Excessive substance use by people with schizophrenia has most of the same adverse social, health, economic, and psychiatric consequences as it does for other individuals. Moreover, it has additional serious consequences for these multiply handicapped patients. Substances of abuse tend to increase dopaminergic activity (
5), thereby increasing the risk of symptom exacerbation and relapse and compromising the efficacy of neuroleptic medications (
3). Substance use is also thought to decrease compliance with treatment and often serves as a source of conflict in families, a pernicious circumstance for patients with schizophrenia, who are highly vulnerable to heightened stress (
6).
People with serious mental illness, the majority of whom have schizophrenia, are now one of the highest-risk groups for HIV (
7). Ample data indicate that substance use substantially increases the likelihood of unsafe sex practices (
8), the primary source of infection in this population. Women who abuse substances also are at increased risk of physical and sexual abuse (
9). Substance use also has deleterious cognitive effects, a particularly serious consequence for people with schizophrenia, as their information-processing system is already compromised by the illness (
10).
Treatment issues
The problem of substance abuse in schizophrenia has generated a large literature, but to date few well-controlled trials of specific interventions to deal with this pernicious combination have been done. Several recent pilot and demonstration projects have yielded mixed results (
15,
16,
17,
18).
Despite the absence of definitive data on specific intervention techniques, researchers appear to have broad agreement about some general requirements for effective treatment. First and foremost is the contention that dually diagnosed patients need a special program that integrates and coordinates elements of both psychiatric and substance abuse treatment (
3,
19,
20). A related caveat is that the confrontational, highly affective style of many traditional substance abuse treatments is contraindicated for people with schizophrenia.
A second assumption is that treatment is best conceptualized as an ongoing process that involves a number of relatively distinct stages in which motivation to reduce substance use waxes and wanes. For example, Osher and Kofoed (
21) hypothesized that treatment entails four stages: engagement, persuasion, active treatment, and relapse prevention. McHugo and colleagues (
22) further elaborated this model into eight stages through subdivision of the four stages into early and late phases.
These two models may have considerable clinical utility, but they are more descriptive than explanatory. They provide operational definitions for the treatment focus—for example, a patient who is not currently interested in reducing substance use must first be persuaded to reduce use, and an individual who has recently become abstinent needs help in preventing substance use relapse. However, the models fail to explain how or why an individual becomes motivated to change, and they fail to predict what types of persuasion, engagement, or active treatment will be effective at any particular point in time.
Transtheoretical model
A more elaborate conceptualization of the stages of treatment is provided by the transtheoretical model of change (
23,
24). This model has posited several dimensions for understanding the process of intentional change of problem behaviors, which includes both interventions and self-initiated change in the natural environment.
The basic dimensions of the model are stages and processes of change. The stages of change represent the temporal, motivational aspect of the process of change. The processes are the strategies or coping mechanisms that move individuals through the stages. A third dimension, levels of change, represents the multidimensional nature of the problems for each individual that complicates the process of change for any single problem (
23,
24,
25,
26).
A series of studies by Prochaska, DiClemente, and others (
23,
25,
27,
28) has demonstrated that individuals can be classified as being at various points in the stages of change and seem to move systematically through five stages. The first stage is precontemplation, when individuals either are unconvinced that they have a problem or are unwilling to consider change. The second is contemplation, when individuals are considering change in the future, in six months to a year. The third stage is preparation, when individuals have a more proximal goal to change in the next month and make a commitment and initial plans to change the behavior. The fourth is an action stage, in which individuals take effective action to make the change. The fifth is a maintenance stage, where the task is to consolidate the change and integrate it into the individual's lifestyle.
The course of progression through this linear series of stages is often cyclical in the short term for most individuals, and relapse and recycling through these stages is the norm (
23). Over time, many individuals move through this cyclical process repeatedly until they are able to successfully attain sustained change of the problem behavior. Thus both short-term and long-term perspectives are needed to understand movement through this process of change.
Several other variables are also important in the process of change. Decisional balance refers to an individual's subjective evaluation of the pros and cons of engaging in the problem behavior versus changing. Considerations of decisional balance have been important indicators that the individual is in an early stage of change or is moving through the early stages of change (
28,
29).
Self-efficacy deals with the individual's subjective sense of competence to perform specific behaviors, such as resisting temptations to use drugs. Self-efficacy has been shown to be salient as a predictor in the later stages of change (
30).
Several recent studies have supported the validity of components of the transtheoretical model with heterogeneous groups of psychiatric patients (
31,
32). Although patients with schizophrenia were included in these samples, the applicability of the model for schizophrenia was not specifically tested.
Schizophrenia is characterized by deficits in a variety of neurobiological, cognitive, and behavioral capacities that may limit the applicability of the model with this population. For example, the model deals with change in intentional behavior, but patients with schizophrenia have significant difficulty sustaining intentional behavior. The model posits that change entails ongoing judgments about the pros and cons of substance use and one's ability to make changes, but schizophrenia is marked by cognitive deficits that interfere with both introspection and complex problem solving. The model also assumes a set of coping skills—the processes of change—that may not be in the repertoire of many individuals with schizophrenia.
Obstacles to change
An extensive body of research on substance abuse and addiction in the general population indicates that critical factors in abstinence and controlled use of addictive substances include high levels of motivation to quit, the ability to exert self-control in the face of temptation or urges, cognitive and behavioral coping skills, and social support or social pressure (
33,
34,
35,
36). Unfortunately, the abuser with schizophrenia often has limitations in each of these areas.
First, several factors can be expected to diminish motivation among patients with schizophrenia. Most patients suffer from some degree of generalized avolition and anergia as a function of hypodopaminergia in the frontal cortex (
37), medication side effects, or other social, psychological, and biological factors that contribute to negative symptoms (
38,
39). Thus they may lack the internal drive to initiate the complex behavioral routines required for abstinence. Another negative symptom, anhedonia, may compromise the experience of positive affect, thereby limiting patients' ability to experience pleasure and positive reinforcement in the absence of substance use and restricting patients' appraisal of the advantages of abstinence (
40). There are extensive data documenting that people with schizophrenia can learn a variety of skills and acquire new information (
41), but there is scant evidence that they use acquired skills in the community—that is, that they are motivated to apply what they have learned—or that they can sustain motivation in the absence of short-term reinforcement, such as that provided by awarding tokens (
42).
A second factor that might compromise the applicability of the transtheoretical model for people with schizophrenia is the profound and pervasive cognitive impairment that characterizes the disorder. Research in the last ten years has documented that patients with schizophrenia have prominent deficits in attention, memory, and higher-level cognitive processes such as abstract reasoning. They also have deficits in maintenance of set, which is the ability to sustain focus on a strategy or goals, and in the ability to integrate situational context or previous experience into ongoing processing—that is, to use previous experience to direct current behavior—as well as in other executive functions (
43). They have been shown to have profound deficits in problem-solving ability on both neuropsychological tests (
44) and more applied measures of social judgment (
45).
Several lines of evidence suggest that cognitive impairment is largely independent of symptoms and that many of these higher-level deficits may result from a subtle neurodevelopmental anomaly that is reflected in frontal-temporal lobe dysfunction (
44,
46). Moreover, cognitive performance deficits are not substantially ameliorated by treatment with typical antipsychotic medications (
47). The new atypical antipsychotic medications may have a more beneficial effect on cognition (
48), but overall the data on this issue are mixed.
The higher-level cognitive deficits would make it very difficult for patients with schizophrenia to engage in the complex processes thought to be central to intentional change in behavior and the transtheoretical model. They have difficulty engaging in self-reflection or evaluating previous experiences to formulate realistic appraisals of self-efficacy. Deficits in the ability to draw connections between past experience and current stimuli (
49) may impede the ability to relate their substance use to negative consequences over time and to modify their decisional balance accordingly.
Deficits in problem-solving capacity and abstract reasoning may impede the ability to evaluate the pros and cons of substance use or formulate realistic goals. Motivation to change and inclination to resist urges and social pressure vary over time for anyone coping with substance abuse, but people with schizophrenia have the added burden of being unable to reliably recall their intentions and commitments to change. Frontal-temporal impairments are associated with a phenomenon referred to as "forgetting to remember," which may make it difficult for patients to recall commitments to change or to use coping skills in the face of temptations or cravings to use substances.
Another constraint on change is related to the marked social impairment that characterizes the illness. People with schizophrenia are often unable to fulfill basic social roles, they have difficulty initiating and maintaining conversations, and they frequently are unable to achieve goals or have their needs met in situations requiring social interaction (
50). The precursors of adult social disability can often be discerned in childhood (
51) and may be associated with early attentional impairments (
52). Pronounced social impairment would leave patients with schizophrenia and substance abuse vulnerable in a number of ways: they would have difficulty developing social relationships with individuals who do not use drugs, resisting social pressure to use drugs, and developing the social support system needed to reduce use.
A new treatment approach
Many patients do reduce their use of substances, with or without treatment, so sustained motivation and self-directed behavior change is clearly possible. However, existing treatment models do not adequately account for the specific learning and performance deficits that are characteristic of the illness. We have developed a new treatment approach that takes into account the unique deficits in motivation, cognitive ability, and social skills associated with schizophrenia.
The treatment protocol contains four modules that are implemented sequentially. First, social skills and problem-solving training enables patients to develop social contacts with others who do not abuse substances and to be able to refuse social pressure to use substances. The second module focuses on education about the reasons for substance use, including habits, triggers, and craving, and the particular dangers of substance use for people with schizophrenia. The third module consists of motivational interviewing, goal setting for decreased substance use, and development of contingency contracts for clean urine screening tests. The fourth module consists of training in behavioral skills for coping with urges and high-risk situations and training in relapse prevention skills.
The basic training techniques have been used effectively for more than 20 years to teach social skills to persons with schizophrenia (
41) and include instruction, modeling, role play, feedback and positive reinforcement, and homework. Patients repeatedly rehearse both behavioral skills, such as refusing unreasonable requests, and didactic information, such as information about the role of dopamine. They receive social reinforcement for their efforts.
Training is done in a small-group format, with six to eight participants, to ensure sufficient individual attention and opportunities to rehearse skills within the session. The 90-minute sessions are held twice a week for approximately six months. This treatment duration allows time for participants to develop motivation to change and work toward their goals. As an accommodation to the disorganization and life problems that are typical of these patients, make-up sessions are offered as needed.
Several steps are taken in consideration of cognitive deficits. Sessions are highly structured, and there is a strong emphasis on behavioral rehearsal. The didactic material is broken down into small units. Complex social repertoires, such as making friends and refusing substances, are divided into component elements, such as maintaining eye contact and being able to say no. Patients are first taught to perform the elements, and then gradually learn to combine them.
The intervention emphasizes overlearning of a few specific and relatively narrow skills that can be used automatically, thereby minimizing the cognitive load for decision making during stressful interactions. Extensive use is made of learning aides, including handouts and flip charts, to reduce the requirements on memory and attention. Patients are prompted as many times as necessary, and there is also extensive repetition of material within and across sessions.
The module focused on motivational interviewing is based on work done by Miller and Rollnick (
53). We have found it difficult to engage patients in discussions of the general benefits of continuing to use substances versus quitting. Rather than attempting to make a broad shift in the decisional balance about substance use, we attempt to identify one or two specific negative consequences that have a strong impact on the patient and that can serve as a prompt for change. For example, a patient who is on conditional release from jail may be engaged in a discussion of how continued substance use may lead to a return to jail. These negative consequences then serve as a stimulus for patients to make and adhere to substance use goals.
Because many patients with schizophrenia may not be able or willing to accept complete abstinence as a goal (
16,
54), identifying more modest goals that the patient is comfortable with is particularly important. Consistent with a harm-avoidance model (
20), we attempt to shape behavior by gradually encouraging the patient to focus on increasing periods of abstinence or on using less drugs or alcohol.
The motivational interviewing sessions are supplemented by a financial contingency for clean urine screening tests. The reward for clean tests is $1 during initial sessions and increases to $2 over successive sessions. The urine testing helps ensure honest reporting of drug use and also adds a modest incentive for abstinence. Clients whose urine samples are positive for drugs receive training in coping skills to help prevent future slips.
We do not require patients to set any goals about substance use when they enter the program. Rather, we first attempt to engage them in the program. One factor that deters many patients from committing to reduced use is low self-efficacy based on a long history of failure in achieving any goals. Consequently, we first attempt to enhance a sense of efficacy by building the experience of success into the program. Patients first achieve success in learning skills that are directed at starting conversations and refusing unreasonable requests. Only then is the focus shifted to social situations involving substance use. Patients have the opportunity to practice saying no before they make a commitment to saying no. Hence, when we do help them to set goals, they have already had some success in practicing some of the skills needed to achieve their goals.
The content of the sessions on coping skills and relapse prevention is adapted from substance abuse programs based on social learning theory that have proven to be effective with less impaired patients (
55,
56,
57). Patients with schizophrenia have considerable difficulty with abstract concepts and in generalizing principles of action across situations. Rather than teaching generic problem-solving skills and coping strategies that can be adapted to a variety of situations, we focus on specific skills that are effective for handling a few key high-risk situations—for example, what to do when your brother or a specific friend offers you cocaine, rather than what to do when anyone offers it to you. Similarly, management of negative affect and interpersonal distress is addressed concretely using scenarios provided by the patient to practice alternate responses and highlight behaviors that could evoke more satisfying interactions.
Our treatment approach has evolved in the context of a behavioral treatment development grant from the National Institute on Drug Abuse. Between May 1996 and Jan 1998, a total of 80 patients have been enrolled in the program. We have been highly successful in engaging and retaining participants: the attrition rate of 38.5 percent compares very favorably with rates for other substance abuse programs reported in the literature, and no subject who completed the first three weeks of the program dropped out after that.
We have also been able to teach therapists to conduct the treatment according to a detailed manual. Independent ratings of adherence to the manual have averaged 96.6±4.9 percent, and ratings of competence in presenting the material in the manual have averaged 5±.2 on a 5-point scale. The results for the first 12 subjects to complete the entire finalized protocol are quite promising. Based on the in-session urine testing, six of the 12 had negative tests an average of 90.8 percent of the time (range=72.7 to 100 percent) over a 4.5-month period. Subjects who had good outcomes attended 80.9 percent of all sessions. Subjects who had poor outcomes attended 73.2 percent of sessions before urine testing began and only 43.6 percent afterwards, suggesting it had a pronounced impact. The next phase of the treatment development process will be a controlled clinical trial.
Further research is needed to elucidate why people with schizophrenia abuse drugs and alcohol, how they reduce substance use on their own, and how cognitive deficits influence the process of change. We are currently evaluating which elements of intentional change are relevant for individuals with schizophrenia who abuse substances and how to adapt constructs of the transtheoretical model of change for this population, including how to modify specific decisional considerations that will promote reduction of substance use and which processes of change are used most frequently. The findings will offer insight into how people with schizophrenia change their behavior and the extent to which traditional and innovative approaches are appropriate.