Therapists' in-session feelings in psychotherapy can be seen as indications of the development of the therapeutic relationship and the therapeutic process. To manage them appropriately, it is important to know to what extent they may be influenced by patients' pretreatment characteristics. This study aims to improve the understanding of therapists' emotional reactions in the psychotherapeutic setting by investigating if patients' pretreatment mentalization ability and attachment style predicted therapist in-session feelings. In a sample of 87 therapy dyads treated with interpersonal psychotherapy and cognitive behavioral therapy for depression, patient attachment was measured using self-reported Experiences in Close Relationships (ECR) and mentalization using Reflective Functioning (RF). ECR and RF were hypothesized to predict therapist feelings measured by the Feeling Word Checklist-24 at different treatment phases over the full course of treatment. Treatment method, patient age, gender, and pretreatment depression were evaluated as potential confounders. Multilevel modeling was used to analyze the data. Lower RF in patients predicted more negative therapist feelings in the mid- to late-treatment phases and less positive feelings in the late-treatment phase. Self-reported attachment anxiety or avoidance did not predict therapist feelings. Findings indicate that patients' ability to mentalize is important to consider when conducting psychotherapy, as it can influence therapists' feelings in the therapeutic process. Limitations of the present study's approach are discussed, and directions for future research are considered.

Objective: The objective of this study was to analyze the effect of adherence to both specific technique factors and facilitative condition variables (e.g., therapists involvement, understanding and support) in Cognitive Behavior Therapy (CBT) and Interpersonal Psychotherapy (IPT). In addition, we were interested in whether the effect of therapist adherence would depend on the level of the working alliance. Method: Three sessions each from 74 patients diagnosed with Major Depressive Disorder who were randomized to 14 sessions of IPT or CBT were rated for adherence using a modified version of The Collaborative Study Psychotherapy Rating Scale-6 (CSPRS-6). Data was analyzed using Multilevel Modeling. Results: No effects of adherence to specific factors on outcome were found in neither CBT nor IPT. Facilitative conditions were associated with better outcome in CBT but not in IPT, even after adjustment for the quality of the working alliance. No interaction effects were found. Conclusions: Our findings highlight the importance of relational factors in CBT, but do not support the need for specific adherence to any of the two treatments. Possible explanations of the findings and directions for future research are discussed.

Objective: Several studies have shown that the quality of the working alliance predicts symptomatic improvement sessionby-session, including in cognitive-behavioral therapy (CBT). We wanted to explore what characterizes CBT sessions with high and low alliances further using qualitative analysis. Method: Ten CBT-sessions were selected from eight patients therapies in a larger research project on psychotherapy for patients with major depression. Five sessions were chosen from high- and five from low-alliance sessions, based on therapist- and patient-reported Working Alliance Inventory scores. Transcripts of these sessions were analyzed using thematic analysis. Results: The analysis yielded four themes, each structured into two sub-themes: Therapist style, Person in focus, Content focus, and Therapeutic direction. In contrast to low-alliance sessions, high-alliance sessions were characterized by a more exploring as opposed to expert therapist style; a focus on the patients thoughts, feelings, and behavior, rather than a diffuse focus or a focus on other peoples actions/external events; and a sense of moving forward rather than stagnation. Conclusion: Our qualitative analysis showed theoretically and clinically meaningful processes in CBT sessions of high- vs low working alliance. This method is a useful complement to quantitative within-patient analyses, to expand on the meaning of quantitative findings.

Interpersonal Psychotherapy (IPT) and Cognitive Behavioral Therapy (CBT) are both evidence-based treatments for Major Depressive Disorder (MDD). Several head-to-head comparisons between these methods have been made, most of them in the US. There is a need for more trials in different treatment settings. This thesis is based on a randomized controlled trial of CBT and IPT for MDD in a community-based psychiatric outpatient clinic. In the trial, treatment outcome and mentalization change was compared between the methods. In addition, the significance of pre-treatment mentalization for subsequent alliance and outcome was analyzed. Ninety-six patients, about half of them with personality disorders, were randomized to 14 sessions of CBT or IPT. The hypothesis was that IPT would not be inferior to CBT which was confirmed. CBT had a significantly higher drop-out rate. Initial capacity for mentalization predicted alliance and outcome in both IPT and CBT. The level of mentalization was changed in IPT but not in CBT

Interpersonell Psykoterapi (IPT) och Kognitiv Beteendeterapi (KBT) är båda evidensbaserade psykoterapier för egentlig depression. Flera direkta jämförelser mellan dessa terapimetoder har gjorts, de flesta i USA. Det finns dock ett behov av fler studier i olika socio-kulturella kontexter och behandlingsmiljöer. Denna avhandling är baserad på en randomiserad kontrollerad studie av KBT och IPT för egentlig depression på en psykiatrisk öppenvårdsmottagning. Behandlingsutfall och förändring i mentaliseringsförmåga jämfördes mellan metoderna. Dessutom studerades betydelsen av mentaliseringsförmågan före behandlingen som prediktor för allians och utfall. Nittiosex patienter, varav ca hälften med personlighetsstörning, randomiserades till 14 sessioner med IPT eller KBT. Hypotesen var att IPT inte skulle ha sämre utfall än KBT, vilket bekräftades. KBT hade signifikant högre grad av terapiavbrott. Initial mentaliseringsförmåga predicerade allians och utfall, både i IPT och KBT. Mentaliseringsförmågan ökade i IPT men inte i KBT.

Aims: Although considerable attention has been paid to the concept of mentalization in psychotherapy, there is little research on mentalization as predictor of psychotherapy process and outcome. Using data from a randomized controlled trial of cognitive-behavioral therapy and interpersonal psychotherapy for depression, we studied mentalization in 85 outpatients with major depressive disorder (MDD) according to the Diagnostic and Statistical Manual of Mental Disorders. It was hypothesized that patients showing lower capacity for mentalization would experience poorer quality of alliance and worse outcome. Method: Depressive symptoms were measured each session using the Beck Depression Inventory-II. Mentalization was measured as reflective functioning (RF) on a slightly shortened version of the Adult Attachment Interview. A measure of depression-specific reflective functioning (DSRF), measuring mentalization about depressive symptoms, was also used. The Working Alliance Inventory-Short Form Revised was completed after each session by both therapist and patient. Longitudinal multilevel modeling was used to analyze data. Results: The patients had on average very low RF (M = 2.62, SD = 1.22). Lower pretreatment RF/DSRF predicted significantly lower therapist-rated working alliance during treatment. RF did not affect patient-rated alliance, but lower DSRF predicted lower patient-rated alliance across treatment. Patients with higher RF/DSRF had better outcomes on self-rated depression. Conclusions: The findings showed lower than normal capacity for mentalization in patients with MDD. Lower RF/DSRF predicted worse treatment outcome. More research is needed to understand how RF affects psychotherapy response and how RF is affected after recovery from depression.

Aims: Despite considerable clinical attention to the concept of mentalization in psychotherapy, research on mentalization and the psychotherapy process is virtually absent. Using data from two ongoing randomized controlled trials of psychotherapy for depression, we studied mentalization in 45 patients diagnosed with DSM-IV Major Depressive Disorder. The initial psychotherapy process was hypothesized to be experienced as more difficult with patients showing lower capacity for mentalization because of their difficulty understanding behavior in terms of mental states.

Methods: Mentalization was measured as Reflective Functioning (RF) on the Adult Attachment Interview, conducted before therapy start. Additionally, a measure of Depression-Specific Reflective Functioning (DSRF) measuring mentalization about depressive symptoms was tested. Psychotherapy process was measured by the Working Alliance Inventory – Short form (WAI-S) and the Feeling Checklist (FC), which were completed after each session by both therapist and patient.

Results: Pre-treatment RF on the AAI was on average low (M = 3.1), but variation in RF did not predict any aspect of the initial (first four sessions) psychotherapy process. Higher DSRF predicted better working alliance and more positive feelings as rated by the patient. RF on the AAI did not predict any process measure significantly.

Discussion: The low RF in depression replicates two previous studies, and may help explaining why maternal depression is a risk factor for infant developmental problems. If the results for DSRF are replicated, this measure might be used for identifying patients who are not easy candidates for psychotherapy.