Aphasia and stroke therapeutic alliance measure (A-STAM): Development and preliminary psychometric evaluation

Purpose: The therapeutic alliance, also known as the therapeutic relationship, may influence treatment process and outcome in aphasia rehabilitation; however, we currently lack a reliable tool to measure this relationship. This study aimed to develop a novel measure of the therapeutic alliance applicable to this population and provide preliminary evidence of the measure's psychometric properties. Method: Statements were generated from the: (1) therapeutic alliance literature, (2) qualitative interviews with stakeholders, and (3) Q methodological insights with people with aphasia (PWA) (n = 455). A representative sample of statements was identified from the data set (n = 57) and reviewed by expert panels (professionals and PWA), culminating in a 42-item clinician and patient version of the aphasia and stroke therapeutic alliance measure (A-STAM). Reliability and validity of both the clinician and patient versions of A-STAM were investigated with 34 Clinician-patient dyads engaging in therapy. Result: Internal consistency and test-retest reliability were excellent for both clinician (α = 92; ICC = 0.93) and patient versions of A-STAM (α = 0.92; ICC = 0.97). In both versions, scores correlated highly with psychotherapeutic measures of therapeutic alliance, indicative of good construct validity (r_s = 0.75; r_s = 0.77). Conclusion: The findings establish the preliminary reliability and validity of A-STAM and support further investigation into the measure's psychometric properties in larger samples.