Conceptual Invariance, Trajectories, and Outcome Associations of Working Alliance in Unguided and Guided Internet-Based Psychological Interventions: Secondary Analysis of a Randomized Controlled Trial

BACKGROUND: The role of working alliance remains unclear for many forms of internet-based interventions (IBIs), a set of effective psychotherapy alternatives that do not require synchronous interactions between patients and therapists. OBJECTIVE: This study examined the conceptual invariance, trajectories, and outcome associations of working alliance across an unguided IBI and guided IBIs that incorporated clinician support through asynchronous text messaging or video messaging. METHODS: Adults with high education attainment (n=145) with subclinical levels of anxiety, stress, or depressive symptoms were randomized to 1 of 3 treatment conditions for 7 weeks. All participants received treatments from MyCompass, an unguided IBI using cognitive behavior therapy. Participants in condition 2 and 3 received supplemental, asynchronous clinician support through text and video, respectively. Working alliance with the IBIs was measured weekly using select items from the 12-item version of the Agnew Relationship Measure. Symptom and functional outcomes were assessed at baseline, at the end of treatment, and 1-month follow-up. RESULTS: Working alliance with the IBIs was conceptually invariant across the 3 conditions. Working alliance followed a quadratic pattern of change over time for all conditions and declined significantly only in the text-support condition. After controlling for baseline symptoms, higher baseline levels of working alliance predicted less depression and less functional impairment at follow-up, whereas faster increases in working alliance predicted less worry at the end of treatment and at follow-up, all of which only occurred in the video-support condition. CONCLUSIONS: Working alliance with the IBIs was generally established in the initial sessions. Although working alliance is conceptually invariant across IBIs with or without clinician support, the associations between working alliance and treatment outcomes among IBIs may differ depending on clinician involvement and the modalities of support. TRIAL REGISTRATION: ClinicalTrials.gov NCT05122429; https://clinicaltrials.gov/ct2/show/NCT05122429