Trust as a Predictor of Eating Disorder Severity and Therapeutic Alliance in an Adolescent Clinical Sample




Krol, Heather Paige Lefkof

Journal Title

Journal ISSN

Volume Title



Trust is recognized as an important construct across many disciplines. Despite this recognition, trust has proven difficult to define and measure, due to its abstract nature and variety of properties. Economic exchanges have become the standard measure of behavioral trust. A recent version, the "Trust Game," provides a unique opportunity to examine behavioral trust in an interactive, interpersonal situation. The Trust Game has been useful in demonstrating patterns of trust behavior in several psychiatric populations. Currently, the Trust Game has yet to be examined with adolescents suffering from eating disorders (EDs). These individuals are at high risk for deleterious outcomes, including an increased mortality risk, and thus early intervention and effective treatment are critical. Limited research exists on the role of trust in patients with EDs, and even less is known about the relationship of trust to factors important to treatment. The present study sought to address this gap in the literature by examining the relationship between behavioral trust and factors relevant to ED treatment, including attachment, self-reported trust, depression, ED severity, and the therapeutic alliance. This study also examined if behavioral trust predicts factors associated with treatment outcomes, including ED severity and therapeutic alliance, above and beyond depression. The sample consisted of 40 adolescents (ages 12-18) who were admitted to either inpatient or partial-hospitalization levels of care for an ED. Participants completed questionnaires and played the Trust Game with a computer simulated "healthy stranger." This pilot investigation found that behavioral trust was partially associated with attachment to parents, and only one aspect of the game related to self-reported trust. The results showed that higher overall game earnings explained unique variance in ED severity, after controlling for depression. Finally, this study found limited relationship between behavioral trust and the therapeutic alliance, although important aspects of the alliance were explored. The findings from this study increase our understanding of behavioral trust, as measured by the Trust Game, in adolescents with EDs and the relevance with factors important to the treatment of this population. Implications for clinicians, limitations of the methodology, and suggested areas for future research are discussed.

General Notes

Table of Contents


Related URI