Patient alliance with the program predicts treatment outcomes whereas alliance with the therapist predicts adherence in internet-based therapy for panic disorder.

This study examines relationships among different aspects of therapeutic alliance with treatment outcome, adherence and attrition in internet delivered cognitive behavioral therapy (ICBT) for panic disorder. We examined alliance-outcome relationships in ICBT ( = 74) using a newly developed self-report alliance measure that disentangles alliance with program content (Internet Patient's Experience of Attunement and Responsiveness with the program; I-PEARp) and with the therapist (I-PEARt). We compared ICBT outcomes of patient rated and therapist-rated alliance with conventional alliance scales (WAI-6 and WAI-T). Consistent with our hypothesis, I-PEARp and I-PEARt distinguished between different aspects of the alliance and predicted outcomes better than standard alliance scales. Furthermore, higher ratings of I-PEARp were associated with subsequent lower symptoms and lower symptoms were associated with higher subsequent alliance. In contrast, I-PEARt predicted adherence, but not symptoms. Although therapists' ratings of alliance (thI-PEAR) improved significantly during treatment, they did not predict subsequent symptoms, adherence, or dropout. Results indicate that the patient experience of the alliance in ICBT includes two aspects, each of which uniquely contributes to outcomes; patient connection to the program is related to symptom outcomes whereas the dyadic relationship with the therapist serves as the glue to allow the treatment to hold.