The Theory and Practice of Group Psychotherapy (51 page)

SUMMARY

Selection of clients for group therapy is, in practice, a process of deselection: group therapists exclude certain clients from consideration and accept all others. Although empirical outcome studies and clinical observation have generated
few inclusion criteria
, the study of failures in group therapy, especially of clients who drop out early in the course of the group, provides important
exclusion
criteria.

Clients should not be placed in a group if they are likely to become groups deviants. Deviants stand little chance of benefiting from the group experience and a fair chance of being harmed by it. A group deviant is one who is unable to participate in the group task. Thus, in a heterogeneous, interactional group, a deviant is one who cannot or will not examine himself and his relationship with others, especially with the other members of the group. Nor can he accept his responsibility for his life difficulties. Low psychological-mindedness is a key criterion for exclusion from a dynamic therapy group.

Clients should be excluded from long-term groups if they are in the midst of a life crisis that can be more efficiently addressed in brief, problem-specific groups or in other therapy formats.

Conflicts in the sphere of intimacy represent both indication and contraindication for group therapy. Group therapy can offer considerable help in this domain—yet if the conflicts are too extreme, the client will choose to leave (or be extruded) by the group. The therapist’s task is to select those clients who are as close as possible to the border between need and impossibility. If no markers for exclusion are present, the vast majority of clients seeking therapy can be treated in group therapy.

Chapter 9

THE COMPOSITION OF THERAPY GROUPS

A
chapter on group composition might at first glance seem anachronistic in the contemporary practice of group psychotherapy. Economic and managed care pressures on today’s group therapist may make the idea of mindfully composing a psychotherapy group seem an impractical luxury. How can one think about the ideal method of composing therapy groups when pressures for target symptom relief, homogeneous groups, structured meetings, and brevity of therapy are the order of the day? Moreover, empirical research indicates that
the briefer and more structured the group, the less important are compositional issues
.
¹
To make matters worse, research in group composition is doubtless one of the most complex and confusing areas in the group therapy literature. So what is the point of including a chapter on group composition in this text?

In this chapter my aim is to show that the principles of group composition are relevant
in all forms of therapy groups, even the most structured and seemingly homogeneous.
Group composition principles help group leaders understand the process within each group and tailor their work to meet the requirements of each client. If therapists fail to attend to issues of diversity in interpersonal, cognitive, personality, and cultural dimensions, they will fall prey to a simplistic and ineffective “one-size-fits-all” approach to group therapy. The research on group composition is voluminous and complex. Readers who are less interested in research detail may prefer in this chapter to focus on the section summaries and the final overview.

Let us begin with a thought experiment. Imagine the following situation: An ambulatory mental health clinic or counseling center with ten group therapists ready to form groups and seventy clients who, on the basis of the selection criteria outlined thus far, are suitable group therapy candidates.
Is there an ideal way to compose these ten groups?

Or imagine this more common, analogous situation: An intake coordinator deems a client a suitable candidate for group therapy, and there are several groups operating in the clinic, each with one vacancy. Into which group should the client go? Which group would offer the best
fit?
† Both situations raise a similar question:
Is there a superior method of composing or blending a group
? Will the proper blend of individuals form an ideal group? Will the wrong blend remain inharmonious and never coalesce into a working group?

I believe that it is important to establish valid compositional principles to help us determine which clients should go into which groups. We grope in the dark if we try to build a group or fill a vacancy without any knowledge of the organization of the total system. The stakes are high: first, a number of comembers will be affected by the decision to introduce a particular client into a group, and second, the brief frame of contemporary group treatment leaves little time for correction of errors.

As in preceding chapters, I will devote particular attention to groups with ambitious goals that focus on here-and-now member interaction. But principles of composition also apply to homogeneous, problem-specific, cognitive-behavioral, or psychoeducational groups. Keep in mind that even in such groups, homogeneity in one dimension, such as diagnosis, can initially mask important heterogenity (for example, stage and severity of illness) that may powerfully interfere with the group’s ability to work well together.

First, let me clarify what I mean by right and wrong “blends.” Blends of what? What are the ingredients of our blend? Which of the infinite number of human characteristics are germane to the composition of an interactional therapy group? Since each member must continually communicate and interact with the other members, it is the interaction of members that will dictate the fate of a group. Therefore, if we are to deal intelligently with group composition, we must aim for a mix that will allow the members to interact in some desired manner. The entire procedure of group composition and selection of group members is thus based on the important assumption that we can, with some degree of accuracy,
predict the interpersonal or group behavior of an individual from pretherapy screening
. Are we able to make that prediction?

THE PREDICTION OF GROUP BEHAVIOR

In the previous chapter, I advised against including individuals whose group behavior would render their own therapy unproductive and impede the therapy of the rest of the group. Generally, predictions of the group behavior of individuals with extreme, fixed, maladaptive interpersonal behavior (for example, the sociopathic or the floridly manic client) are reasonably accurate:
in general, the grosser the pathology, the greater the predictive accuracy
.

In everyday clinical practice, however, the problem is far more subtle. Most clients who apply for treatment have a wider repertoire of behavior, and their ultimate group behavior is far less predictable. Let us examine the most common procedures used to predict behavior in the group.

The Standard Diagnostic Interview

The most common method of screening clients for groups is the standard individual interview. The interviewer, on the basis of data on environmental stresses, personal history, and inferences about motivation for treatment and ego strength, attempts to predict how the individual will behave in the group. These predictions, based on observations of a client’s behavior in the dyadic situation, are often hazy and inaccurate. Later in the chapter I will present some strategies to increase the validity of these preliminary inferences.

One of the traditional end products of the mental health interview is a diagnosis that, in capsule form, is meant to summarize the client’s condition and convey useful information from practitioner to practitioner. But does it succeed in offering practical information? Group therapists will attest it does not! Psychiatric diagnoses based on standard classificatory systems (for example, DSM-IV-TR) are, at best, of limited value as an indicator of interpersonal behavior. Diagnostic nomenclature was never meant for this purpose; it stemmed from a disease-oriented medical discipline. It is based primarily on the determination of syndromes according to aggregates of certain signs and symptoms. Personality is generally classified in a similar fashion, emphasizing discrete
categories
of interpersonal behavior rather than describing interpersonal behavior as it is actually manifested.
²

The 2000
Diagnostic and Statistical Manual of Mental Disorders
(DSM-IV-TR) is an improvement over earlier psychiatric diagnostic systems, and it pays far more attention to personality. It codes personality on a specific axis (Axis II) and recognizes that an individual may demonstrate clustering of personality pathology in more than one area, apart from (or in addition to) Axis I psychiatric disorders. The DSM-IV-TR provides a sharper demarcation between severe and less severe personality disorders and in general has a more empirical foundation than previous DSM systems.
³

Nonetheless, the DSM-IV-TR, along with the most recent
International Classification of Disease
(ICD-10), has marked limitations for practitioners working with clients whose interpersonal distress and disturbance do not fit neatly into syndrome definitions. Contemporary diagnosis also emphasizes discrete and observable behavior, with little attention paid to the inner life of the individual.
⁴

Overall,
the standard intake interview has been shown to have little value in predicting subsequent group behavior
.
⁵
For example, one study of thirty clients referred to group therapy demonstrated that the intake interviewers’ ratings of five important factors—motivation for group therapy, verbal skills, chronicity of problems, history of object relations, and capacity for insight—had no predictive value for the client’s subsequent group behavior (for example, verbal activity and responsivity to other members and to the leader).
⁶

That a diagnostic label fails to predict much about human behavior should neither surprise nor chagrin us. No label or phrase can adequately encompass an individual’s essence or entire range of behavior.
⁷
Any limiting categorization is not only erroneous but offensive, and stands in opposition to the basic human foundations of the therapeutic relationship. In my opinion, the less we think (during the process of psychotherapy) in terms of diagnostic labels, the better. (Albert Camus once described hell as a place where one’s identity was eternally fixed and displayed on personal signs: Adulterous Humanist, Christian Landowner, Jittery Philosopher, Charming Janus, and so on.
⁸
To Camus, hell is where one has no way of explaining oneself, where one is fixed, classified—once and for all time.)

Standard Psychological Testing

The standard psychological diagnostic tests—among them the Rorschach test, the Minnesota Multiphasic Personality Inventory (MMPI), the Thematic Apperception Test (TAT), the Sentence Completion test, and the Draw-a-Person test—have failed to yield predictions of value to the group therapist.
⁹

Specialized Diagnostic Procedures

The limited value of standard diagnostic procedures suggests that we need to develop new methods of assessing interpersonal behavior. Slowly, the field is beginning to assess personality traits and tendencies more accurately to improve our methods of matching clients to therapy.
¹⁰
Recent clinical observations and research suggest several promising directions in two general categories:

1. A formulation of an interpersonal nosological system. If the critical variable in group therapy selection is interpersonal in nature, why not develop an interpersonally based diagnostic scheme?

2. New diagnostic procedures that directly sample group-relevant behavior

An Interpersonal Nosological System.
The first known attempt to classify mental illness dates back to 1700 B.C.,
¹¹
and the intervening centuries have seen a bewildering number of systems advanced, each beset with its own internal inconsistency. The majority of systems have classified mental illness according to either symptoms or presumed etiology. The advent of the object-relations and interpersonal systems of conceptualizing psychopathology, together with the increase in the number of people seeking treatment for less severe problems in living,
¹²
stimulated more sophisticated attempts to classify individuals according to interpersonal styles of relating.† In previous generations psychotherapy researchers interested in the impact of personality variables on the individual’s participation in groups measured such variables as externalization and resistance,
¹³
perceived mastery and learned resourcefulness,
¹⁴
dogmatism,
¹⁵
preference for high or low structure,
¹⁶
social avoidance,
¹⁷
locus of control,
¹⁸
interpersonal trust,
¹⁹
and social risk-taking propensity.
²⁰

It is of interest to note that some of the contemporary empirical schema of interpersonal relationships draw heavily from earlier clinical conceptualizations. Karen Horney’s midcentury model has been particularly relevant in new formulations. Horney viewed troubled individuals as moving exaggeratedly and maladaptively
toward, against
, or
away from
other people and described interpersonal profiles of these types and various subtypes.
²¹