The Role of Parent Therapeutic Alliance in Cognitive Behaviour Therapy for Children with Autism

Background: Therapeutic alliance (TA) (i.e., the collaborative working relationship between a therapist and client) is an important component of treatment success across a range of treatment modalities among both child and adult populations. In empirically supported interventions designed to address emotional and behavioural challenges for children with ASD, TA is currently an under-studied treatment variable, especially parent-therapist alliance.

Objectives: The objectives of this study were to examine: (1) the psychometric properties of a behavioural measure of alliance, the Therapy Process Observational Coding System–Alliance Scale (TPOCS-A; McLeod, 2001), and (2) how parent-therapist and child-therapist alliance related to post-treatment outcomes in a sample of children with ASD who participated in a 10-session cognitive behavior therapy program targeting emotion regulation difficulties.

Methods: Twenty children (19 males) ages 8 to 12 years (M=9.8; SD=1.29) with ASD and their parents (15 mothers, 5 fathers) completed the Secret Agent Society: Operation Regulation; Beaumont, 2013), as part of a randomized waitlist controlled trial. Parents and children completed measures of child emotion regulation (Emotion Regulation Checklist,Shields & Cicchetti, 1997; Children’s Emotion Management Scale, Zeman et al., 2010) and psychopathology (Behavior Assessment System for Children, 2^ndEdition, Reynolds & Kamphaus, 2004) pre- and post-intervention. Trained coders rated early (Session 2), middle (Session 5), and late (Session 9) therapy sessions using the TPOCS-A. Interrater reliability was “excellent” (ICC=.95). Therapist-reported client involvement, alliance, and adherence were also assessed.

Results: Behavioural ratings of parent and child alliance were correlated (r=.48, p=.04) across sessions. Internal consistencies were “respectable” to “very good”. Parent behavioural ratings were correlated with therapist-rated parent alliance (r=.63,p=.003) and a similar pattern was evident for child behavioural ratings and therapist-rated child alliance (r=.58, p=.008). Child behavioural ratings were related to therapist-reported treatment adherence (homework completion) (r=.69, p<.001) and therapist-reported client involvement (r=.55, p=.01). Regarding associations with child outcomes, higher levels of parent-therapist alliance early in treatment were related to improvements in child emotional inhibition [F(1, 17)=5.41, p=.03]; higher child-therapist alliance early in treatment was related to improvements in dysregulated expression [F(1, 16)=10.48, p=.005)]and improvements in internalizing symptoms [F(1, 17)=4.36, p=.05] post-intervention. Higher levels of child-therapist alliance late in treatment were related to improved emotion coping skills [F(1, 15) = 5.35, p= .04], improvements in dysregulation [F(1, 15)=4.52, p=.05], and less internalizing problems [F(1, 16)=6.59, p=.02]. Higher parent-therapist alliance at the end of therapy was related to improvements in child internalizing problems [F(1, 15)=5.36, p=.04] and adaptive regulation [F(1, 15)=9.72, p=.007].

Conclusions: Our findings provide support for the psychometric properties of a behavioural measure of TA and reveal positive associations between TA and improvements in child psychopathology and emotion regulation among a sample of children with ASD following intervention. Research and clinical implications of examining relational processes on child outcomes in empirically supported treatments for children with ASD and their families will be discussed.