(PS7-22) Dimensions of Clinical Perfectionism in Anorexia Nervosa, Atypical Anorexia Nervosa, and Bulimia Nervosa

Clinical perfectionism, characterized by unrealistically high personal standards (PS) and excessive concern over mistakes (CM), is theorized to contribute to eating disorder (ED) development and maintenance. Prior studies show higher perfectionism in anorexia nervosa (AN) than bulimia nervosa (BN), but it is unclear how specific dimensions of clinical perfectionism may differ across ED diagnoses, including atypical AN (AAN; defined by AN symptoms in individuals not clinically underweight). Further, it may be valuable to compare diagnostic subtypes focused on restriction or binge/purge behaviors. Thus, this study examined dimensions of clinical perfectionism in AN, AAN, and BN, which may lead to a more fine-grained understanding of these diagnoses and guide intervention approaches. This study included 546 participants(M_age = 30.1, SD=9.9; n=510 women, n=9 men, n=16 nonbinary, n=1 transgender man, n=10 undeclared) diagnosed with AN (AN-restricting [AN-r] n=137, AN-binge/purge [AN-bp] n=107), AAN (AAN-r n=123, AAN-bp n=87), or BN (n=92). Participants completed a diagnostic screening and the Frost Multidimensional Perfectionism Scale (FMPS). We used analyses of variance (ANOVAs) to examine differences on the FMPS PS and CM subscales between AN, AAN, and BN and between AN/AAN-r, AN/AAN-bp, and BN. There were significant between-group differences on the PS subscale (F(2,543)=6.35, p=.002). Pairwise comparisons indicated that PS scores in AN were higher than in BN (t =3.27, p=.003). PS scores were also higher in AAN than BN (t =3.33, p = .003), but did not differ between AN and AAN. Between-group differences for CM were not significant (p=.063). AN/AAN-r, AN/AAN-bp, and BN differed on PS scores (F(2,543)=9.42, p< .001). Pairwise comparisons showed that PS scores were higher in AN/AAN-r than in AN/AAN-bp (t =-2.46, p = .038). PS scores were also higher in AN/AAN-r than BN (t =4.19, p< .001), but did not differ between AN/AAN-bp and BN (p=.076). CM did not differ between groups (p=.090). Findings suggest that PS as a dimension of clinical perfectionism may differ based on ED behaviors (i.e., restriction and binge/purge) that characterize disorders and subtypes, while CM is similar across different EDs. Our results indicate that excessively high PS may be more associated with AN/AAN compared to BN. This diagnostic difference may be further explained by our subtype analyses, which suggest that high PS are more strongly linked to restriction but may be less relevant for binging and/or purging. Clinically, addressing the tendency for excessively high standards may be beneficial in treating restrictive behaviors in AN/AAN, although more research is needed to better understand the association between PS and restriction. Our results may also suggest that interventions targeting heightened CM may have transdiagnostic utility. Future studies should examine the efficacy of interventions targeting different aspects of perfectionism for ED behaviors in AN, AAN, and BN.