The Eating Disorder Inventory-2 Perfectionism subscale (EDI-P) was originally construed as a unidimensional measure of perfectionism. However, research in non-clinical samples suggests that the EDI-P measures two dimensions of perfectionism: self-oriented and socially prescribed perfectionism. This study aimed to investigate the factor structure of the EDI-P in a transdiagnostic sample of females seeking treatment for an eating disorder, and to determine the unique association between EDI-P dimensions, weight and shape concern, and dietary restraint in anorexia nervosa (AN), bulimia nervosa (BN), and eating disorder not otherwise specified. Two hundred and ninety nine females seeking treatment for an eating disorder at an outpatient eating disorder service completed the Eating Disorder Examination and the EDI-P. Confirmatory factor analysis supported a two-factor model of the EDI-P comprising self-oriented and socially prescribed perfectionism. Self-oriented perfectionism, but not socially prescribed perfectionism, accounted for unique variance in weight and shape concern and dietary restraint in both AN and BN. Results highlight the potential importance of self-oriented perfectionism in eating disorders and support the argument that self-imposed standards are central to perfectionism in eating disorders.
Perfectionism has been implicated in the development and maintenance of all forms of eating disorders, including anorexia nervosa (AN), bulimia nervosa (BN), and eating disorders not otherwise specified (EDNOS) (Fairburn et al., 2003, Pratt et al., 2001, Schmidt and Treasure, 2006, Shafran et al., 2002 and Slade, 1982). The Eating Disorder Inventory-2 Perfectionism subscale (EDI-P; Garner, 1991) has been widely used in the study of perfectionism and eating disorders. The EDI-P was originally construed as a unidimensional measure of perfectionism. However, it has been observed that the EDI-P measures beliefs about perfectionism in the intrapersonal and interpersonal domains, which correspond to the “self-oriented” and “socially prescribed” perfectionism dimensions outlined in the Hewitt and Flett (1991) multidimensional model of perfectionism (Sherry, Hewitt, Besser, McGee, & Flett, 2004). Specifically, the EDI-P comprises three items that appear to assess self-oriented perfectionism (the belief that perfection is required in personal performance) and three items that appear to measure socially prescribed perfectionism (the belief that perfection in personal performance is expected by others). The EDI-P may therefore be composed of two correlated factors of self-oriented and socially prescribed perfectionism. Two studies have used confirmatory factor analysis (CFA) to evaluate the factor structure of the EDI-P in samples of university students (Joiner and Schmidt, 1995 and Sherry et al., 2004), and both studies supported the two factor model. However, it has yet to be tested if the EDI-P is best represented by a two factor structure in a clinical eating disorder sample.
Moreover, it is unclear how self-oriented and socially prescribed perfectionism dimensions are associated with eating disorders. A number of studies have compared self-oriented and socially prescribed perfectionism in eating disorder diagnostic groups and non-eating disorder control groups. Self-oriented perfectionism has been shown to be elevated in both AN and BN relative to psychiatric and healthy controls (Bastiani et al., 1995, Castro-Fornieles et al., 2007, Cockell et al., 2002 and Pratt et al., 2001). In contrast, socially prescribed perfectionism has been shown to be elevated in AN relative to controls in some studies (Bastiani et al., 1995 and Cockell et al., 2002), but not others (Castro et al., 2004 and Castro-Fornieles et al., 2007), and no difference in socially prescribed perfectionism has been observed between BN and controls (Castro-Fornieles et al., 2007 and Pratt et al., 2001). Overall, these results suggest that self-oriented perfectionism may be more strongly associated with AN and BN than socially prescribed perfectionism.
Only one study has investigated the relationship between EDI-P dimensions and eating disorder psychopathology in AN and BN. Watson, Raykos, Street, Fursland, and Nathan (2011) observed that EDI-P self-oriented perfectionism, but not socially prescribed perfectionism, was uniquely associated with dietary restraint, eating concern, weight concern, and shape concern in a sample of 201 females seeking treatment for an eating disorder (AN, BN, or EDNOS). However, the relationship between EDI-P dimensions, dietary restraint, and weight and shape concern has not been compared across eating disorder diagnoses.
The current study aimed to use CFA to investigate the factor structure of the EDI-P in a transdiagnostic eating disorder sample. It was expected that a two factor model of the EDI-P would be supported. As it has been proposed that perfectionism operates to maintain eating disorders by encouraging determined striving to achieve in the valued domain of eating, weight, and shape, and their control (Fairburn et al., 2003), it was hypothesised that perfectionism would be associated with weight and shape concerns and dietary restraint. Therefore, this study also aimed to identify the association between weight and shape concern, dietary restraint, and EDI-P dimensions in AN, BN and EDNOS. It was hypothesised that self-oriented perfectionism would be more strongly associated with weight and shape concern and dietary restraint than socially prescribed perfectionism in all diagnostic groups.
