A Group Cognitive-Behavioral and Process-Oriented Approach to Treating the Social Impairment and Negative Symptoms Associated With Chronic Mental Illness

Abstract

Major changes in health insurance have brought challenges in managed mental health care services to the forefront. Given slim capitation margins, brief treatment with proven efficacy has become the standard. These developments have significant ramifications for the treatment of chronic mental illnesses, which often require lifelong treatment. Efficacious short-term group treatment may help fill the gap between quality of care and the ideal economic allocation of mental health care services for people with chronic mental illnesses. The present treatment outcome study was designed to determine the extent to which Interactive-Behavioral Training (IBT)— a group psychotherapy model that actively combines cognitive-behavioral and group process techniques—can provide significant gains for people who suffer from chronic and debilitating social impairment and negative symptoms.

Social impairment is widely believed to play a major role in the pathogenesis as well as the course of chronic mental illness.^1–3 Some researchers, for example, view social dysfunction as the “most debilitating and refractory aspect of psychopathology” for the chronically mentally ill.⁴ Numerous investigations have also demonstrated clear overlaps in the expression of negative symptoms (such as diminished emotional range, insufficient social drive, social withdrawal, and poverty of speech) and those symptoms that are markers for social impairment.^4,5 Moreover, social skills deficits not only seem to persist over time, but also are more resistant to treatment when they are secondary to negative symptoms.^6,7

Behavioral and cognitive-behavioral treatment (CBT) approaches to social skills training are used extensively in treating the social impairments found in people with severe mental illness.^4,6,8–12 Many studies appear to demonstrate that by learning social cue recognition, appropriate behavioral responses, and how to generate alternative solutions to interpersonal problems, the individual can develop significantly enhanced overall social competence,^1,2,13 which is also assumed to provide positive long-term generalizable effects. However, treatment outcome studies reveal that enhancing specific cognitive-behavioral social skills components does not necessarily lead to enhanced interpersonal relationships outside of the treatment setting.^1,6,14,15 These studies have shown that for people diagnosed with schizophrenia, the treatment effects of current CBT models of social skills training are notably inconsistent with respect to generalization and impact on relapse rates.^l6–19 For example, more recent findings by Penn and Mueser¹⁷ show that while behavioral skills may be enhanced by social skills training, improvement in symptoms and community functioning are less prominent. Moreover, many investigators have found that cognition skills training has little effect on psychiatric relapse,^16,17,20 since between 35% and 50% of community service–connected people with a chronic mental illness relapse within a year and between 68% and 85% relapse within 5 years of a psychiatric hospitalization.^21–26

A review of the literature also reveals that negative symptoms (estimated to affect between 30% and 60% of people with chronic mental illnesses²⁷) significantly interfere with social learning and thus lead to further resistance to standard cognitive-behavioral social skills training.^7,28 Unsuccessful interpersonal interactions, poor coping skills, and high relapse rates reflect the adverse impact of negative symptomatology on social functioning and its relationship to social skills learning for the chronically mentally ill. Mueser et al.,⁵ for example, found that negative symptoms were highly associated with poor social functioning and quality of life in chronically mentally ill populations. Although the presence of negative symptoms in schizophrenia and schizoaffective disorders is widely accepted as a significant barrier to pharmacological and psychological treatments,^6,7,27 current cognitive-behavioral social skills training models generally do not target actively the goal of negative symptom reduction.

It then seems likely that CBT social skills training outcomes are mixed in part because current models do not adequately examine how to motivate group members so that skills learned in training can be internalized and integrated into modifiable social models, which could then be used outside of the treatment environment. The abilities to internalize social skills, interpret instances of related knowledge, and integrate new information should lead to positive long-term generalizable improvement in overall social functioning and quality of life. The establishment of social skills training treatments that focus on the active facilitation of engaged relationships among group members should create viable models of social interactions while actively targeting negative symptom reduction. Many studies show that the facilitation of group process (through means such as altruism, group cohesiveness, self-disclosure, and instillation of hope) leads to enhanced interactions and engagement among group participants. Yalom²⁹ also found that for people with severe and persistent mental illnesses, emergence of group process factors was followed by long-term positive effects that improved overall social functioning. Although many CBT social skills training models do not actively explore how and to what extent group process factors should be fostered, an important modification of standard social skills training models may be the active inclusion of interpersonal group process strategies to facilitate motivation for learning and interpersonal connections between group members. This approach should, in practice, reduce negative symptomatology and significantly improve social skills acquisition.

The present study assesses the efficacy of Interactive-Behavioral Training (IBT),³⁰ an approach to social skills training with a combined focus on cognitive-behavioral techniques (such as instruction, modeling, and behavioral rehearsal) and group process strategies. The IBT format directly facilitates therapeutic group process and uses established cognitive-behavioral strategies. The blending of cognitive-behavioral and group process interventions is therefore postulated to increase motivation for learning, improve social skills acquisition, and enhance overall social competence. It appears, then, that IBT should improve the effectiveness of current social skills training models, inasmuch as therapeutic group process itself reduces negative symptoms by increasing motivation for social learning. Through the use of cognitive-behavioral and interpersonal group process strategies, group members may learn to participate fully and to act as vehicles for social learning by serving as clarifiers of affect, reality testers, and interpersonal behavioral and problem-solving models. Thus, the combination of cognitive-behavioral and interpersonal group process strategies^29,31 may offer the most comprehensive and dynamic social skills treatment package. The application of cognitive-behavioral social skills techniques with group process strategies may then fill the current gap in outcome research between social skills and social competence.

Preliminary data³² comparing the efficacy of IBT with standard social skills training³³ suggest that IBT training results in positive changes in psychiatric symptoms and social functioning. Clinical observations also provide encouragement for further examination of the IBT model. The following hypotheses were tested:

1. IBT will increase the overall social competence of people with chronic schizophrenia and schizoaffective disorders.

2. IBT will improve the negative symptoms that are often associated with poor treatment outcome for people diagnosed with schizophrenia or schizoaffective disorder.

3. IBT will facilitate the emergence of those therapeutic group process factors found to enhance social competence in people with chronic schizophrenia and schizoaffective disorders.

Methods

Subjects

Subjects were recruited from the Long Island Jewish Medical Center/Hillside Division Ambulatory Outpatient Clinic and the Adult Continuing Day Treatment Program. These sites were selected because of their high utilization rate during the recovery process for those with severe and persistent mental illnesses.

All willing patients who met the DSM-IV³ diagnostic criteria for a schizophrenia or schizoaffective disorder were screened and evaluated by a doctoral-level clinician and experienced diagnostician. Patients with medication noncompliance as a current and clinically significant problem or with a history of alcohol/substance abuse or dependence in the preceding year were excluded. Those with a history of moderate to severe neurological impairment or mental retardation as documented in medical records were excluded, as well as those who were significantly psychiatrically unstable as defined by scores of 5 (maximum score per item = 7) or more in any of the following Positive and Negative Syndrome Scale³⁴ domains: conceptual disorganization, hallucinatory behavior, or unusual thought content.

After being screened and giving informed consent, 40 patients (27 men and 13 women, mean age = 33.7 years, age range 19–61) were included in the study. Twenty patients were receiving outpatient treatment from the Adult Continuing Day Treatment Program, and 20 were enrolled in the Ambulatory Outpatient Clinic. The total sample (N = 40) included the following diagnostic categories: paranoid schizophrenia (n = 24), schizoaffective (n = 12), undifferentiated schizophrenia (n = 3), and catatonic schizophrenia (n = 1). Mean age of illness onset was 21.56 years (SD = 9.25, range 12–36 years), and total number of hospitalizations was 3.26 (SD = 2.56, range 0–10).

Assessment

Clinical rating scales were administered to measure changes in social functioning and negative symptomatology. Each participant was rated by the same trained single-blind rater (interrater reliability = 0.92) at each assessment stage. The following instruments were administered to each participant at baseline and immediately following treatment:

The Clinical Global Impressions (CGI) scales³⁵ are widely used in schizophrenia research because of their ease of administration and their validity as a measure of global functioning. The CGI-S was used at baseline to clinically assess global illness severity. The CGI-I was used to assess global clinical improvement at posttest.

The Quality of Life Scale (QLS)³⁶ is an instrument for rating the schizophrenic deficit syndrome and was developed to evaluate major areas of functioning. It is widely used and has been shown to have adequate reliability and validity. This scale has several subscales: interpersonal relations (IR), instrumental role (IN), intrapsychic foundations (IF), and common objects and activities (COA). Overall, the QLS assesses interpersonal relations and social network, occupational functioning, and role functioning. The QLS was used to assess changes in the patient's quality of life experiences over the course of the study.

The Modified Scale for the Assessment of Negative Symptoms (SANS)³⁷ is a modified scale used in the assessment of negative symptoms for clinical trials. The scale assesses the severity and duration of negative symptoms with subscales assessing the severity of affective flattening or blunting (AF), alogia (AL), avolition-apathy (AA), and asociality (AS). The SANS was used to assess changes in negative symptoms from baseline to posttreatment.

The Behavioral Assessment Task (BAT) ³⁸ is a four-part videotape of analogue scenarios that has been validated with schizophrenic populations. This assessment tool addresses the accuracy of social perception, the transformations or mental processes involved in social interactions, and the behavioral responses likely to occur. Based on standardization trials, this assessment tool has adequate psychometric properties and has been found to be effective in distinguishing groups with known differences in social skills.^38,39 The first two scenarios are comparable and were individually administered to all participants at baseline and posttraining. After presentation of each scenario, each participant was asked to respond to a set of nine questions regarding what she or he had viewed. Questions 1 through 4 assessed whether the participant encoded and received relevant details of the scenario. Questions 5 through 9 assessed the participant's ability to recognize that a problem existed, evaluated how problem recognition was executed, and assessed the participant's skills in defining and generating alternative solutions to the problem.

The Global Assessment of Functioning Scale (GAF) ³ is a 90-item scale used to assess overall psychosocial functioning and symptom level.

The Brief Psychiatric Rating Scale (BPRS) ⁴⁰ is an 18-item scale used to assess current psychopathology including thought disorder, delusions, and hallucinations.

The Positive and Negative Syndrome Scale for Schizophrenia (PANSS) ³⁴ is a 30-item, four-scale instrument that measures both positive and negative symptoms along with general illness severity. This instrument has been shown to have reliability as well as criterion-related and predictive validity. In the current study, the PANSS instrument was used as a screening tool to assess for the presence of positive symptoms.

Treatment

Following the baseline assessment phase, the 20 patients from each of the two sites were randomly assigned to a treatment group (n = 10) or a waitlist group (n = 10). Each of the two treatment groups followed a 16-session format, meeting for 50 minutes per session. Each group met twice per week and was led by two leaders. The group leader was required to have a minimum of 3 years' supervised IBT group trainer experience. The co-leader was a hospital clinical staff member with a minimum of 12 years of experience working with chronic mentally ill populations. Each session was videotaped, using a camera positioned unobtrusively in the group training room.

Each IBT session³⁰ was divided into four stages. In the Orientation and Cognitive Networking stage, the leaders encouraged and facilitated social interactions among group members. For instance, the leader might make the following statement: “Ruth, your comments seem to suggest that you understand the issue John is raising. What might John be feeling about the issue?” This stage began as a 10- to 15-minute segment. However, with the emergence of group process factors (such as altruism and self-disclosure), this stage averaged 3 to 5 minutes by session 10.

During the Warm-up and Sharing phase, the second stage of the training model, there was a strong emphasis on self-disclosure. It was during this 15-minute phase that members were encouraged to share concerns or personal issues with other group participants. During the 20-minute Enactment phase of the group, the individual enacted an interpersonal situation that included group members as active participants. This stage involved five basic elements: 1) selecting a participant, 2) assigning an auxiliary, 3) using an interpersonal group process technique, 4) directing an encounter, and 5) using cognitive-behavioral strategies. An average of three enactments per group session became typical over the course of treatment.

In order to foster the therapeutic group process factors typically associated with positive treatment outcome (such as cohesion, universality, and learning/modeling),²⁹ we included the following interpersonal group process techniques³¹ among those used during the Enactment phase: future projection, role reversal, and doubling.

With doubling, the participant (protagonist) is encouraged to express feelings evoked by an interpersonal situation. A group member is then asked to represent and establish identity with the protagonist by verbalizing what the group member feels that the protagonist is feeling. The group member confirms the accuracy of his or her feeling statements by “checking in” with the protagonist. In multiple doubling, more than one group participant is asked to identify with the protagonist. This creates multiple perspectives for both the protagonist and other group members and provides the protagonist with the feeling of being understood and supported.

The “role reversal” technique provides both role clarification and reality testing as the protagonist is asked to “step into the other's shoes.” Finally, with “future projection,” the protagonist acts out how she wants her future to shape itself by selecting a point in time, a place, and the people with whom she expects to be involved at that time.

Following the use of group process techniques, the individual was given affective, behavioral, and cognitive feedback. He or she, along with other group participants, gained practice in cognitive-behavioral strategies including modeling and behavioral rehearsal. The group also generated alternative solutions to the interpersonal problem presented, and the individual then role-played or enacted the new response. Overall, techniques introduced in the Enactment stage offered group members skills training in social cue recognition, empathy, self-awareness, reality testing, creation of boundaries, insight, self-efficacy, interpersonal behavioral, and problem solving.

Finally, the Affirmation stage took place during the last 5 minutes of the group. The leaders and group members specifically identified and verbally reinforced socially competent behaviors displayed by individual members in the group. For example, self-disclosure was seen as a positive step toward social relatedness and involvement, and it was reinforced as follows: “Mary, many of the members seemed to relate to what you were feeling about taking medication. Sharing your personal feelings helped other group members to open up and share their feelings as well.” This phase also allowed for positive emotional closure of each session, which itself further enhanced group process (by fostering universality and cohesiveness) and increased motivation for participation.

Results

Outcome Measures

1. IBT will increase the overall social competence of people with chronic schizophrenia and schizoaffective disorders.

Enhanced social functioning was defined as overall clinical improvement in CGI, GAF, BPRS, BAT, and QLS scores. Six of the 40 participants did not complete the study and were therefore excluded from the following analysis. At baseline, the entire sample showed notable functional impairment in overall social functioning. The means and standard deviations of all clinical scales for the sample are given in Table 1. Study participants scored in the moderately impaired range of social functioning as measured by the QLS (mean = 52.45) and BAT (mean = 7.77). Because the screening process excluded those in the severe range of psychopathology, BPRS mean score (34.77) was in the mild to moderate range. Assessment of global functioning, as measured by the GAF (mean = 43.91) and CGI (mean = 3.82), showed moderate overall functional impairment for all participants.

TABLE 1.

Posttreatment scores revealed that global functioning, as measured by the GAF (mean = 45.70), showed significant improvement for the treatment groups (F = 4.32, df = 1,31, P = 0.046). As shown in Table 1, there was a significant increase (treatment groups mean score at baseline = 43.88; posttreatment mean = 50.83) in overall psychosocial and occupational functioning. The waitlist group showed a nonsignificant increase on the same measure (mean score = 43.94 at baseline vs. 45.13 at posttreatment assessment).

Scores on the BAT (F = 3.01, df = 1,27, P = 0.094), CGI (F = 2.29, df = 1,31, P = 0.140), QLS (F = 2.02, df = 1,31, P = 0.166), and BPRS (F = 3.39, df = 1,31, P = 0.075) were not significantly different following treatment. However, posttreatment scores on all measures showed overall improvement for the treatment groups, suggestive of a positive trend toward enhanced social functioning. In addition, although the difference was not statistically significant, the QLS-IR, a subscale score used as an assessment of the quality of interpersonal relationships, demonstrated improvement for the treatment groups (baseline mean = 2.62, posttreatment mean = 3.10; F = 3.49, df = 1,31, P = 0.071).

2. IBT will improve the negative symptoms that are often associated with poor treatment outcome for people diagnosed with schizophrenia or schizoaffective disorder.

Improvement in negative symptoms was defined as an overall reduction in total SANS and subscale scores. At baseline, SANS (mean = 49.80) scores for both groups were in the moderately severe average range. Following training, the treatment groups showed a nonsignificant decrease in total SANS score (baseline mean = 46.41; posttreatment mean = 42.53). There was also a nonsignificant decline in the asociality subscale score (SANS-AS) for the treatment group (F = 3.37, df = 1,31, P = 0.076). The waitlist group showed little overall and subscale score change.

Process Measure

3. IBT will facilitate the emergence of those therapeutic group process factors found to enhance social competence in people with chronic schizophrenia and schizoaffective disorders.

On the basis of operationally defined ratings of each videotaped session (interrater reliability = 0.90), raters coded whether operationally defined verbal and nonverbal expressions of established therapeutic group process factors occurred (Table 2): both treatment groups were successful in facilitating the emergence of numerous group process factors. The following group process factors showed the largest absolute scores over the course of treatment for the Adult Day Treatment Program group: acceptance/cohesion (26), universality (58), guidance (33), vicarious learning and modeling (31), self-disclosure (91), and the development of socializing techniques (21). The Ambulatory Outpatient Clinic group showed similar results: cohesion (19), universality (38), guidance (26), vicarious learning and modeling (45), self-disclosure (89), and the development of socializing techniques (35). Self-understanding, catharsis, instillation of hope, corrective recapitulation of the primary family, altruism, imparting of information, and existential factors did not emerge with similar frequency, averaging 3 to 4 occurrences over the course of treatment.

TABLE 2.

Discussion

The first hypothesis was that IBT would lead to an increase in overall social competence for people diagnosed with schizophrenia or schizoaffective disorders. GAF scores showed significant improvement for the treatment groups only. The treatment groups also improved across the board on all outcome measures. The significant increase in GAF scores suggests that measuring specific cognitive-behavioral components (as is typical in social skills training outcome studies) may reflect a limiting approach to addressing the dynamic process of interpersonal learning and functioning. Rather, social functioning may reflect an inclusive and interdependent assemblage of components that, combined, results in global social competence.

In testing the second hypothesis, the effects of IBT on negative symptoms were examined. Although nonsignificant, total SANS scores showed positive changes for both treatment groups. Similarly, SANS asociality subscale scores approached statistical significance for those who participated in the treatment groups (Table 1). Positive changes in SANS scores, combined with significant GAF score changes for the treatment groups, lend further credence to the proposition that a reduction in negative symptoms increases overall psychosocial and occupational functioning.

The third hypothesis allowed us to explore whether IBT facilitates the emergence of therapeutic group process factors. Both treatment groups showed cumulative occurrences of numerous therapeutic process factors. The results of this study suggest that the IBT model is a salient facilitator of many of the therapeutic group process factors found by researchers to be significantly instrumental in enhancing overall social functioning.²⁹ Because of the relatively short length of treatment,^29,31 factors such as “recapitulation of the primary family” and “existential issues” did not emerge with the same frequency. Although there is little research exploring what is “good enough” process for people with chronic mental illnesses, this study points to the IBT model as a step toward quantifying group process in a meaningful way.

In general, the data support all three hypotheses and illustrate that IBT, by its structure, may create the conditions within which social skills can be enhanced and translated to psychosocial and occupational areas. The IBT model may provide individuals with a learning prototype in which to integrate the skills learned in training into numerous social situations beyond those practiced in group sessions. Overall, these findings suggest that the IBT approach, seemingly irrespective of outpatient treatment site, promises to more succinctly and successfully address the multidimensionality of social functioning and competence and to offer an inclusive approach to the reduction of negative symptoms and the acquisition of social skills.

Future studies should evaluate the efficacy of offering IBT on a weekly basis, a treatment consistent with most insurance and managed care regulations. Recent investigations have begun to explore the roles of verbal memory, executive functioning, vigilance, and other neurocognitive deficits found in schizophrenia in social skills acquisition and social competence.^16,17,20 Once specific neurocognitive processes are adequately identified and linked to functional outcomes (such as social and occupational functioning), the IBT model, because of its inclusive and focused structure, may be useful in providing an integrated cognitive remediation and social skills training model. Lastly, and as part of investigating whether IBT treatment gains are maintained, follow-up assessments of at least 6 months posttreatment should be administered. In addition, other cognitive-behavioral and interpersonal process strategies may be introduced to assess whether additional factors emerge, enhance factors already present, and improve generalization and relapse rates. Negative symptomatology may also be measured at intervals (such as every 4 weeks) to monitor stability over time and to further assess which group process factors may be more or less instrumental in reducing negative symptoms.

Acknowledgments

The author thanks Simca Pollack, Ph.D., for his statistical expertise, and also thanks Daniel Tomasulo, Ph.D., Samuel S. Siris, M.D., Gregory Hinricksen, Ph.D., Steven Mattis, Ph.D., Doris Francis, M.S., Maryann Hamilton, R.N., Lenore Goldstein, R.N., Barbara Linden, R.T., Susan Rose, C.S.W., Jeffrey Ponnell, C.S.W., Sandra Kalman, C.S.W., and Heather Perrin-Boyle, M.S. Special thanks go to the study participants and Ellen Keller, Psy.D. at the Young Adult Institute for her continued support in providing excellent IBT group trainers: Judith Bailey, Gloria Lindsay, and Bobra Fyne. This article is based on research supported by a grant to the author from the Long Island Jewish Medical Center Faculty Research Competitive Pool Grant Program.