Fat Phobia In Eating Disorders: Explicit Attitudes And Implicit Associations

Objective: To understand the patterns and mechanisms of fat phobia in different types of eating disorders(EDs),and to identify potential markers for ED diagnosis and assessments on both the explicit and implicit levels of cognition.Methods: Subjects: The study included 57 female first-onset,currently ill ED outpatients or inpatients(aged 13-30 years)and 31 healthy volunteers matched for gender,age and education.Measures: Measures of ED specific and general psychopathology were administered,including the Eating Disorder Examination Questionnaire(EDE-Q),Eating Disorder Inventory(EDI),Beck Depression Inventory,Beck Anxiety Inventory and Perceived Stress Scale.Fat phobia was assessed on the explicit level using 4 different measures(EDE-Q item fear of weight gain,EDE-Q Shape Concern and Weight Concern subscales,EDI Drive for Thinness subscale).An Implicit Association Test(IAT)was used to measure the implicit fat-bad / thin-good associations.Statistical analysis: We use "desired body weight is lower than actual body weight and lower than 85% of the lowest healthy weight for the age" as the operational criterion for the presence of fat phobia.Pearson's Chi-squared test was used to analyze the correlation between the presence of fat phobia and ED diagnosis.Subjects were then divided into different groups according to ED diagnosis and the presence of fat phobia.Welch's ANOVA was used to examine the differences in fat phobia(as measured by different means)and other clinical psychological features among different groups of subjects.Pearson's correlation coefficient was used to identify factors related to the implicit fat-bad / thin-good associations.Results:(1)Using our operational criterion for the presence of fat phobia,26.3%(95% CI: 17.0%-40.6%)among all ED patients(N = 57)and 48.4%(95% CI: 33.6%-69.6%)among patients with a diagnosis of anorexia nervosa(AN,n =31)were classified as non-fat phobic,whereas all of the patients with a diagnosis of bulimia nervosa(BN,n = 26)were classified as fat phobic.(2)Group comparisons among non-fat phobic AN(NFP-AN)patients(n = 15),fat phobic AN(FP-AN)patients(n = 16),BN patients(n = 26)and healthy controls(HC,n = 31)showed that NFP-AN patients differed from FP-AN patients in various clinical(including EDE-Q global and restraint subscale score,level of depression,anxiety and stress as measured by BDI,BAI and PSS)and psychological(including interoceptive awareness,ascetism,and impulse regulation as measured by EDI)variables.(3)In each of the 4 explicit measures of fat phobia,FPAN and BN patients demonstrated stronger fat-phobic attitudes than HCs,whereas NFP-AN patients did not differ from HCs.(4)The IAT showed stronger fat-bad / thin-good associations in FP-AN and BN patients compared to HCs,while NFP-AN did not differ from HCs.Implicit fat-bad / thin-good associations were positively correlated with EDE-Q Shape Concern,Weight Concern and EDI Drive for Thinness subscale scores.Compared with explicit measures,the IAT showed greater group difference between the FP-AN and NFP-AN groups(Cohen's d = 1.22).Conclusions: This study indicates that fat phobia in AN and BN may be related to implicit fat-bad / thin-good associations.Beside the absence of explicit fat phobic attitudes,NFP-AN patients also showed weak fat-bad / thin-good associations that are comparable to those of healthy controls.They also differed from FP-AN patients in various clinical and psychological variables.These results suggest that NFP-AN may result from mechanisms other than fat phobia and cannot be fully explained by denial of illness or deception.Taken together,the study provides further insights into the psychopathology of ED at the level of implicit cognition.It also suggests that IAT may help differentiate fat phobic and non-fat phobic EDs and may therefore be potentially useful in the diagnosis and assessments of EDs.