Testing the Factor Structure and Psychometric Properties of the Toronto Alexithymia Scale-20 Among a Sample of Individuals Diagnosed with Eating Disorders

Alexithymia is implicated in eating disorder phenomenology. The Toronto Alexithymia Scale-20 (TAS-20) is a commonly used measure of alexithymia, which has been used to investigate associations between alexithymia and eating disorders. However, the TAS-20 has never been validated among individuals across the eating disorder spectrum. The purpose of this study was to test the factor structure and psychometric properties of the TAS-20 in a sample of patients with eating disorders. Participants were 710 patients admitted to a partial hospitalization program for eating disorder treatment. Participants completed self-report instruments at admit and one-month into treatment. Analyses tested the three-factor structure of the TAS-20 using confirmatory factor analyses (CFA) and measurement invariance across eating disorder diagnosis. Next, analyses tested the psychometric properties (i.e., convergent validity, internal consistency and test-retest reliability) of this model. CFA determined that the original three-factor structure of the TAS-20 fit the data after removal of five items in the Externally Oriented Thinking factor. This structure was invariant across diagnoses. The TAS-20 showed strong test-retest reliability, good internal consistency (excluding Externally Oriented Thinking), and moderate convergent validity. This structure of the TAS-20 may be more appropriate to use in future clinical work and research among those with eating disorders.