Client-therapist temporal congruence in perceiving immediate therapist self-disclosure and its association with treatment outcome

Abstract

Objective: Immediate therapist self-disclosure (Im-TSD) can be a powerful intervention. When engaged in judiciously, it can provide clients with a unique opportunity to explore their interpersonal relationship in real time. Relational theories suggest that for Im-TSD to be effective, both client and therapist must have temporally congruent perceptions of its occurrence. The present study examined (a) whether clients and therapists are temporally congruent in their session-by-session ratings of Im-TSD; and (b) whether this congruence is associated with therapy outcomes. Method: After each session, clients (n = 102) and therapists (n = 60) at a university-based clinic indicated whether Im-TSD was present during the session. Before each session, clients self-reported their functioning. They rated session quality after each session. Results: Therapists’ ratings of their Im-TSD tended to be temporally congruent with their clients’ Im-TSD ratings. Greater temporal congruvdence was associated with greater improvement over time in clients’ experience of the session as helpful, but not with changes in clients’ functioning. Conclusion: The findings highlight the importance of establishing a stronger temporal congruence of Im-TSD ratings between therapists and clients to further improve clients’ experiences in treatment. The findings’ implications are discussed as well as situations in which temporal congruence may not be beneficial.

Keywords:

• Therapist self-disclosure
• Congruence
• Truth and Bias Model
• Process-outcome research

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

1 The following DSM-IV diagnoses were assessed in the affective disorders cluster: major depressive disorder (17.65%), dysthymia (13.51%) and bipolar disorder (1.20%). The following DSM-IV diagnoses were subsumed under the anxiety disorders cluster: generalized anxiety disorder (16.67%), social anxiety disorder (11.76%), panic disorder (3.90%) and agoraphobia (1.98%).

2 One coefficient represents the expected between-person reliability on one fixed session; it can be thought of as a proxy for the average session-specific alpha-coefficients across all sessions. The other coefficient assesses the precision of the measurement of systematic change of therapists over sessions. It represents the within-person reliability of session-to- session change on the scale.

3 When attempting to estimate three-level models (i.e., taking into account therapist effects), the models did not converge. This is likely the result of the low average number of clients treated by the same therapist in the sample (in which most therapists treated only one client); this limits the extent to which therapist effects could be examined. All the analyses were re-run with one client per therapist (the one who provided the most session-by-session reports; Nclients = 60). The pattern of results was almost identical to the one obtained for the entire sample of clients, suggesting that consistent with recent simulations (Schiefele et al., 2017), therapists' effects, if they existed, were unlikely to have biased the within-dyad effects. These results are available upon request.

4 In order to determine whether the findings could be influenced by a general increase or decrease in Im-TSD over time, we fitted a Truth-and-Bias model which included both time, as a separate parameter in the model, and the interaction between time and truth. There was no interaction between temporal congruence and time, and hence no changes in congruence over time, implying that the findings cannot be ascribed to linear trends in the therapists’ and clients' ratings. We found that the directional bias decreased over time, although we had formulated no a-priori question related to this issue.