Therapist strategies early in therapy associated with good or poor outcomes among clients with low proactive agency.

OBJECTIVE: Within a mixed methods program of research the present study aimed at expanding knowledge about interactions in the initial therapeutic collaboration by combining focus on client interpersonal style and therapist contribution. METHOD: The study involves in-depth analyses of therapist-client interactions in the initial two sessions of good and poor outcome therapies. Based on interpersonal theory and previous research, the Inventory of Interpersonal Problems (IIP-64-C) was used to define poor outcome cases, that is, low proactive agency cases. To compare good and poor outcome cases matched on this interpersonal pattern, cases were drawn from two different samples; nine poor outcome cases from a large multi-site outpatient clinic study and nine good outcome cases from a process-outcome study of highly experienced therapists. RESULTS: Qualitative analysis of therapist behaviors resulted in 2 main categories, fostering client's proactive agentic involvement in change work and discouraging client's proactive agentic involvement in change work, 8 categories and 22 sub-categories. CONCLUSION: The findings revealed distinct and cohesive differences in therapist behaviors between the two outcome groups, and point to the particular therapist role of fostering client agency through engagement in a shared work on change when clients display strong unassertiveness and low readiness for change. Clinical or Methodological Significance Summary: The present analysis combines focus on client interpersonal style, therapist strategies/process and outcome. The categories generated from the present grounded theory analysis may serve as a foundation for identifying interactions that are associated with agentic involvement in future process research and practice, and hence we have formulated principles/strategies that were identified by the analysis.

Treatment failure in psychotherapy: the pull of hostility.

Client-therapist interactions were studied in 14 positive-change (PC) and 14 negative-change or nonchange (NC) therapies with the same therapists and similar clients. Aggregated structural analysis of social behavior (SASB) scores showed increasingly dissimilar interaction styles between client and therapist in NC therapies. First-lag transition analyses of SASB codings of Sessions 3, 12, and 20 showed the following differences: Stable hostile complementarity characterized NC within and across sessions. Hostile complementarity was nevertheless relatively rare. Therapists met clients' invitations to hostile responses most frequently in nonhostile ways, yet they initiated more belittling and ignoring interactions with NC clients, pointing to the subtly hostile therapeutic climate created. Rejection of therapists' interventions predicted negative outcome most strongly and escalated with time. Clients' skepticism may make therapists vulnerable to feelings of inadequacy and, if not dealt with therapeutically, may easily release the therapists' own hostility.

Validation of the SASB Introject Surface in a Norwegian clinical and nonclinical sample.

In this study, we examined the reliability and construct validity of the Structural Analysis of Social Behavior Introject Surface, Intrex long form A (SASB-IS; Benjamin, 1995) in 2 Norwegian samples. The fit of the 8 SASB-IS scales to the structural requirements of a circumplex model with relaxed equal spacing constraints was reasonably good in an outpatient sample, but poor in a normal reference sample. The deviations from the equal spacing based on an ideal circumplex model, however, seem to have minimal implications for the utility of the instrument in clinical assessment. The reliability of the SASB-IS was acceptable on most scales, but two scales had unacceptable low reliability. Correspondence with external criteria supported the validity of the SASB-IS in both samples. Profile patterns related to different segments of the introject circumplex model were systematically related to severity of psychopathology: Hostile and accepting patterns of self-relatedness formed polar opposites; control patterns and intermediate patterns gave intermediate pathology scores.