Primary and Secondary Outcomes from a Double Randomized Clinical Preference Trial of two Panic-Focused Psychotherapies

Abstract: Approximately 2-3% of adults in Sweden will experience unexpected panic attacks that cause them to develop fears of future attacks and to alter their lives, defined as Panic Disorder with or without Agoraphobia (PD/A). The condition often onsets in late adolescence or early adulthood, with high levels of psychiatric comorbidity, social difficulties, diminished study and work ability, an elevated risk for substance use, health problems, and mortality. Cognitive behavioural therapy (CBT), including Panic Control Treatment (PCT), is the most evaluated and recommended treatment, however far from all PD/A patients complete or respond to CBT. A promising psychological treatment for PD/A is the brief Panic-Focused Psychodynamic Psychotherapy (PFPP). Three randomized controlled trials (RCTs) have found PFPP to achieve comparable outcomes to CBT for PD/A severity up to 12 months post-treatment. Further, well-designed RCTs comparing PFPP to CBT, involving longer follow-ups of psychiatric symptoms and broader indices of functioning are needed.Project POSE (Psychotherapy outcome and Self-selection Effects) was a doubly randomized clinical preference trial (DRCPT) designed to test whether patient preferences for either PCT or PFPP, delivered in routine care, influenced outcomes for the two treatments. Adults (n=221) with primary PD/A were randomly allocated to: 1) to choose either PCT or PFPP; 2) to be randomly allocated to PCT or PFPP; or 3) to a treatment Waitlist. The primary outcome measures were the Panic Disorder Severity Scale (PDSS), work status and sick leave, assessed together with secondary outcomes, at post-treatment, 6-, 12-, and 24-month follow-ups. Treatment was delivered by 45 therapists trained in either PFPP or PCT. This thesis uses data from Project POSE to address three aims: 1) the relative efficacy of PCT and PFPP at post-treatment and during a two-year follow-up; 2) to explore apparent differences in the trajectory of weekly, self-reported panic symptoms in PCT and PFPP during the treatment phase; and 3) to investigate the effects of PCT and PFPP on Work ability and its relations with improvements in PD/A. Study I was the trial protocol and presented the theoretical and empirical justifications for a DRCPT of two psychotherapies for PD/A. Study II presents the primary and secondary outcomes for Project POSE at post-treatment and all follow-ups. Study III presents the results of an exploratory investigation of a resurgence of self-reported panic symptoms in the termination phase of PFPP (Termination Setback – TS) and patient characteristics that might help to explain the TS. Study IV presents the findings for the self-report Work Ability Index (WAI) and its relation to symptom severity and occupational status at post-treatment and during follow-up. Irrespective of assignment to the Choice or Random conditions, both treatments yielded clinically significant improvements for the primary and secondary outcomes, comparable to those found in previous trials of the two therapies. PCT was significantly superior to PFPP at post-treatment, possibly owing to a TS that occurred for patients in the PFPP treatment. Individuals with less avoidant attachment and less severe interpersonal problems were more likely to experience a TS during PFPP. However, PFPP was significantly superior to PCT during follow-up, so that the two treatments were equally effective at the 24-month follow-up for both the primary and secondary outcomes. Both treatments were well tolerated with no differences in drop-out rates. Further gains during the follow-up phase did not appear to be associated with further treatment seeking. High rates of employment were present at every assessment point, but significant improvement were observed for Work Ability at post-treatment and follow-up, apparently mediated by a reduction in self-reported panic symptoms during the treatment phase. Thus, it appears that a clinically meaningful reduction in panic symptoms is associated with significant improvements in the patient’s beliefs about their capacity in meeting current and future work demands.