Use of Legal Drugs by Psychiatric Outpatients: Benefits, Costs, and Change

Abstract

Caffeine, nicotine, and alcohol use by persons with a severe mental illness occurs frequently but is poorly understood. We used qualitative methods to elicit information regarding the functional relationships between legal substance use and its antecedents and consequences. This report summarizes responses provided by 37 psychiatric outpatients who participated in focus groups and key informant interviews regarding the use and functions of alcohol, caffeine, and nicotine. We describe major themes regarding positive and negative consequences, triggers for use, change efforts, and offer observations regarding substance use in this under-served population.

Licit Substance Use Among Psychiatric Outpatients: A Qualitative Analysis

Caffeine, nicotine, and alcohol are used extensively by persons with severe mental illnesses (SMI), often at rates that far exceed rates found in the general population. Test, Wallisch, Allness, and Ripp (1989) reported that 86% of their patients with schizophrenia drank coffee, averaging 6 cups per day; this quantity has been associated with caffeine withdrawal and related disorders (American Psychiatric Association, 1994). In contrast, only 9% of the general population consumes as much caffeine (Greden, Fontaine, Lubetsky, & Chamberlain, 1978). Excessive caffeine use can have deleterious effects on mental status, increasing anxiety and depression (Greden et al., 1978; Winstead, 1976), and exacerbating psychotic symptoms in persons with schizophrenia (Carey & Carey, 1989; Mikkelsen, 1978).

As many as three-quarters of patients with schizophrenia smoke (Goff, Henderson, & Amico, 1992; Ziedonis, Kosten, Glazer, & Frances, 1994) compared to one-quarter of the general population. In addition to the established risks for cancer and heart disease, additional deleterious effects may accrue for smokers receiving psychiatric treatment. Cigarette smoking increases hepatic enzyme action, hastening the elimination of drugs that depend on hepatic metabolism, thereby reducing plasma concentration of some therapeutic drugs (Perry, Miller, Arndt, Smith, & Holman, 1993). Persons with schizophrenia who smoke have also received higher ratings of positive symptoms (Goff et al., 1992; Ziedonis et al., 1994). Although cause and effect relationships are not clear, nicotine may exacerbate psychotic symptoms and reduce the effectiveness of antipsychotic medications.

Alcohol use by the SMI is also common. Among persons diagnosed with schizophrenia, one-third meet criteria for an alcohol use disorder, representing an odds ratio of 3.3. Similar odds ratios for alcohol use disorder range from 1.9 among persons with mood disorders (22%) to 1.5 for persons with anxiety disorders (18%; Mueser, Bellack, & Blanchard, 1992). The literature demonstrates a link between alcohol use and psychiatric re-hospitalization (e.g., Drake, Osher, & Wallach, 1989), exacerbation of delusions and hallucinations (Noordsy et al., 1991) and depressive symptoms (Cuffel & Chase, 1994). Larger studies have also confirmed that comorbid alcohol abuse increases the likelihood of violent behavior (Swanson, 1993) among persons with a SMI.

Drinking coffee, smoking cigarettes, and consuming alcohol occur frequently among persons with SMI, with documented negative effects on clinical status. In order to respond to these potential destabilizing behaviors, clinicians need to understand the functions of licit drug use for this population. Prevention and treatment programs will be most effective if they are based on a population-sensitive analysis of the antecedents and consequences associated with substance use. Few of the previously published descriptive studies have elicited data relevant to perceived functions of caffeine, nicotine, or alcohol among persons with SMI. Although the perceived costs and benefits of substance use have been addressed in other populations (e.g., Velicer, DiClemente, Prochaska, & Brandenburg, 1985), the relevance of previous research to persons with SMI remains unknown. It is likely that the contingencies maintaining substance use may differ across sub-populations of substance users. Similarly, certain individuals may be more sensitive to particular negative effects of substance use, because of the psychotropic medications they are taking, or enduring social or psychological vulnerabilities. Furthermore, little is known about psychiatric patients’ interest in changing their use of caffeine, nicotine, and alcohol, or about the nature of their self-control efforts. Thus, the goal of this research is to provide information that can guide a functional analysis of caffeine, nicotine, and alcohol use among persons with SMI.

The current study employs qualitative research methods (viz., focus groups and elicitation interviews) to enhance our understanding of licit substance use by persons living with a SMI. Qualitative research allows investigators to explore under-studied problems in specific populations, and to generate hypotheses for future quantitative research. These methods have been used in market research and by health educators (Basch, 1987; O’Brien, 1993) to conduct needs assessments for program development (Emery, Ritter-Randolph, Strozier, & McDermott, 1993), to evaluate health care services (Richter, Bottenberg, & Roberto, 1991), and to enhance the cultural sensitivity of intervention efforts (Carey, Gordon, Morrison-Beedy, & McLean, 1997). The value of qualitative research in the context of substance use has been recognized (Dennis, Fetterman, & Sechrest, 1994). Although qualitative research has been underutilized by behavioral investigators, it can be of considerable heuristic value to guide mental health clinicians in the process of case formulation and treatment planning.

Methods

Participants

Thirty-seven psychiatric outpatients (16 men, 21 women) participated; all were receiving treatment at a public hospital. Patients were invited to participate contingent upon (a) the presence of a severe mental illness; (b) alcohol use in the last year; and (c) the absence of symptoms that would preclude participation in a discussion group. Ages ranged from 23 to 55 years (M = 38); the modal participant was high-school educated (M = 12.4 years), Caucasian (85%; 11% African-American; 4% Hispanic and Native-American), and unmarried (88%). Nearly half lived in their own apartment or home (47%), 22% lived in supervised housing, 12.5% lived in rooming houses or hotels, 6% had no permanent address, and 12.5% had unknown residence. Only one of the participants was known to be working part-time; the rest were unemployed and/or disabled due to their mental illness. Primary diagnoses, obtained from the medical chart, consisted of schizophrenia-spectrum disorders (47%), major depression (16%), bipolar disorder (12.5%); the remaining 16% were a mixture of other disorders, including anxiety and personality disorders.

Procedure

Recruitment materials were posted at psychosocial clubs and clinic common areas and distributed to therapists. These materials recognized potential participants as key informants who would help us to understand the programming needs of persons with SMI, and offered $5 in exchange for participation. Research assistants obtained informed consent, conducted brief screens, and reviewed medical records to determine eligibility for participation. Most (90%) patients approached agreed to the brief screen and were interested in participating in the focus groups. Eighty patients were screened, and 37 (46%) were eligible; the primary reason for exclusion was that patients did not report alcohol consumption in the previous year.

At the beginning of each focus group, the written consent was reviewed with each patient by the group facilitators. Patients understood that the groups would be audiotaped, transcribed, and used to inform substance use reduction and other health promotion programs at the treatment site. They knew that their responses were confidential, would not be shared with their treatment team, and would not affect their treatment. Participants were assured that they could refrain from participation or leave the group if the discussion was too distressing.

Consistent with focus group methods (Taylor & Bogdan, 1984), facilitators followed a semi-structured outline to prompt group discussion, but also allowed participant responses to guide the flow of topics. Each session covered a standard sequence of topics, outlined in Table 1. After a general orientation to the group, the discussion started with non-drug related information about participants’ current life circumstances. The purpose of this was to get individuals to start talking about topics they were relatively comfortable with. Facilitators then introduced caffeine, nicotine, and alcohol in that order; in doing so, the discussion progressed from less sensitive to more sensitive topics. Within each drug category, open-ended questions were posed regarding use patterns, effects of the drug, response when drug is unavailable, experience with quit attempts, whether there are reasons for concern, and interest in interventions for that particular drug.

Table 1.

Focus Group Topics and Illustrative Questions.

Topics	Sample Questions
Introduction and orientation	Purpose of focus group Confidentiality, respect for all opinions, honesty Reasons for audiotaping Reminder about duration of group
Contextual factors related to substance use	What kind of place are you living in right now? How do you support yourself financially? Do you have any family in the area? Are you currently in a relationship? How long have you been coming to the clinic?
Caffeine use	What types of caffeinated beverage do you drink? What happens when you drink something with caffeine? How do you feel when you have to go without caffeine? Should people worry about their caffeine use? Should the treatment staff talk about caffeine use?
Nicotine use	Do you smoke? What are some things that you like about smoking? What are some things that you do not like about smoking? If you smoke, have you ever tried to quit? Should people worry about their nicotine use? Should the treatment staff talk about nicotine use?
Alcohol use	When do you drink alcohol? What are some good things about drinking? Has drinking ever caused you problems? Have you ever tried to cut down on your drinking? Why do you think other people drink alcohol?

Nineteen focus groups (9 with men, 10 with women) were conducted in private rooms at outpatient clinics and psychosocial clubs. Group size was intentionally small to encourage individual participation; thus, we conducted 1 group of 5, 1 group of 4, two groups of 3, and 7 groups of 2. Due to no-shows, 8 patients were seen individually; in these cases, the same topic outline was used but adapted as an elicitation interview. The groups were facilitated by trained personnel with the gender of the facilitator matched to the group. Facilitators had experience in group facilitation and substance use assessment, and adopted a nonjudgmental stance. Refreshments were served throughout the session.

Data Management and Qualitative Analysis

All focus group audiotapes were transcribed verbatim. Raters reviewed the transcripts and assigned content codes to individual comments. This step was taken to facilitate identification of similar content across multiple transcripts despite variations in vocabulary and idiosyncratic verbal styles. The initial content categories reflected the conceptual framework that guided our elicitation research, namely, a biopsychosocial model of substance use and abuse. Raters allowed the response content to guide the development of unanticipated content domains; thus, the coding process included both top-down and bottom-up strategies. Next, two doctoral-level psychologists with substance use expertise reviewed the content categories and transcripts to organize the coded content categories into major themes (e.g., unusual use patterns, positive effects, negative effects, triggers, and efforts to control use). We identified themes in this paper if they were elicited in 2 or more groups (O’Brien, 1993). Representative quotes were then identified to illustrate the themes that emerged from the participant responses.

Results

Caffeine

Participants in 12 of the groups reported consuming large quantities of caffeinated beverages on a daily basis, such as 2-3 pots of caffeinated coffee, 3-4 20 oz bottles of Coke, or 2-3 pots of tea. It was also not unusual for participants to consume combinations of several cups of coffee along with caffeinated soft drinks. Table 2 summarizes common triggers for caffeine use as well as the positive and negative effects of caffeine that participants identified. Efforts to moderate caffeine use were mentioned spontaneously in 4 groups. These efforts included cutting down “because medications weren’t agreeing with coffee,” and substituting other beverages when possible.

Table 2.

Qualitative Summary -- Caffeine

Themes	Content	Representative Quotes
Triggers	Withdrawal symptoms (5) Dry mouth (2) Craving (2)
Positive Aspects	Combats drowsiness; wakes you up; increases energy; helps you think clearer (8) Likes taste, aroma (7) Relieves depression (2) Relaxing (3) Social contact, meet friends to drink coffee (2)	“I think I think clearer. When you’re really groggy … it gives you a little bit more energy.” “It gets the bad thoughts off my mind … better thoughts about the future and stuff.” “It relaxes my body the more coffee I drink. I get excited about the coffee. Then I react when I drink it. It relaxes me and makes my heart slow down.”
Negative Aspects	Makes you nervous, “hyper” (9) Insomnia (6) Stomach problems (4) Appetite increase; weight gain (4) Caffeine withdrawal symptoms: headaches; feeling tired, cranky, restless, depressed, anxious (8)	“I’d be all over the place, I wouldn’t be able to sit still for more than like 5 minutes.” “You can’t sleep at night.” “If I can’t get it, then I get all nervous, my nerves get all frazzled.”
Control Efforts	Substituting decaffeinated drinks; restricting the number of caffeinated drinks per day (3)	“I have cut myself down; I used to drink coffee throughout the day, but now I am only drinking about one cup a day.” “I try to only drink one cup of caffeinated a day, then the rest decaff.”

Note. Numbers in parentheses indicate the number of groups (out of 19) in which a topic was mentioned.

Nicotine

Nearly all participants smoked cigarettes. Both men and women described chain smoking (i.e., smoking 6-10 cigarettes, one after another, at a single sitting). For example, one woman stated, “I’d smoke 5 and a half packs of cigarettes that one day when I got fired. And I was really depressed. I didn’t even sleep for two days, you know, I just stayed up bugged-eyed and smoking cigarettes one after another.”

As depicted in Table 3, the management of negative affect plays a prominent role in both triggers for smoking and positive aspects of smoking. However, participants readily identified many negative effects of smoking, including long-term health risks, social awkwardness, expense of buying cigarettes on a limited budget, as well as aesthetic consequences of smoking. The frequency of quit attempts was an unexpected finding; one participant reported having quit between 25-30 times, and another “hadn’t smoked for a whole year.” Although the desire to quit smoking was common, participants expressed considerable ambivalence about their ability to quit. In fact, one participant mentioned that she wants to quit, but her doctor told her not to try, as they “don’t want another nervous breakdown!” The ambivalence felt by many of our participants is summarized in the following quote:

Table 3.

Qualitative Summary -- Nicotine

Themes	Content	Representative Quotes
Triggers	Feeling depressed, frustrated, upset (3) Seeing others smoke, smelling smoke (3)	“Usually when I get nervous, or tense or whatever, or anxious I’ll smoke.”
Positive Aspects	Relaxes your nerves; relieves tension (14) Something to do when bored (11) Like the taste, aroma, feel of inhaling (5) Like the stimulant effect, the “rush” or “buzz” (4) Social nature of smoking (4)	“I have large stress and I tell myself that it eases my tension and so it does. It calms me down.” “It relaxes me to a certain extent. And it stimulates me, you know. I use it for both reasons.” “I smoke when I don’t know what to do with myself … something to do with my hands.”
Negative Aspects	Health risks: asthma, heart disease, emphysema, lung cancer, second hand smoke (10) Buying cigarettes is expensive (8) Smoking hurts your lungs, you gag and cough, have a hard time breathing (7) Interpersonal tension: it bothers other people, others don’t respect you, it sets a bad example (4) Smell, dirty ashtrays, butts on the ground (3) Fire hazard (2) Nicotine withdrawal: agitated and irritable (5) Cigarettes seen as addictive (5)	“If I smoke around my kids, it’s bad. They are sucking in the smoke … They don’t need that in their lungs.” “I bum. I don’t like to bum, but I bum.” “It’s disgusting but I still do it.” “I’m just ashamed that I smoke.” “It’s very uncomfortable for me not to smoke. I’m very addicted to it.”
Control Efforts	Many quit attempts in the past (10) Desire to quit (7)	“It’s a terrible habit, but I’d go crazy if I didn’t smoke.” “I’ll quit when I get my life back to a balance point.”

Note. Numbers in parentheses indicate the number of groups (out of 19) in which a topic was mentioned.

“I get upset when my physician tells me I need to quit smoking. After that, I don’t want to go back to them. I’m afraid that if people started pushing, especially mental patients, because they have disorders, nervous disorders or something, but you know, sometimes as sad as this is to say, sometimes cigarettes are like the only thing people have to make them feel better, and if you’re going to take that away, they don’t want to hear it. At least I don’t.”

Alcohol

Many of our focus group participants had histories of heavy alcohol use or abuse, although they may not have been drinking regularly at the time of the study. Thus many participants could share their views of the pros and cons of drinking alcohol, as well as their experiences with controlling or limiting their drinking behavior. As seen in Table 4, social facilitation emerged as a primary positive aspect of drinking alcohol. The disinhibition provided by alcohol appeared to help individuals overcome social anxiety and/or social skills deficits. Many participants considered tension reduction and improved affect to be a positive result of drinking. A sizeable minority described using alcohol to distance themselves from their present circumstances. Nonetheless, serious negative consequences of drinking were also readily identified. Although intoxication did exacerbate psychiatric symptoms for some, more participants reported social and interpersonal costs that substantially affected quality of life.

Table 4.

Qualitative Summary – Alcohol

Themes	Content	Representative Quotes
Triggers	Having a bad day: bothered by psychiatric symptoms or negative emotions (5) Having money to spend (6) Hot day (3) Freedom from consequences: no commitments to work or school (3) In relationship with someone who drinks (3) Boredom (2)	“Like if I had a bad day I would have a drink. Like it was going to solve my problems for the day. I know now it don’t solve nothing.” “Triggers for me are stress, strain, emotions. You see, I don’t have control over my emotions, feelings, or thoughts. It’s part of the diagnosis. I don’t try to keep them in check. The first thing I’m going to do is pick up a beer.” “Every time I get off probation I start drinking again” “There is nothing else to try, you might as well drink all day.”
Positive Aspects	Social facilitation: you are more talkative, make friends more easily (9) Bars enable social contact (10) Relaxation, “I don’t care” attitude (6) Feel happier, more self-esteem (4) Helps you forget problems, numb feelings, forget your failings (5) Likes taste (3)	“I like to see what is going on in bars … a feeling of fitting in with people.” “made me outgoing when I felt shy.” “I like to relax, that’s why I do it” “you live today and forget tomorrow” “it helps me forget how I’m feeling”
Negative Aspects	Fights/arguments when drunk (5) Domestic violence/victimization (7) Unwanted sexual encounters (2) Symptom exacerbation: hospitalization, delusions, self-injury, suicidal ideation (5) Medical consequences: seizures, blackouts, passing out, vomiting, ulcers, chronic liver disease (5)	“I used to beat up my wife when drunk.” “Get locked up because of my temper. It makes my temper worse.” “One time I was sold to some people. I didn’t know where I was … Hammered.” “If I drink too much, I’ll think about jumping in front of cars or something like that.” “It really brings out my mental illness … terrible depression, sometimes I’d have fits of violence, sometimes I would hurt myself.”
Control Efforts	Many quit attempts, some for extended periods (7) Avoid mixing alcohol, medications (2) Having a job to go to helps limit alcohol consumption (2) Abstain to avoid other problems (2) Use self-help groups to cope with urges to drink (3)	“At any moment they can pop a blood test on us and end up in family court … I wouldn’t get to see my kids.”

Note. Numbers in parentheses indicate the number of groups (out of 19) in which a topic was mentioned.

Quit attempts were common, and participants reported various attempts to avoid problems related to alcohol. For example, one woman did not drink because a positive test for alcohol could result in her losing visitation rights for her children. Another woman stated that she does not drink because drinking leads to a craving for Valium, which she finds almost impossible to resist. Others described decisions not to mix alcohol and medications. Whereas it was clear that many participants felt vulnerable to the effects of alcohol, several expressed the opinion that “drinking is not bad” and that they could limit their drinking so as not to get drunk or sick.

Substance Use in the Context of Severe Mental Illness

For all three drugs, participants described the place of substance use in the context of their experiences. For some, substance use was one of many concerns, as illustrated in the following quote:

“I think that there’s more pressing things. For me, caffeine and cigarettes are the least of my problems, so I think that that’s probably why I think that they’re OK. I don’t really. They’re not OK, and eventually I would really like to stop, but the thing is, I think that if people were going to start a support group or counseling, whatever, I think that they would, their time would be better invested on something other than cigarettes.”

Some participants enjoyed drinking, felt that they appropriately managed the effects, and did not wish to give it up. However, alcohol abuse was considered an appropriate focus of intervention within the context of their psychiatric treatment:

“The only thing that I can say, that’s from my experience, is that if you have a mental health issue, if you’re coming to a place like this, that you really need to address a substance abuse issue if you have one. Because you’re not going to get better. My addiction and my mental illness feed right into each other. If I don’t take care of one without the other, I’m not going to make it. So, the thing is, I can go to all the therapy in the world, and leave here and go out and drink, and it’s worthless. Or else I can quit drinking and still have all these childhood traumas and things like that, and not deal with those, and I’m still not going to get better.”

Discussion

The primary purpose of our research was to use qualitative research methods to enhance our understanding of the functions served by licit substance use in persons living with a SMI. Our sample was recruited from the general outpatient psychiatric population; having a substance use disorder was not required for participation. Nonetheless, use of caffeine, nicotine, and alcohol was common in our sample, consistent with previous descriptive studies. Participants used these drugs in amounts and patterns that are likely to exacerbate symptoms of their mental disorder as well as increase their risk for other chronic physical illnesses. For all three substances, participants identified positive and negative consequences, and reported efforts to reduce or eliminate substance use.

The process of eliciting qualitative information from our participants varied in its resemblance to published focus group guidelines (e.g., Taylor, & Bogdan, 1984). We noticed much less cross-talk among participants and more of a tendency for participants to interact primarily with the facilitator. Some individuals were highly verbal and provided a lot of unprompted information; however, facilitators often took a more active role in keeping discussions on topic and eliciting multiple responses to leading questions. For example, participants may offer fairly simple, one-dimensional responses to a prompt such as “Has drinking alcohol ever caused any problems for you?” Knowing that alcohol use often has interpersonal, financial, legal, and psychological consequences, facilitators might use additional prompts to elicit more information, such as “Some people find that they have conflicts with other people, has that happened to you?” or “Has it ever changed your thinking or feelings?” Facilitators tried to prompt attention to multiple dimensions, but they did not suggest specific responses.

What did we learn from this qualitative study that we may have missed with a more structured, quantitative approach? First, we did not anticipate the extent of the interrelations between substance use and psychiatric status. Positive functions of caffeine use (e.g., to enhance arousal and reduce negative mood) parallel those identified by the general population (Graham, 1988). However, several perceived beneficial effects were more interrelated with psychiatric status, insofar as they were linked to medication side effects (e.g., drowsiness and dry mouth) and affect management. Smoking also appears to serve many important functions for these outpatients, including relaxation, social facilitation, and filling in unstructured time. Although most participants wished they did not smoke, few felt capable of quitting. Concerns about quit attempts provoking psychiatric destabilization may be justified (Glassman, 1993).

Second, most participants were generally adequately informed regarding the health effects of caffeine and nicotine use. However, few understood the addictive process, that is, how regular heavy use provokes withdrawal symptoms that are relieved by continued use. Because participants routinely consumed caffeine in large amounts, it is likely that many of the desired effects of caffeine were instrumental in reducing withdrawal discomfort. Nicotine was the only drug that was repeatedly identified as addictive by our participants.

Third, it became clear that the relative costs and benefits of caffeine and nicotine use must be considered in light of the psychosocial context of psychiatric disorders, and possible interactions with pharmacological treatments often used for those disorders. Based on the reports of our focus group members, the use of caffeine and nicotine do result in clinically significant concerns; thus more routine assessment and monitoring of these legal drugs appears warranted (see Larson & Carey, 1998). With a better understanding of the long- and short-term outcomes of caffeine and nicotine use for persons with SMI, we will be able to determine the priority of intervening to modify use of these substances of abuse relative to other problems that compete for attention from understaffed mental health settings.

Fourth, many motives for alcohol use were similar to those that would be found in samples without psychiatric impairment (e.g., Cooper, 1994): drinking is often used to facilitate social interactions, for tension reduction, to manage negative affect, and to enhance positive affect. We suspect that these perceived benefits may be enhanced for our participants because of their observable social skills deficits, limited opportunities for social intimacy, and high levels of negative affect. Participants recognized the costs of alcohol use to their physical health, interpersonal relationships, and emotional stability; clearly, the elevated rate of alcohol abuse in this population is not due entirely to ignorance regarding the risks of drinking. Rather, adults with a SMI remain vulnerable to alcohol’s effects but they have few alternative, productive activities with which to structure their time. This observation is consistent with the finding that substance use is greater among outpatients with relatively few competing sources of positive reinforcement (Correia & Carey, in press).

Fifth, we had not anticipated the extent to which participants reported self-control efforts and successful quit attempts. Some were based on external contingencies supporting abstinence (e.g., a steady job, probation, fear of losing visitation rights), whereas others were self-generated (e.g., restricting caffeine use, not wanting to mix alcohol with medication). This finding suggests that psychiatric outpatients do have resources to bring to the task of harm reduction, but that these may need to be supplemented with environmental supports. In other words, clinicians can help by mobilizing both internal and external sources of motivation for change.

We have argued previously that alcohol abuse, psychiatric disorders, and environmental contexts are interdependent (Stasiewicz, Carey, Bradizza, & Maisto, 1996). Antecedents of drinking events include environmental, cognitive, and affective variables. Alcohol use may then produce both positive and negative changes in psychological status. Similarly, antecedents of symptom exacerbation may involve interpersonal, environmental and substance use factors. Thus, interventions that focuses on both substance abuse and psychiatric symptoms are indicated for most dually disordered individuals.

Despite the prevalence and intensity of legal drug use among persons with SMI, relatively little information is available regarding the use of these drugs and their consequences in this population. Our qualitative findings suggest that psychiatric outpatients acknowledge both risks and benefits in their use of all three substances. These data begin to fill a gap in the behavioral treatment literature. Information about the positive effects identified by users helps to identify possible contingencies maintaining substance use. Similarly, information about perceived negative effects of substance use could identify concerns that could help to motivate patients to modify their use. The behavioral treatment literature indicates that identifying the functional relationships among antecedent and consequence variables is essential to modify drug use (Hester & Miller, 1995). Such functional relationship patterns are presumed to be idiosyncratic (e.g., Hayes, Nelson, & Jarrett, 1987). Therefore, an increased awareness of the functions and context of drug use among understudied populations such as the SMI may help to expand the application of behavioral treatments for the substance use disorders.

We acknowledge the limitations of the qualitative methodology used in this study. Our qualitative data analysis does not allow us to evaluate quantitative relationships among variables, nor can we make strong statements about the external validity of our findings. Our study was designed to be exploratory and heuristic, rather than confirmatory. We sought to extend our knowledge about perceived causes and consequences of licit substance use among the SMI so as to generate hypotheses for future research. However, we did confirm that a behavioral analysis of antecedents and consequences of substance use by persons with SMI will consist of a blend of content previously identified in non-psychiatric samples and considerations specific to persons with psychiatric disorders. Thus, the general behavioral literature on motives, decisional balance, and behavioral assessment of caffeine, nicotine, and alcohol use can serve as a point of departure for clinicians who desire to address these behaviors in psychiatric outpatients. However, we recommend careful consideration of the cognitive, affective, and social context of use among persons struggling with multiple disabilities; and an awareness of the relative importance of a variety of health concerns in this population.