Changes after PBSP
group therapy
Changes in symptoms, interpersonal problems and
self-concepts after participation in
Pesso Boyden System Psychomotor group therapy
Trondheim, Norway

P A T R I C K A . V O G E L
O D D H A R A L D R Ø K E N E S,
Trondheim, Norway

Summary
This is a non-experimental study of the effectiveness of a form of group therapy, Pesso Boyden System Psychomotor (PBSP). Twenty-eight outpatients were assessed on questionnaires prior to their participation in five different PBSP groups, and at fixed intervals during the semi-open groups. Significant reductions on standardized measures of psychiatric symptoms and interpersonal problems were found, as well as improvements in self-concepts. The size and clinical significance of these results were comparable to those obtained by other types of therapy.

It has been said that there are hundreds of forms of psychotherapy available for the treatment of disturbing psychological symptoms, dysfunctional interpersonal patterns of behavior, and distorted perceptions of the self and other. It is difficult for therapists or patients to judge the relative effectiveness of these different treatment methods. Personal affinities for certain theories or the recommendation of colleagues or friends is often the guide used for selecting a given treatment. Today there seems to be an increasing demand that therapies demonstrate that they do produce the types of changes that are deemed desirable by patients and by other interested parties like third-party reimbursors of treatment costs.

Since the early criticisms of the field of psychotherapy research (Eysenck, 1952), there has been a great increase in the amount and sophistication of psychotherapy research that has been conducted. Several types of therapy have been shown to be efficacious in the treatment of specific problems, and are considered to be evidence-based (Barlow & Hofmann, 1997; DeRubeis & Crits-Christoph, 1998). These therapies have been primarily short-term individual therapies. These efforts have contributed useful knowledge to the field about which psychological treatments have proved efficacious in the treatment of depression, obsessive-compulsive disorder, panic, and eating disorders, among others.

Some criticize this approach on philosophical grounds that exclude human problems from the scope of natural science methods of inquiry (Rønnestad, 2000; Strupp, 1982). Others from a psychodynamic persuasion have criticized the diagnositic systems (DSM-IV, ICD-10) that serve as a crucial basis for the evidence supported therapies (Jacobson & Cooper, 1993). More pragmatic criticisms have come from clinicians that have not felt that the stringent methodology of treatment efficacy trials is applicable to their clinical reality (Persons, 1991).
The lack of studies on therapies of a long-term nature leads to their being underrepresented in lists of evidence supported treatments. The statistical complexity of performing rigorous randomized clinical trials (RCT) on these types of therapies is a significant obstacle preventing more of these therapies from being assessed according to the criteria for an evidence supported treatment (Seligman, 1995). In addition, the high degree of control in RCT’s makes them less generalizable to treatment as actually performed in the field. Seligman proposes the use of effectiveness trials as an alternative methodology to make the study of long-term therapies more feasible.

Another form of therapy notably lacking from the lists of evidence supported treatments are many forms of group therapy that are of a long-term nature. Most group treatment studies are about short-term cognitive behavioral interventions. One recent survey of the treatment literature found only six studies dealing with long-term group psychotherapy (Lorentzen, 2000). Lorentzen employed an effectiveness methodology to study the effects of long-term group psychoanalytic group therapy (Lorentzen, Bøgwald, & Høglend, 2002).
Lorentzen and his colleagues studied 67 outpatients participating in an average of 32.5 months of group psychoanalytic therapy. Only 2 patients terminated treatment with less than a minimum of 6 months of weekly therapy sessions. Participants produced very large reductions in symptomatic distress and interpersonal problems. The effect sizes averaged 1.4 across three outcome measures (two self-reports and one independent interview rating). The lack of control group inflates these effect sizes, but they were very comparable to those found in other studies with higher dropout rates and effect sizes of around 0.7 on similar outcome measures (Budman, Demby, Soldz, & Merry, 1996; Piper, Debbane, Bienvenu, & Garant, 1984).

Psychic trauma, often in relation to negative experiences with important attachment figures, seems to play an important role in the development of many types of psychological disorders (Weisaeth, 1998). One form of group therapy, Pesso Boyden System Psychomotor (PBSP), has been developed that utilizes empathic observations of body language to facilitate expression of these previously unresolved conflicts and emotional scars usually in connection with early attachment-related experiences (Pesso, 1969, 1997; Pesso & Crandell, 1991). Some group exercises are used to increase sensitivity to body cues. Carefully controlled psychodrama “scenes” are constructed at the client’s direction and pace. According to PBSP theory the client is a conscious collaborator in finding “new solutions” that are meaningful and satisfying emotionally as well as cognitively. This gives the client a “new map” for behavioral pattern changes that can result in lower levels of symptoms, more satisfying relationships, and improvements in self-concepts.

Although widely practiced in several countries, PBSP has been subjected to little research. A few studies have showed that it produces improved self-concept, and a decreased need for social desirability (Foulds & Hannigan, 1974, 1976).

In the present study, it was hypothesized that completers of PBSP group therapy would show significant improvements in their levels of psychiatric symptoms, interpersonal problems, and in important aspects of their self-concepts on standardized self-report measures.

Method
Subjects were recruited from PBSP group therapy patients in psychiatric outpatient clinics or in private practice settings. Thirty-four group participants in 5 different semi-open groups completed self-report questionnaires prior to their participation. Therapists also completed a questionnaire about each group member prior to the inclusion of the new member. Groups planned to meet for one year. After one-year it would be decided if participants wished to continue in a new group. Twenty-eight subjects (82%) completed at least one-half year of group therapy (mean = 9.1 months, SD = 5.7, range = 3 – 18 months) and completed the questionnaires again when leaving the group. Participation in the study was voluntary and patients were promised full anonymity.

Treatment procedure
Forty-three individuals were offered treatment. Four individuals refused treatment and did not attend any sessions. Five individuals refused to participate in the study, but participated in the groups. All group members were interviewed by their therapist prior to acceptance in the group. As these were patients in routine clinical settings, all had a ICD-10 psychiatric diagnosis. The most common principal diagnosis was depression (n= 17, 50%), followed by personality disorders (n=6, 18%), anxiety disorders (n=5, 15%), while four patients (12%) had no symptom or personality diagnosis. There was missing data as to diagnosis for two patients (5%).

Most sought group therapy to work on recurring symptoms and problems with work or relationships. Patients with psychoses, active substance abuse problems, suicidality, or impulse control problems that could interfere severely with regular participation in the group meetings, were excluded. Patients had to commit to attending group sessions for at least one-half year. Group sessions varied from two to four hours in length and were held weekly or bi-weekly. A plan allotted group time to two or three individuals during each session.

Therapists
Four male therapists led the groups. One group had two of the therapists as co-therapists who equally shared group responsibilities. The other two therapists led two groups each alone. All the therapists were well experienced in the PBSP-method. They had completed a 3-year intensive training program led by the founder of the therapy method Albert Pesso. Afterwards they have regularly received over 6 years of supervision on videotape material from on-going groups that they have led from Albert Pesso and other certified trainers in PBSP.

Statistics
Paired t-tests were administered comparing each individual’s introductory ratings on the self-reports with the ratings they received when they had left the groups. Because of the exploratory nature of the study a conservative significance level (p < .01, one-tailed) was chosen to minimize chance findings.

Measures
SCL-90-R Derogatis, 1983)
The Symptom Checklist-90-revised is a widely used self-report instrument. It measures 10 subscales and one total index (GSI) of psychiatric symptoms. It asks informants to rate their recent level of disturbance on a five-point rating scale (0 = none – 4 = extreme) on 90 questions. A score of less than .40 is considered asymptomatic. Scores over .99 indicate moderate symptoms often requiring treatment.

IIP-C (Alden, Wiggins, & Pincus, 1990
The Inventory of Interpersonal Problems-Circumplex version assesses relationship problems on eight subscales and yielding a total score, that often improve as a result of psychotherapy. Informants are asked to rate their degree of problems in relation to a significant other on a five-point rating scale (0 = none – 4 = extreme) on 64 questions. A score of less than 1.189 is considered asymptomatic.

SASB-Introject Version (Benjamin, 1974)
The Structural Analysis of Social Behavior-Introject-Version assesses the degree of self-affirming versus self-critical self-concepts, through eight subscales. Composite scores measuring Self-attack (ATT) and Control (CTR) are also derived. Informants are asked to rate their agreement with statements about their self-worth on 36 items on a 11-point rating scale (0 = total disagree – 10 = total agreement). Norms for normal ratings of self-concept are available.

Results
Table 1 describes the demographic characteristics of the treatment completers. Chi-square and t-tests performed between the characteristics of the completers versus the dropouts found no significant differences. There were also no significant differences between completers and dropouts found on the self-report measures.

Table 1. Demographic characteristics of completers in means
(sd/range) or n (%) N = 28

Age                                    37.1 years (10.2/22-60)
Sex                                     10 men (36%)  18 women (64%)

Civil Status
Single                                  8 (29%)
Married                               6 (21%)
Cohabiting                          5 (18%)
Separated/Divorced             9 (32%)

Education
Elementary School                1 (3%)
High School                       10 (36%)
University/College             14 (50%)
Other (Trade Schools)          3 (11%)

Employment Status
Student                               7 (25%)
Employed (Full-time)        12 (43%)
Sick Leave                           4 (14%)
Disability/Rehabilitation    5 (18%)

Figure 1 describes the group members global level of functioning prior to their group participation (GAF scores). The median scores of patients at the time of entry in the groups corresponds to a level of functioning characterized by marked impairment in either social relations or occupational functioning, or moderate impairment in both

Figure 2, 3, and 4 depicts the results of the paired t-tests for the completers of treatment on SCL-90-R, IIP-C, and SASB.

 

 

Table 2. Selected Effect Sizes (Cohen 1988) for SCL-90-R GSI, IIP-C Total, SASB Clusters

Scale                                  Effect Size (relative size)
SCL-90-R GSI                    .57                       medium
IIP-C Total                        .58                       medium
SASB Self-blame                .64                       medium
SASB Self-attack                .96                       large
SASB Self-ignore                .65                       medium

Completers showed significant decreases in psychiatric symptoms and interpersonal problems on 6 of 10 subscales and GSI total scale of the SCL-90-R, and 6 of 8 subscales and on the Total scale of IIP-C. Patients showed also significant decreases in levels of 3 SASB subscales reflecting self-critical self-concepts. However, they showed no significant changes in the levels of 5 other SASB subscales, nor on total scales. The Effect Sizes obtained ranged from medium to large.

Discussion
The participants in the PBSP groups showed the predicted decreases in psychiatric symptoms and interpersonal problems. These preliminary findings in general were of medium size and were at a level somewhat lower than in other studies of long term group therapy. Reductions in negative self-concepts were also found. The lack of increases in positive self-concepts is open to several explanations and invites further investigation. Caution needs to be employed in generalizing from this small and uncontrolled sample.

 The lack of data from non-participants and dropouts may have inflated the size of the positive changes obtained. However, the participants were not ending participation in a fixed length treatment group. This may have tended to lower the sizes of the changes observed. Patients with more symptoms would have an easier time justifying their continuation in the groups for another year. Patients in the treatment literature on the other hand were completing groups and could be expected to have a bias in the opposite direction.

 Participation in PBSP group therapy has demonstrated some effectiveness, but the lack of experimental controls makes it impossible to determine if these changes are caused by the participation in these groups (treatment efficacy). Further study of PBSP under more experimental control seems warranted.