy determine whether coping behavior will be initiated or not, the amount of effort that will be expended in attempting to cope, and how long a coping attempt will continue in the face of obstacles and aversive experiences (Bandura, 1977). He also suggested that self-efficacy exerts an influence on the individual's behavior through cognitive, motivational, and emotional systems (Bandura, 1977). If a person has low self-efficacy due to a lack of necessary coping skills, she might be expected to have negative or distorted thoughts and beliefs about herself and her situation, have reduced motivation to even try to cope, and may be depressed and perceive herself as helpless.
Cognitive-behavioral approaches to substance abuse disorders postulate that low levels of self-efficacy are related to substance use and an increased likelihood of relapse after having achieved abstinence (Annis and Davis, 1988b, 1989b; DiClemente and Fairhurst, 1995; Marlatt and Gordon, 1985). A model of relapse that is based on the role of self-efficacy and coping is depicted in Figure 4-15.
Self-efficacy has been thought of as both the degree of a client's temptation to use in substance-related settings and his degree of confidence in his ability to refrain from using in those settings (Annis and Davis, 1988b; DiClemente et al., 1994; Sklar et al., 1997). The role of self-efficacy has been examined for alcohol (Evans and Dunn, 1995; Solomon and Annis, 1990), cocaine (Coon et al., 1998; Rounds-Bryant et al., 1997), marijuana (Stephens et al., 1993), opiates (Reilly et al., 1995), and across all of these substances of abuse (Sklar et al., 1997). This research generally supports the hypothesis that those with lower levels of self-efficacy are more likely to abuse substances.
As substance use is reinforced by the positive effects of the substance being taken, it is also likely that the individual will develop a set of cognitive expectancies about these anticipated effects on her feelings and behavior. They represent the individual's expectation that certain effects will predictably result from substance use. Although there has been more research on alcohol-related effect expectancies (Goldman, 1994), there has been an increased interest in drug-related expectancies (Brown, 1993). Given that drugs have differing effects, it has been necessary to develop measures specific to the effects anticipated from these different drugs, such as marijuana (Schafer and Brown, 1991) and cocaine (Jaffe and Kilbey, 1994; Schafer and Brown, 1991).
The initial focus in studying alcohol-related expectancies was on the positive effects that individuals anticipated from alcohol (Goldman and Brown, 1987). Drinkers anticipated that alcohol would serve as a global elixir, having positive effects on mood, social and interpersonal behavior, sexual behavior, assertiveness, and tension reduction. Positive effect expectancies for marijuana include relaxation and tension reduction, social and sexual facilitation, and perceptual and cognitive enhancement (Schafer and Brown, 1991). Positive cocaine-related expectancies include global positive effects, generalized arousal, euphoria, enhanced abilities, and relaxation and tension reduction (Jaffe and Kilbey, 1994; Schafer and Brown, 1991).
More recently, there has been an increased interest in the expectations of negative outcomes that individuals hold about substances. Negative expectancies about alcohol include cognitive and behavioral impairment, risk and aggression, and negative self-perception (Fromme et al., 1993). Negative consequences expected from cocaine include global negative effects, anxiety, depression, and paranoia (Jaffe and Kilbey, 1994; Schafer and Brown, 1991). It is thought that the anticipated positive effects of substances serve as an incentive or motivation to use. Conversely, negative expectancies are thought to act as a disincentive and contribute to reduced drinking or drug use (McMahon and Jones, 1993; Michalec et al., 1996).
Research supports these hypothesized actions of positive and negative expectancies (Jaffe and Kilbey, 1994; Jones and McMahon, 1994b; Rounds-Bryant et al., 1997). Positive alcohol- and cocaine-related expectancies are associated with a greater likelihood of relapse and poorer substance-related outcomes (Brown et al., 1998; Rounds-Bryant et al., 1997), whereas negative alcohol effect expectancies are related to decreased likelihood of relapse and less alcohol consumption (Jones and McMahon, 1994a; McMahon and Jones, 1996).
Research also indicates that alcohol-related effect expectancies were negatively correlated with clients' ratings of self-efficacy at the beginning of treatment (Brown et al., 1998); that is, the lower the perceived self-efficacy, the greater the level of anticipated positive effects of alcohol. Both these sets of expectancies changed over the 4-week course of treatment, with self-efficacy increasing and alcohol effect expectancies decreasing. Lower self-efficacy judgments, positive alcohol expectancies, and reliance on avoidant, emotion-focused coping strategies were significantly associated with increased alcohol consumption and alcohol-related problems among heavy drinking college students (Evans and Dunn, 1995).
Over time, with repeated exposure, aspects of a situational context (e.g., the people, places, feelings, activities) can come to serve as conditioned cues that can elicit a strong craving or desire to use. To the extent that substance abuse allows the individual to avoid or escape such problem situations or their resultant emotional reactions, the use of substances will be reinforced through operant learning. Thus the likelihood is increased that substances will be abused and will come to be relied on in the future when the individual encounters similar situations.
Marlatt and colleagues have characterized a number of situations in which substances are abused (Chaney et al., 1982; Cummings and Gordon, 1980; Marlatt and Gordon, 1980, 1985). While the original taxonomy of these situations focused on settings in which relapse occurred following a period of abstinence from a substance, the settings appear to represent situations in which substance use in general will be more likely to occur (Annis and Davis, 1988a; Marlatt, 1996). The situations as originally categorized are found in Figure 4-16 .
These situations have been classified into categories. At the broadest level, they are considered either interpersonal (i.e., involving a present or recent interaction with someone else) or intrapersonal-environmental (i.e., factors that are either internal to the individual or reactions to nonpersonal environmental events). There are a number of more specific situations within each of these broader categories. These situations include many emotional, interpersonal, and environmental settings in which people commonly abuse substances and where they are likely to relapse. Therefore, these are called "high-risk" situations. These situations also serve as the foundation from which a number of measures of substance-related self-efficacy have been developed (Annis and Davis, 1988b; DiClemente et al., 1994; Sklar et al., 1997).
While there appears to be considerable overlap in high-risk situations across substances (Cummings and Gordon, 1980), there are also a number of substance-specific patterns. Emotional and situational risk factors have been examined among a clinical sample of individuals who were primary abusers of alcohol, cocaine, marijuana, sedatives and tranquilizers, or heroin/opiates. They found that positive social experiences and negative emotional states were important risk factors for patients who were dependent on alcohol or cocaine. Positive emotional and situational factors were most important for those using marijuana. Individuals dependent on sedatives and tranquilizers or heroin/opiates reported that negative physical states and interpersonal conflict were the most important risk factors. Again, it is the individual's appraisal of such situations, in terms of its threat to maintaining abstinence relative to their available coping abilities, that determines the situational risk for the individual (Myers et al., 1996).
The cognitive-behavioral approach attempts to integrate all of these theoretical details into a meaningful model of substance abuse disorders (Mackay et al., 1991; Marlatt et al., 1988). Figure 4-17 presents a flowchart that depicts this model of substance abuse and dependence.
The cognitive-behavioral model assumes that substance abusers are deficient in coping skills, choose not to use those they have, or are inhibited from doing so (Monti et al., 1994, 1995). It also assumes that over the course of time, substance abusers develop a particular set of effect expectancies based on their observations of peers and significant others abusing substances to try to cope with difficult situations and through their own experiences of the positive effects of substances. They have come to believe that substances have positive benefits that are more immediate and prominent than their negative consequences. They also come to rely on substances as a means of trying to cope with these situations.
To the extent that the individual is lacking in the coping skills necessary to deal with the demands of high-risk substance abuse or relapse situations, his sense of self-efficacy decreases. As personal efficacy decreases, the anticipated positive effects of substance abuse increase and become more salient (Brown et al., 1998). Under such conditions, the individual is likely to use (Moser and Annis, 1996). When confronted by similar situations in the future, the likelihood of using continues to be quite high, unless new coping skills have been learned. Given the interaction of self-efficacy, substance-related effects expectancies, and high-risk situations, "the decision to drink or exercise restraint (self-control) is ultimately determined by self-efficacy and outcome expectations formulated around a current situational context" (Abrams and Niaura, 1987, p. 152).
Attributional processes and emotional responses also play a role in an individual's decision to use (Marlatt and Gordon, 1985). Should the client attribute her substance abuse to internal, stable, and global characteristics (e.g., "I'm nothing but an addict; there's nothing that I can do to stop using"), then it is likely that she will feel angry, depressed, hopeless, and helpless. These reactions are less likely to occur and to be less pronounced for individuals who are more firmly committed to the goal of abstinence or moderation and for those who have maintained such goals longer. These negative emotions represent yet another high-risk situation. If the individual does not have the necessary restorative coping skills to deal with them and to counteract the impact of a negative attributional style, it is more likely that an initial slip will continue on as a full-blown relapse (Stephens et al., 1994).
Cognitive-behavioral therapy (CBT) derives, in part, from both behavioral and cognitive theories. While sharing a number of procedures in common, CBT is also distinct in many ways from these other therapies (Carroll, 1998). In comparison to cognitive therapy, CBT places less emphasis on identifying, understanding, and changing underlying beliefs about the self and the self in relationship to substance abuse. It focuses instead on learning and practicing a variety of coping skills, only some of which are cognitive. A greater emphasis is also placed on using behavioral coping strategies, especially early in therapy. CBT tries to change what the client both does and thinks.
In comparison to behavioral treatments such as the community reinforcement approach, CBT focuses more on cognitions, beliefs, and expectancies. Also, CBT generally does not incorporate contingency management approaches such as the use of vouchers to reinforce desired behaviors. CBT is usually confined to the treatment session (although therapists often give homework to clients to be completed outside the therapy session), whereas the community reinforcement approach stresses the importance of incorporating interventions into real world settings and taking advantage of community resources. Figure 4-18 lists a number of features thought to be unique to cognitive-behavioral interventions.
CBT uses learning processes to help individuals reduce their drug use. It works by helping clients recognize the situations in which they are likely to use, find ways of avoiding those situations, and cope more effectively with situations, feelings, and behaviors related to their substance abuse (Carroll, 1998). To achieve these therapeutic goals, cognitive-behavioral therapies incorporate three core elements: (1) functional analysis, (2) coping skills training, and (3) relapse prevention (Rotgers, 1996).
Behavioral, cognitive, and cognitive-behavioral treatments all rely heavily on an awareness of the antecedents and consequences of substance abuse. In all of these therapeutic approaches, the client and therapist typically begin therapy by conducting a thorough functional analysis of substance abuse behavior (Carroll, 1998; Monti et al., 1994; Rotgers, 1996). This analysis attempts to identify the antecedents and consequences of substance abuse behavior, which serve as triggering and maintaining factors. Antecedents of use can come from emotional, social, cognitive, situational/environmental, and physiological domains (Miller and Mastria, 1977). The functional analysis should also focus on the number, range, and effectiveness of the individual's coping skills. While a major emphasis in cognitive-behavioral therapy is on identifying and remediating deficits in coping skills, it is also important to assess the client's strengths and adaptive skills (DeNelsky and Boat, 1986).
The functional analysis will also assess features in the client's emotional states and thoughts and in her environment that are highly associated with substance abuse. This allows the identification of situations that are particularly high risk for the individual. In addition, it is important to determine what the person thought, felt, and did both during and after high-risk situations. Gaining information about high-risk situations in which the person drank or used drugs and those in which a relapse crisis was encountered but averted is helpful in assessing coping abilities, self-efficacy perceptions, substance-related effect expectancies, and attributional processes.
Without such a thorough assessment, CBT treatment cannot proceed and is likely to fail (Rotgers, 1996). This detailed analysis serves to inform the treatment process and individualize the specific interventions and treatment plan for the client. The therapist and client can then use the results of the functional analysis to anticipate high-risk situations and develop specific methods to avoid or cope with them.
Questionnaires, interviews, and role-playing measures are available to assist the therapist in the assessment and functional analysis. The therapist should try to evaluate the number and type of high-risk situations, the temptation to use in these situations, confidence that one will not use in high-risk situations, substance abuse-related self-efficacy, frequency and effectiveness of coping, and substance-specific effect expectancies. More detailed information on the assessment process in cognitive-behavioral approaches to substance abuse and its treatment is available in a number of sources (Donovan, 1998; Donovan and Marlatt, 1988; Monti et al., 1994; Sobell et al., 1988; and Sobell et al., 1994). For a review of assessment tools that can be used in developing a functional analysis see TIP 35, Enhancing Motivation for Change in Substance Abuse Treatment (CSAT, 1999c).
A major component in cognitive-behavioral therapy is the development of appropriate coping skills. Deficits in coping skills among substance abusers may be the result of a number of possible factors (Carroll, 1998). They may have never developed these skills, possibly because the early onset of substance abuse impaired the development of age-sensitive skills. Previously developed coping skills may have been compromised by an increased reliance on substances use as a primary means of coping. Some clients continue to use skills that are appropriate at an earlier age but are no longer appropriate or effective. Others have appropriate coping skills available to them but are inhibited from using them. Whatever the origin of the deficits, a primary goal of CBT is to help the individual develop and employ coping skills that effectively deal with the demands of high-risk situations without having to resort to substances as an alternative response.
A number of published treatment manuals are available to guide skills training with substance users (Carroll, 1998; Kadden et al., 1992; Monti et al., 1989). These manuals provide a session-by-session overview of the intervention. The material covered in these interventions can be categorized into a number of broad classes. The skills to be taught are either specific to substance abuse (e.g., coping with craving, refusing an offer of alcohol or drugs) or apply to more general interpersonal and emotional areas (e.g., communication skills, coping with anger or depression). They are either cognitive or behavioral in nature. Some might be viewed as essential and would be expected to be used for all clients, while others would be viewed as more elective in nature and would be selected for a particular individual based on the functional analysis. The ability to individually tailor the skills training to the client's needs represents one of the strengths of CBT.
Figure 4-19 presents a list of session topics (Monti et al., 1989) which served as the foundation for the CBT delivered in Project MATCH (Matching Alcohol Treatment to Client Heterogeneity Project) (Kadden et al., 1992), a large multisite study of treatment matching funded by the National Institute on Alcohol Abuse and Alcoholism (NIAAA). While the topics used in this particular example were developed for use with clients with alcohol abuse disorders, they are easily adapted to the needs of clients who are abusing other substances.
According to Carroll, teaching coping skills is the core of CBT (i.e., helping clients recognize the high-risk situations in which they are most likely to abuse substances and to develop other, more effective means of coping with them) (Carroll, 1998). The therapist teaches the client specific behavioral skills for forming and maintaining interpersonal relationships. For example, a client may be taught how to refuse a drink in a social situation (which might include some form of assertiveness training, as described in Figure 4-20 ). Learning how to develop new social contacts with people who are not substance abusers is another example.
Skills training sessions follow a relatively standardized format. The client is given an overview of the session, describing the area to be addressed and the rationale for the specific intervention to be used. This is facilitated by skill guidelines that focus attention on the most important aspects of the approach as it applies to substance abuse. After discussing the issues involved in the session, the therapist models the effective coping skill for the particular topic. The therapist then asks the client to participate in a role-playing scenario in which he can rehearse the new coping behaviors. The therapist provides feedback and guidance while the client continues in the behavioral rehearsal. Between sessions, therapists often give homework assignments that provide the client with an opportunity to try behaviors learned in sessions in real-life settings. The next session usually begins with a review of this homework and the client's reactions to it.
Skills training approaches have been evaluated more than many other approaches to substance abuse disorders. Monti and colleagues evaluated a coping skills training intervention for cocaine-dependent clients (Monti et al., 1997). A cocaine-specific skills training intervention, administered as individual counseling, was added to a more comprehensive treatment program along with a placebo control. The approach involved the identification of high-risk situations based on a functional analysis and the teaching of coping skills to deal with these situations. In comparison to the control condition, clients who received individualized coping skills training had significantly fewer days of cocaine use and significantly shorter periods of binge use of cocaine over a 3-month followup period. Although the two groups did not differ in their rates of relapse, the pattern of use and the harm associated with it clearly favored the skills training condition.
The third core element of CBT is relapse prevention. While there are a number of different models of relapse (Donovan and Chaney, 1985), the two best articulated within the cognitive-behavioral model are those presented by Annis and Davis and Marlatt and Gordon (Annis and Davis, 1988b; Marlatt and Gordon, 1985). Relapse prevention approaches rely heavily on functional analyses, identification of high-risk relapse situations, and coping skills training, but also incorporate additional features. These approaches attempt to deal directly with a number of the cognitions involved in the relapse process and focus on helping the individual gain a more positive self-efficacy.
Although self-efficacy is related to the availability of coping skills and would be expected to increase as the client learns new skills, this does not always occur spontaneously. It is often necessary to help the client change the passivity and sense of helplessness that often accompany low self-efficacy. Bandura noted that there are a number of ways to increase self-efficacy (Bandura, 1977). However, the model that appears to have the greatest impact and lasting influence uses the idea of performance accomplishments to enhance client self-efficacy. In this model, the client is coached to do something that she previously was unable to do. Annis and Davis use graduated homework assignments to help in this process (Annis and Davis, 1988b). The client gradually exposes herself to increasingly difficult situations with greater relapse risk but does so without using. The rate of the exposure is calculated to be at a level that can be handled by the client. The accomplishment of these homework tasks serves as a point of discussion to reinforce the client's growing sense of self-efficacy.
The therapist practicing CBT will also challenge the attributional process and emotional aftermath of a relapse. If a slip occurs, the therapist should try to bring the more negative attributions for relapse (internal, stable, and generalized) to the client's attention so that he can identify these tendencies and learn how to change them. Clients can be helped to see the relapse as caused by a lack of appropriate coping skills for the particular situation (i.e., external), alterable with training or practice (i.e., unstable), and not implying that everything the person does is wrong (i.e., specific). This change in perspective will help reduce the client's sense of helplessness and loss of control. Addressing the attributional process should be done in the broader context of educating the client about the relapse process.
Research has consistently shown that people who expect more positive effects from substances are more likely to abuse them (Brown, 1993; Goldman and Rather, 1993). It has also become clear more recently that individuals who are aware of and concerned about the more negative consequences associated with substance abuse are less likely to use (Jones and McMahon, 1996). There are also significant differences in the way men and women react to expectancies concerning substances; males are more affected by positive expectancies, whereas the positive expectancies of females are more balanced by negative expectancies (Romach and Sellers, 1998).
The therapist can work to challenge a client's positive expectancies about the effects of substances. There are two strategies that the therapist can use concerning expectancies in order to decrease substance abuse: change the client's belief in the positive effects of the substance or get her to pay more attention to her knowledge and experience of its negative effects.
For a long time, researchers did not believe that positive expectancies concerning substance effects could be changed, but a study on heavy-drinking college students showed that expectations regarding alcohol effects could be altered (Darkes and Goldman, 1993). In group sessions, several techniques were used to make the students aware that some of their alcohol-related expectancies were false. For example, the heavy-drinking college students were told that the beverages they were drinking contained alcohol, but they were actually given nonalcoholic drinks, disguised to look, smell, and even taste like alcohol. They then engaged in group party games, in which most displayed the uninhibited behavior that is associated with alcohol intoxication. Later, when they were told that their beverages were actually placebos, they were surprised. Group discussion and other information on placebo effects altered their perceptions of the positive effects of alcohol. A significant decrease in alcohol consumption was noted in this group after the intervention, compared to a control group that received conventional information on the effects of alcohol. Challenging social beliefs about the effects of a substance may alter its use.
Another way to use substance expectancy information in therapy is to have the client consider both the positive and negative effects of the substance. Many clients have automatic scripts like "I'll feel more relaxed if I drink" without considering other scenarios, like: "I'll drink too much. I'll have a fight with my girlfriend, and then I'll sleep in and not go to class." The therapist helps the client acknowledge that the other consequences exist and are not being attended to. It is possible to use a decisional balance procedure in this process, wherein the client is asked to list all the positive and negative things associated with drug use. By acknowledging the substance's positive effects, the therapist gains credibility and reduces resistance from the client. The client can more easily acknowledge the negative aspects of substance abuse and make those beliefs more salient. This technique is a mainstay of motivation enhancement therapies that are largely cognitive in nature (Miller and Rollnick, 1991). (TIP 35, Enhancing Motivation for Change in Substance Abuse Treatment [CSAT, 1999c], gives more detailed information on these approaches.)
Relapse prevention also stresses the importance of preparing for the possibility of a relapse and planning ways to avoid it or, failing this, stop the process quickly and with minimal harm when it does occur. Clients are sometimes apprehensive about talking so directly about the possibility of relapse. The therapist can help dispel these concerns by using an analogy of fire drills. Having a drill and being prepared for a fire does not necessarily mean that a fire will occur. However, if one does, it will be possible to get out of the situation without getting burned. It is helpful to have very concrete emergency plans, including the phone numbers of individuals supportive of the client's recovery process. Including family members in the planning process is important because they are often better able than the client to see the warning signs of an impending relapse.
Relapse prevention also stresses the development of a more balanced and healthier lifestyle. Marlatt and Gordon posit that one source of possible relapse risk has to do with the degree of stress or daily hassles that the client experiences (Marlatt and Gordon, 1985). They suggest that when the demands and obligations a client feels ("shoulds") outweigh the pleasures the individual can engage in ("wants"), then his life is out of balance. This often results in feelings of deprivation and resentment. In response to these feelings, the individual could begin making decisions that gradually lead toward possible relapse. The goal is to help the individual find a better balance, increasing involvement in pleasant and rewarding activities while reducing the level and sources of stress.
The following case study involves a young male cocaine user who has sought outpatient treatment. It reflects interactions early in the course of the session and is meant to depict some of the questions the therapist could ask to gain information about the antecedents, consequences, and cognitive mediators involved in his use.
Therapist: So, can you tell me about your cocaine use and why you are coming to treatment now?
Client: Well, I finally came to the end of my rope. I kept using even though I didn't want to, and I felt that I was nothing but a junkie who had no future. It's just hopeless.
Therapist: What makes you say that?
Client: Well, I just can't stop using. Even when I've gone through treatment in the past, I end up using in no time. When I look at my track record, I don't see much of a future.
Therapist: I wouldn't give up hope yet. We'll work together to help you get a better look at your cocaine use, some of the things that trigger it, and some of the benefits you think you get from it. Sometimes by looking at your use from a different perspective, you can help put it into context and things don't seem so hopeless. Now why don't you tell me about how you slipped and started using after your last time in treatment. What was going on in your life? What were you feeling? What were you thinking about yourself and your life?
Client: Well, when I got out I still had some doubts about whether I would make it or not. I mean I felt better about myself, but there was still a lot of crap going on in my life. I had bills to pay. My relationship was falling apart. I was still being hassled by my probation officer. I was feeling kind of overwhelmed. Here I thought I would walk out of there a new man, but I walked out with all the same problems.
Therapist: Was there any time after treatment when you felt you could handle all the problems facing you?
Client: Well, for a while, then I started to feel depressed. I mean you go through treatment, and this stuff shouldn't be happening.
Therapist: What did you try to do to deal with it all?
Client: At first I thought I would get myself organized and get a plan. But it didn't work. As much as I tried, I couldn't figure out a way to put all this stuff in its place and handle it. So I just threw up my hands and said, "Screw it!" I felt like the best thing to do was to pull the blankets over my head and hope that it would all blow over.
Therapist: So, did it blow over?
Client: No. Things just kept getting worse. I couldn't pay my bills. My relationship was gone, and I got booted out of my apartment.
Therapist: As all this was happening, did you think about using cocaine?
Client: You bet I did! I kept thinking, "Damn, it sure would feel good to get all this off my mind." And I knew that if I used coke it would all go away--at least for a while.
Therapist: So, as you thought about the cocaine, what positive things did you think you would feel if you used?
Client: I knew I'd feel a rush, I'd feel damn good--and I'd just forget. I could get out of the depression and funk I'd been in. I was just looking to feel better.
Therapist: Did you think of any negative things about using?
Client: Yeah. I always seem to crash after using. So I lose the high and find myself sometimes even more depressed than before. But that didn't seem to bother me. I'm willing to put up with it for a while. I'll take the high any day. It lets me get away from all this crap--at least for a while.
Therapist: So what were the circumstances of your starting to use again?
Client: Well, like I said, I got booted from my apartment. And I couldn't go stay with my girlfriend since she booted me too. So I had to find a place to stay. I called an old friend who said I could stay at his place for a while. We used to do a lot of drugs together. I knew he might not be the best person to be staying with, but he was the only one I felt would put up with me. So, I moved in. I was feeling pretty low, thinking about everything that had happened to me and was not sure what I was going to do. My friend pulled out some coke and asked if I'd like some. I just kept thinking of how lousy I felt and how good I would feel if I used. So I said yeah, why not.
In this case study, it is clear that the client has a low sense of self-efficacy predicated in part by his past treatment failures and his inability to cope with difficult situations. As a result, he feels depressed and helpless. He makes a half-hearted attempt at problemsolving but fails in this attempt. Then he switches to passive-avoidant approaches in order to cope (e.g., pulling the blanket over his head and hoping it will all blow away). His depression continues unabated as the daily hassles mount. The positive expectancies he has about cocaine as the "magic elixir" are quite strong and seem to outweigh potential negative consequences. His situational context contains two high-risk situations. The first is the negative mood states that he experiences, when he has abused substances in the past. The second is the indirect social pressure involved in returning to a setting that had been associated with substance abuse in his past. There is also the proximal influence of the direct social pressure to use from his friend. The likelihood of relapse was high, and, in fact, relapse occurred.
The therapist in this case might consider using skills training that focuses on problemsolving, stress management to alleviate his depression, developing communication skills, practicing substance refusal skills, and developing a social support network. The therapist should target both this client's low self-efficacy and his positive cocaine-effect expectancies. Clearly the full intervention plan would require further assessment and a functional analysis; however, a direction for further treatment can already be seen in this brief interchange.
Two advantages of CBT are that it is relatively brief in duration and quite flexible in implementation. CBT typically has been offered in 12 to 16 sessions, usually over 12 weeks (Carroll, 1998). The form of CBT used in NIAAA's Project MATCH (Kadden et al., 1992) consisted of 12 sessions, administered as individual therapy, meeting once per week. The sessions included eight "core" sessions that dealt with alcohol-related issues (e.g., coping with craving, drink refusal, relapse emergency planning) and general problem-solving skills that all clients were expected to receive, and four "elective" topic areas chosen from a menu of more general social and interpersonal issues based on individually assessed problem areas. A 12-session CBT for cocaine addicts suggested that this length of treatment is sufficient to achieve and stabilize abstinence from cocaine (Carroll, 1998). However, not all clients will respond in that amount of time. In such cases, an initial trial CBT can serve as preparatory to a more intensive treatment experience.
Varieties of cognitive-behavioral therapy are applicable to a wide range of substance abusers. The outpatient CBT program developed by Carroll for cocaine users excluded a number of different clients as inappropriate for that form of treatment (see Figure 4-21 ). However, even though these criteria were derived from cocaine users, they appear to be applicable to clients using other substances.
While reliance on the results of the functional analysis makes skills training particularly well suited for individual therapy, these interventions can easily be adapted for use in group settings (Monti et al., 1989). Similarly, they can be used with inpatients or outpatients and can be administered as part of an intensive phase of treatment or as part of less intensive aftercare or continuing care. CBT is also compatible with a number of other elements in treatment and recovery, ranging from involvement in self-help groups to pharmacotherapy (Carroll, 1998).
In contrast to many other therapies, cognitive-behavioral therapy for the treatment of substance abuse disorders has substantial research evidence in support of its effectiveness.
The research findings on the use of coping skills training with alcohol- and cocaine- dependent clients indicate that this strategy has strong empirical support. A review of outcome studies evaluating the efficacy of relapse prevention interventions indicates that the support for relapse prevention is more equivocal (Carroll, 1996b). Relapse prevention was found to be superior to no treatment, but the results have been less consistent when it is compared to various control conditions or to other active treatments. There are some outcomes on which relapse prevention may have considerable impact (Carroll, 1996b); for instance, although not necessarily reducing the rate of relapse, clients treated in relapse prevention appear to have less severe relapses when they occur.
Overall, behavioral, cognitive, and cognitive-behavioral interventions are effective, can be used with a wide range of substance abusers, and can be conducted within the timeframe of brief therapies.