Send to

Choose Destination
J Med Internet Res. 2012 Jun 28;14(3):e88. doi: 10.2196/jmir.1848.

Working alliance in online cognitive behavior therapy for anxiety disorders in youth: comparison with clinic delivery and its role in predicting outcome.

Author information

School of Psychology, University of Queensland, Brisbane, QLD, Australia.



Substantial evidence exists that positive therapy outcomes are related to the therapist-client working alliance.


To report two studies that examined (1) the quality of the working alliance in online cognitive behavior therapy (CBT), with minimal therapist contact, for anxiety disorders in youth, and (2) the role of working alliance and compliance in predicting treatment outcome.


Study 1 participants were 73 adolescents aged 12 to 18 years who met diagnostic criteria for an anxiety disorder, plus one or more of their parents. Participants were randomly assigned to clinic or online delivery of CBT, with working alliance being assessed for youth and parents after session 3. Study 2 participants were 132 children and adolescents aged 7 to 18 years who met diagnostic criteria for an anxiety disorder, plus one or more of their parents. Youths and parents participated in a minimally therapist-assisted online CBT program supported by brief, weekly emails and a single, short phone call.


Study 1 revealed a strong working alliance for both online and clinic CBT, with no significant difference in working alliance between conditions for adolescents (F(1,73 )= 0.44, P = .51, η(p) (2 )= 0.006, Cohen d = 0.15). Parents also reported high working alliance in both conditions, although a slight but significantly higher working alliance in clinic-based therapy (F(1,70 )= 6.76, P = .01, η(p) (2 )= 0.09, Cohen d = 0.64). Study 2 showed a significant and substantial decrease in anxiety symptoms following online therapy (P < .001 for all outcome measures). Adolescents improved significantly more in overall functioning when working alliance (beta = .22, t(79 )= 2.21, P = .03) and therapy compliance (beta = .22, t(84 )= 2.22, P = .03) were higher, with working alliance also predicting compliance (beta = .38, F(1,80 )= 13.10, P = .01). No such relationships were evident among younger children.


Working alliance is important in determining clinical outcome for online treatment for anxiety among adolescents, with minimal therapist assistance, although this was not the case for younger children.


Australian New Zealand Clinical Trials Registry: ACTRN12611000900910;

[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for JMIR Publications Icon for PubMed Central
Loading ...
Support Center