Adaptive depends on context: An examination of the Intuitive Eating Scale-2 in an eating disorder sample

Abstract

Though there is a robust literature base linking intuitive eating (IE) to better psychological health and reduced eating pathology, little is known about the psychometric properties of IES-2 measurement or how IE relates to eating psychopathology within eating disorder clinical samples. Importantly, some seemingly adaptive eating behaviors in non-clinical populations could conceivably result from disordered eating within clinical eating disorder populations (e.g., not eating when not hungry because of distorted hunger-fullness cues, furthering restriction). This study examined the factor structure, psychometric properties, and correlates of the Intuitive Eating Scale-2 (IES-2) in a transdiagnostic eating disorder sample (N = 224). Participants at an eating disorder specialty treatment center completed the IES-2 upon assessment, along with other symptomology measures. A confirmatory factor analysis was performed, finding a poor model fit in this transdiagnostic sample, CFI = 0.889, TLI = 0.869, RMSEA = 0.095, TLI = 0.869, and SRMR = 0.099. Through exploratory factor analysis, we identified a factor solution for the measure that can be used transdiagnostically in this population, with revised scoring and modifications. As expected, the IES-2 demonstrated construct validity, with higher IE being associated with lower eating pathology (r = −0.36, p < .01), lower clinical impairment (r = −0.26, p < .01), lower body image dissatisfaction (r = −0.39, p < .05), and lower depressive symptoms (r = −0.20, p < .01), supporting construct validity. As anticipated, IES-2 was not related to emotion regulation, supporting discriminant validity. This study suggests that the IES-2 does not perform as expected in a transdiagnostic clinical eating disorder population, and instead, a revised scored 21-item IES-2 is suggested, with interpretation of only the total score and two subscales.