Date on Master's Thesis/Doctoral Dissertation


Document Type

Doctoral Dissertation

Degree Name

Ph. D.


Counseling and Human Development

Degree Program

Counseling and Personnel Services, PhD

Committee Chair

Leach, Mark

Committee Co-Chair (if applicable)

Owen, Jesse

Committee Member

Owen, Jesse

Committee Member

Halford, Tyler

Committee Member

Mitchell, Amanda

Author's Keywords

Psychotherapy; process; outcome; alliance


Ruptures in the therapeutic alliance have shown to be a significant predictor of therapy outcomes, with non-repaired ruptures resulting in increased dropout rates of therapy, and repaired ruptures resulting in positive therapy outcomes. This study investigated the impact of alliance ruptures on outcomes of therapy among youth and adolescents, with a specific focus on military youth and adolescents. A sample of 5,640 military adolescents who were treated by 101 therapists were selected for analysis based on inclusion criteria of: a) being aged 13 to 19 years old; and b) attending more than one session of therapy. Each session, clients completed the Outcome Rating Scale to measure well-being, and the Session Rating Scale as a measure of the therapeutic alliance. Clients were coded as either a) No Rupture; b) Rupture Repair; or c) Rupture Non-Repair, based on presence or absence of therapy ruptures, and the final result of ruptures at the end of the therapy process. This study examined comparisons of alliance-outcome relationships by rupture condition, the effect of rupture conditions on premature termination from therapy, and moderation effects of stage-of-therapy ruptures, age, problem type, and therapist effects. Findings showed that there was significantly greater positive change in well-being amongst clients coded as Rupture Repair compared to No Rupture and Rupture Non-Repair. Clients coded as Rupture Non-Repair were more likely to prematurely terminate from therapy and attend a fewer number of sessions compared to Rupture Repair clients. Among Rupture Repair clients, repairs more likely to occur in early sessions of therapy compared to later sessions. Neither age nor problem type was found to act as a moderator on the effect of rupture condition on therapy outcomes. Finally, although there is evidence of between-therapist effects on therapy outcomes, aggregate therapist alliance scores across client caseloads failed to predict an effect on either the intercept and/or slope of the alliance effect on therapy outcomes. In conclusion, monitoring between-session therapy alliance and the presence of ruptures may influence therapy outcomes. Treatment implications and future directions were discussed.