Date of Submission


Degree Type


Degree Name

Doctor of Psychology (PsyD)



Department Chair

Robert A DiTomasso, PhD, ABPP, Chair, Department of Psychology

First Advisor

Petra Kottsieper, PhD, Chairperson

Second Advisor

Marie Kearns, PhD

Third Advisor

Robert A DiTomasso, PhD, ABPP


Adherence to and engagement with treatment is a complex and multifaceted issue encountered throughout many medical and helping professions, particularly in the fields of psychology and psychiatry. Adherence to treatment/service engagement is especially an issue in the treatment of serious mental illnesses; studies have shown that up to 80% of individuals do not adhere to their treatment (Tait et al., 2003). This study investigated if cognitive insight, as measured by the Beck Cognitive Insight Scale ([BCIS] Beck et al., 2004), and clinical insight are related to the recovery concepts of hope and self-stigma as well as symptomatology in predicting treatment adherence/service engagement in a sample of individuals diagnosed with severe mental illness. The sample consisted of volunteers under treatment in outpatient facilities utilizing the Assertive Community Treatment model. Data was collected from 35 participants from two different ACT team treatment sites. Initial T-tests showed that there was a significant difference between the sex of participants t(33) = 2.276, p = .029 and treatment site of participants t(33) = -2.453, p = .020 in regard to scores on the dependent variable measure, a self-designed service engagement scale. A hierarchical multiple regression analysis was conducted in order to test whether the independent variables of cognitive insight as measured by the Self-Reflectiveness and Self- Certainty subscales of the Beck Cognitive Insight Scale, i.e., clinical insight, hope, four factors of self-stigma (aware, agree, apply, and hurts self), and symptomatology, could significantly predict the dependent variable of service engagement/adherence. Model 1 of the regression (site and sex) was found to be significant (p = .011), but model 2 (predictor variables) was found not to be significant (p = .122). Scores from the Beck Cognitive Insight Scale Self-Reflectiveness subscale were positively correlated at the .01 level of significance with scores on the Insight Scale (r = .544), providing further evidence of the construct validity of the BCISR. No significant relationship was found between the Beck Cognitive Insight Scale Self-Certainty subscale and the Birchwood Insight Scale as predicted. Several correlations were found between variables. Small sample size and lack of an accurate measure to gauge service engagement were identified as the greatest limitations of the study. Areas of future research on the constructs of insight, hope, self-stigma, and treatment adherence/service engagement are discussed.