The psychosocial rehabilitation unit at Arizona State Hospital, formerly known as the behavior management unit, typically houses 25 to 30 male patients diagnosed as having psychotic and mood disorders as well as comorbid personality and substance use disorders. Over the past several years the primary reason for admission was one or more episodes of violent or uncontrolled behavior that was perceived as dangerous to self or to others. Until recently, violent behavior by patients toward staff and other patients was an all-too-frequent occurrence on the unit.
To decrease aggressive and violent responses to provocative interpersonal interactions in the psychosocial rehabilitation unit, we adopted a behavioral skills training approach that had been used successfully to teach other skills in an inpatient setting (4). We organized a social skills group program that addressed several topics related to improving interpersonal interactions: learning to be polite and courteous; becoming aware of how one's behavior toward another person can influence that person's response; learning how to avoid personalizing another person's reaction; and strengthening appropriate conversational skills to reduce profanity, putting other people down, and reacting with aggression when staff members are unable to comply with requests. The aim of the program was to increase respect for oneself and for peers and staff. The fundamental principle underlying the program was that teaching adaptive and appropriate behaviors to individuals with serious and persistent mental illness would "displace" intolerable interactions.
The group program was also designed to give participants a chance to express their difficulties with interpersonal interactions openly and to discuss how they perceived and experienced life on the unit. The main focus of the group was on how problematic social interactions with staff and peers resulted in difficulties. For example, participants were helped to realize that an argument between two patients about who gets to use the public telephone first could escalate rapidly to physical aggression. Moreover, participants discussed the immediate and longer-term consequences of their aggressive behaviors—loss of privileges, unmet personal needs, impaired relationships, and longer hospitalizations.
Group role playing with coaching and feedback was encouraged, with the expectation that participants would be responsive to the difficulties faced by their peers. Interdisciplinary staff members were encouraged to participate in the group sessions with the goals of modeling courteous interaction, reinforcing friendly interactions among patients, facilitating personal communication, increasing relevant disclosure, and making use of situations from everyone's daily experience. As time passed, staff members began to understand the antecedents to patients' aggressive behaviors, and they learned to make adjustments in their own behavior and response style that improved patient-staff communication.
We used a modular approach with particular topics designed for each group session along with an open-ended discussion period. The module consists of 11 topics (see box), each of which is the subject of a weekly one-hour group session on the unit. Rather than describe each session of the program in detail, we will use the eighth topic, "What to do when the answer is no," as an example. A clinician's manual of the module is available on request from the authors.
The 11 topics of the interpersonal skills module
Respect yourself and other people
Ways that people like to be treated
Ways that I can treat someone better today
Ways to tell people that now is not a good time for me
Big problems versus small problems
Poor interactions with other people result in me not getting what I want
Using profanity doesn't help get my needs met
What to do when the answer is no
When other people are not responsive to us
I don't put myself up by putting other people down
Deciding on the best response to other people
Patients in the psychosocial rehabilitation unit commonly became frustrated and angry when staff members did not comply with their requests, and many on the unit tended to react aggressively or violently. The first part of the eighth session is devoted to a discussion of how people can control their emotional reactions when they are told that they will not be allowed to do as they wish or that a request will not be granted. Participants are asked to talk about or to write down three positive ways in which they could respond if denied a request. They practice using these responses with a peer, with one participant playing the role of a staff member and the other playing the person who must react to having a request denied. This session allows participants to expand their range of coping techniques to deal with disappointment or frustration. Their requests are not belittled; on the contrary, staff members and other participants are supportive of their needs and are willing to explain what is difficult about agreeing to problematic requests.
To assess the effectiveness of the skills training group program in reducing the incidence of aggressive behaviors, the treatment team—a psychiatrist, a psychologist, a supervising nurse, a social worker, a rehabilitation therapist, and a lead psychiatric technician—selected ten patients to participate in an evaluation. This group included many of the most verbally and physically aggressive patients on the unit. These patients often displayed rapidly fluctuating psychiatric symptoms, which affected their ability to maintain attention and to process information. Most of the patients required prompts and encouragement to maintain their attendance in the group sessions.
The patients completed a 12-item multiple-choice test before the module began and after it was completed to assess their grasp of the concepts presented during the group sessions. Separate surveys were completed by each patient's primary nurse and primary psychiatric technician before and one week after the module was completed to assess aggressive behaviors and response patterns. The survey consisted of ten questions answered on a 5-point scale ranging from 0, indicating not at all, to 4, indicating most of the time. The nurses and the psychiatric technicians were unaware of the purpose or content of the skills training sessions. Both instruments are available from the authors.
All ten patients were male. Their ages ranged from 21 to 67 years, with a mean of 38 years. Eight were Caucasian, and two were Hispanic. Four patients had a diagnosis of schizoaffective disorder, four of bipolar disorder, one of schizophrenia, and one of paraphilia. Five of the patients had a comorbid axis II disorder, two had a seizure disorder, and two had a comorbid substance use disorder.
Half of the patients had been in the psychosocial rehabilitation unit for more than a year before the study, and all had been in the hospital at least that long. Two patients had been in the hospital continuously for more than ten years. Although nine of the ten patients were taking antipsychotic medications and eight were taking mood stabilizer medications, each had at least one—and usually more—verbally aggressive and physically assaultive incident in the month preceding the study.
Overall, the patients demonstrated acquisition of concepts discussed in the group sessions by increasing the number of correct responses on their 12-item multiple-choice tests. The mean± SD score increased from 7.5±3.1 before the module to 9.5±2.6 afterward (t=−2.38, df=9, p<.05, two-tailed repeated-measures t test). The mean± SD score on nurses' surveys of aggressive behaviors decreased from 19.8± 8.4 one week before the skills training module to 15.4±7.9 one week afterward (t=2.27, df=9, p<.05). Five members of the group displayed enough improvement to be transferred to a less restrictive unit. Two of these patients eventually were discharged to community settings.
Because of the success of the pilot study, our skills training approach to the management of aggression has been replicated on several other hospital units. Several dozen patients have completed the 11-session cycle, some of them two or three times. To assure that the groups on other units maintain a behavioral focus—for example, use role playing rather than just discussing a topic—experienced skills trainers periodically attend the groups to model the techniques and provide ongoing supervision. By using these quality assurance methods and by limiting the number in each group to five or six patients, we have found that the skills training techniques can be provided with high fidelity on a variety of units, such as those for forensic, female, or developmentally delayed patients, and yet can be adapted to the goals of each unit and tailored to the needs of each patient.
Interacting with people in a civil, courteous, and nonthreatening manner is an important skill to learn for persons who experience persistent, severe symptoms of psychiatric illness and exhibit aggressive verbal and physical behaviors. The results of this pilot venture suggest that patients in psychosocial rehabilitation units can learn the information and skills necessary for increasing their prosocial behavior and can decrease the number of aggressive behaviors they exhibit on the ward. These social skills not only improve patients' prospects for hospital discharge but also increase their chances of successful community living.
Afterword by the column editors:
Behavior therapy programs on inpatient units such as the one Frey and Weller describe are a critical component of a system of care designed to facilitate successful community living for individuals with serious mental illness. However, remediating behavioral problems such as verbal and physical aggression in this patient group involves the same persistent application of a corrective teaching environment as remedying the client's neurochemical disturbances with medications (9). One way to do so is to train staff to use a "teaching interaction" that can be employed in vivo, 24 hours a day and seven days a week, whenever an aggressive or preaggressive interaction occurs.
In the teaching interaction, the staff member intercedes before, during, or immediately after an aggressive act. The staff member calls a time out and asks the offending patients to describe what happened in the problematic situation. If necessary, the staff member clarifies the event in a factual manner—for example, "He accused you of taking his cigarettes, and you got angry and pushed him." Then the staff member asks both patients to articulate alternative ways that they could have handled the provocation. If the patients cannot conceive of alternatives, the staff member offers prosocial examples to choose from—for example, "I could have gone to the nursing station and asked for help."
Then the patients role play one or more of the alternatives with coaching and reinforcement from the staff member. As a final step, the staff member asks the patients, each in turn, to recite what they have just learned from the teaching interaction and how in the future they might employ the alternative just practiced.
The continual use of the teaching intervention to remediate aggression and to instruct patients on prosocial alternatives meets the need for interventions that can deal with the spontaneous, ongoing events on a ward. Another example is the use of a time-out area on the unit, enabling patients to cool off without receiving any staff or patient attention after a preaggressive or aggressive act. In one study of the use of a time-out area, which was a chair facing a wall at the end of a ward's corridor away from the social center of the ward, 15 minutes of time out resulted in containment of more than two-thirds of aggressive behaviors without the need for medications or physical seclusion, restraint, or locked time out (10). The behavioral learning procedures described here as well as in other reports (11), along with inservice training programs to equip staff with the proactive competencies needed for interacting positively with aggressive patients, should help psychiatric professionals and organizations respond to the regulations on physical restraint and seclusion issued recently by national regulatory agencies.
Skills trainers must also bear in mind the potential for the transfer of skills learned in the ward atmosphere to less controlled settings in the community. Generalization strategies might include overlearning, in vivo practice, and teaching of problem-solving methods to enhance the durability of gains made.
Dr. Frey is a psychologist at Arizona State Hospital, 2500 East Van Buren Street, Phoenix, Arizona 85008, and Ms. Weller is a doctoral candidate in clinical psychology at Arizona State University. Robert Paul Liberman, M.D., and Alex Kopelowicz, M.D., are editors of this column.