Exercise and eating disorders: a 3-year inpatient hospital record analysis

The purpose of this study was to describe the exercise behavior across the DSM-IV eating disorder diagnosis (anorexia nervosa (AN), bulimia nervosa (BN), and eating disorder not otherwise specified (EDNOS)) and determine if differences exist among exercise category (high level and low level), type of exercise (aerobic, strength, and total), Eating Attitudes Test (EAT), Eating Disorders Inventory (EDI-2), and the length of inpatient hospitalization. This study was a retrospective analysis of 199 inpatient hospital records of female patients (age M=20.6, S.D.=7.03) in the following diagnostic categories: AN (n=115, 58%), BN (n=38, 19%), and EDNOS (n=46, 23%). Patients were also grouped by a median split into low- and high-level exercise categories based on their total amount of weekly exercise. No significant relationship was found between diagnosis and exercise category (high and low) or diagnosis and exercise type (total, aerobic, and strength exercise). This is valuable because exercise specialist may be a worthwhile addition to treatment teams working with all eating disorder diagnostic groups. The high-level exercisers showed significantly greater scores on the EAT [F(2,1,110)=5.117, P<.05] and weight preoccupation [F(1,1010)=4.861, P<.05] than the low-level exercise groups. There were significant relationships between total exercise and EAT (r=.354, P<.01), drive for thinness (r=.262, P<.01), and body dissatisfaction (r=.312, P<.001). Total aerobic exercise was significantly related to EAT (r=.352, P<.001), drive for thinness (r=.272, P<.01), and body dissatisfaction (r=.268, P<.01). Finally, total strength exercise was related to body dissatisfaction (r=.249, P<.05). A significant difference was found among exercise group and length of hospitalization [F(1,155)=14.384, P<.000]. The high-level exercisers were in treatment for 3 weeks longer than their low-level exercisers. A forced stepwise multiple linear regression was calculated to predict the length of hospitalization. At Step 1, diagnosis and Body Mass Index (BMI) on admission explained 27% of the variance in length of inpatient hospitalization [F(4,154)=18.08, P<.004, r(2)=.273]. When two exercise variables were included, aerobic activity explained an additional 5% of variance, whereas strength activity explained an additional 0.3% of the variance. Three-fourths of the predictors explained significant components of the total 32% variance. The present study demonstrates that patients with greater total exercise and a greater total aerobic exercise show a significantly greater drive for thinness and require a longer length of hospitalization. This association was not present with total strength exercise levels. Effective clinical methods to modify total aerobic exercise levels need to be determined with inpatient populations.     

The EAT and EDI: Screening,interrelationships, and psychometrics

This study examined the Screening utility, scale interrelationships, and psycometric properties of the Eating Attitudes Test (EAT) and the Eating Disorder Inventory (EDI). 12% and 8% of female college freshman (N =268) were identified as weight preccupied according to the EAT and EDI, respectively. Cross-classification analyses indicated 90% agreement between the total EAT score and the EDI Drive for Thinness subscale. Both instruments demonstrated acceptable levels of internal consistency and scale intercorrelations. The principal component analysis provided moderate support for the factorial integrity, and thus scoring, of the EDI. The screening potential and multidimensional nature of the EDI are discussed.     

Dietary fat reduction behaviors in college students: relationship to dieting status, gender and key psychosocial variables

The primary objectives were to assess dietary fat reduction/avoidance behaviors within a sample of college students, and to assess the strength of the relationship between self reported fat avoidance and a number of variables including body mass index (BMI), self-esteem, and responses to the Eating Disorder Inventory (EDI) and Eating Attitudes Test (EAT). A total of 210 female and 114 male undergraduate students were administered a food habits questionnaire (which assessed four dietary fat reduction behaviors), the EDI, the dieting subscale of the EAT, and the Coopersmith Self-Esteem Inventory. Measured heights and weights were used to compute BMI. Thirty-eight percent of the females and 13% of the males reported that they had dieted with the express purpose of losing weight in the past 12 months. The finding that females in general and female dieters in particular, scored higher on the EAT dieting subscale, and relied on three of the four dietary fat reduction behaviors to a greater extent than did males, supports the assertion that women rely heavily on dietary fat avoidance as a method to reduce caloric intakes. In females, the finding that a greater degree of fat avoidance was associated with significantly lower levels of self-esteem and higher scores on the EAT and on six of the eight EDI subscales suggested that fat avoidance may be a predictor of eating pathology and/or psychosocial problems in college-aged women.     

Patterns of physical activity in college students at possible risk for eating disorder

The relationship between exercise patterns and possible risk for eating disorders was examined in a survey of undergraduates using the Eating Attitudes Test (EAT). Reported hours of jogging per week correlated positively and significantly with total EAT scores and with scores on the “Dieting” factor. The Dieting factor also showed significant correlations with number of activities and total hours of activities. Subjects with EAT scores at or above 30 showed a significantly higher mean number of hours of jogging per week than subjects with EAT scores less than 30. Total EAT scores and each of the three factor scores showed a significant positive correlation with exercising alone. Overall levels of activity and hours spent in activities other than jogging did not relate to total EAT scores nor differentiate between the high and low EAT score groups. The apparent preference for jogging among those possibly at risk for eating disorders is discussed in terms of sociocultural and dynamic factors associated with eating disorders.     

The specific content of core beliefs and schema in adolescent girls high and low in eating disorder symptoms

The specific content of core beliefs and schema in adolescent girls high and low in eating disorder symptoms was investigated using the Eating Disorder Belief Questionnaire (EDBQ)-negative self-beliefs sub-scale and Young Schema Questionnaire (YSQ). Girls with high Eating Attitude Test (EAT) scores had higher scores on both measures (using total and sub-scale scores) than girls with low EAT scores. However, all scores were related to eating disorder as well as depressive symptoms. Analysis of individual items on the EDBQ and YSQ found that scores on all EDBQ, but not all YSQ items, differed between the two groups, with higher scores in the high EAT group. However, a number of individual items (on both measures) in the high EAT group, as well as in the low EAT group, were related specifically to eating disorder symptoms and not also to depressive symptoms.     

The importance of the function of exercise in the relationship between obligatory exercise and eating and body image concerns

This study tested whether exercising in response to negative affect moderates the association between obligatory exercise and eating and body image psychopathology. Participants (n = 226) completed the Eating Disorders Examination-Questionnaire (EDE-Q), Obligatory Exercise Questionnaire (OEQ), and a question assessing whether they ever exercise in response to negative affect. In total, 132 (58.4%) participants endorsed exercising in response to negative affect. Multiple regression analyses revealed significant main effects of negative affect motivated exercise, OEQ total scores, and gender on all four EDE-Q subscales and significant interactions of negative affect motivated exercise and OEQ scores on the Eating Concern, Shape Concern, and Weight Concern scales but not the Restraint scale of the EDE-Q. Obligatory exercisers may not demonstrate elevated eating and body image concerns in the absence of negative affect motivated exercise, providing further support of the importance of the function of exercise.     

Correspondence between the clinical assessment of eating-disordered patients and findings derived from questionnaires at follow-up

Questionnaire scores were compared across three outcome groups in a follow-up of eating-disordered patients with onset of the disease during adolescence. Among a total of 40 patients, 7 continued to suffer from either anorexia or bulimia nervosa. Seven additional patients displayed partial syndromes and 26 had recovered. Both a semantic differential measuring body image and the Eating Attitudes Test (EAT) demonstrated limited evidence of clinical sensitivity by differentiating the three outcome groups. The Eating Disorders Inventory (EDI) did not discriminate the three outcome groups. Obviously, clinical interviews and self-report questionnaires tackle different facets of the eating disorders and, therefore, both should be used in outcome studies. © 1993 by John Wiley & Sons, Inc.     

Decreased resting metabolic rate in ballet dancers with menstrual irregularity.

We studied 21 ballet dancers aged 19.4 +/- 1.4 years, hypothesizing that undernutrition was a major factor in menstrual irregularity in this population. Menstrual history was determined by questionnaire. Eight dancers had always been regular (R). Thirteen subjects had a history of menstrual irregularity (HI). Of these, 2 were currently regularly menstruating, 3 had short cycles, 6 were oligomenorrheic, and 2 were amenorrheic. Subjects completed a weighed dietary record and an Eating Attitudes Test (EAT). The following physiological parameters were measured: body composition by anthropometry, resting metabolic rate (RMR) by open-circuit indirect calorimetry, and serum thyroid hormone concentrations by radioimmunoassay. R subjects had significantly higher RMR than HI subjects. Also, HI subjects had lower RMR than predicted by fat-free mass, compared to the R subjects. Neither reported energy intake nor serum thyroid hormone concentrations were different between R and HI subjects. EAT scores varied and were not different between groups. We concluded that in ballet dancers, low RMR is more strongly associated with menstrual irregularity than is current reported energy intake or serum thyroid hormone concentrations.     

Association between a vasopressin receptor AVPR1A promoter region microsatellite and eating behavior measured by a self-report questionnaire (Eating Attitudes Test) in a family-based study of a nonclinical population

OBJECTIVES: Considerable evidence including twin and family studies suggests that biologic determinants interact with cultural cues in the etiology of anorexia and bulimia nervosa. A gene that makes "biologic sense" in contributing susceptibility to these disorders, and to our knowledge not previously investigated for this phenotype, is the vasopressin receptor (AVPR1A), which we have tested for association with eating pathology. METHODS: We genotyped 280 families with same-sex siblings for two microsatellites in the promoter region of the AVPR1A gene. Siblings completed the 26-item Eating Attitudes Test (EAT) and the Drive for Thinness (DT) and Body Dissatisfaction (BD) subscales of the Eating Disorders Inventory (EDI). The Quantitative Transmission Disequilibrium Test program (QTDT), which employs flexible and powerful variance-components procedures, was used to test for an association between EAT scores and the two AVPR1A promoter region microsatellites, RS1 and RS3. RESULTS: A significant association (p = .036) was detected between the RS3 microsatellite and EAT scores. The strongest association was between RS3 and the Dieting subscale of the EAT (p = .011). A significant association was also observed between the EDI-DT and the RS3 microsatellit (p = .0450). CONCLUSIONS: We demonstrate for the first time an association between a microsatellite polymorphism in the AVPR1A promoter region and scores on the EAT as well as with the EDI-DT. The strongest association was observed between the RS3 microsatellite and the Dieting subscale of the EAT. The relevant phenotype appears to tap severe dietary restriction for weight loss purposes.     

Eating behaviors,nutritional status,and menstrual function in elite female adolescent volleyball players

Nutritional status, eating behaviors and menstrual function was examined in 23 nationally ranked female adolescent volleyball players using a health/weight/ dieting/menstrual history questionnaire, the Eating Disorder Inventory (EDI), and the Body Shape Questionnaire (BSQ). Nutrient and energy intakes (El) and energy expenditure (EE) were determined by 3-d weighed food records and activity logs. Iron (Fe), vitamins C, B12, and Folate status were assessed using serum and whole blood. Mean El (2248 +/- 414 kcal/d) was less than EE (2815 +/- 306 kcal/d). Mean carbohydrate (5.4 +/-1.0g/kg/d) and protein (1.1+/-0.3g/kg/d) intakes were below recommended levels for highly active women. Mean intakes for folate, Fe, Ca, Mg, and Zn were less than the respective RDAs/DRIs and almost 50% of the athletes were consuming less than the RDAs/DRIs for the B-complex vitamins and vitamin C. Three athletes presented with Fe deficiency anemia (Hb <12 mg/dL), while marginal vitamin B12 status (<200 pg/ml) and vitamin C status (<28 mmol/L) were found in 1 and 4 athletes, respectively. Approximately 1/2 of the athletes reported actively "dieting". Mean BSQ and EDI subscales scores were within the normal ranges; yet, elevated scores on these scales were reported by 26% and 35% of athletes, respectively. Past or present amenorrhea was reported by 17% of the athletes and 13% and 48%, reported past or present oligomenorrhea and "irregular" menstrual cycles, respectively. These results indicate that elite adolescent volleyball players are at risk for menstrual dysfunction and have energy and nutrient intakes that place them at risk for nutritional deficiencies and compromised performance.     

Beliefs about exercise: Relationship to eating psychopathology and core beliefs among young female exercisers

ObjectiveThis study had two objectives. First, to determine links between levels of eating psychopathology and beliefs about exercise among young women. Second, to determine the predictive effects of unhealthy core beliefs on exercise beliefs.MethodA convenience sample of 185 young female exercisers completed the Eating Disorders Inventory (EDI), the Exercise Beliefs Questionnaire (EBQ) and the Young Schema Questionnaire (YSQ).ResultsThe participants' mean scores on the EDI were as follows: drive for thinness = 0.69 (SD = 0.82); body dissatisfaction = 1.30 (SD = 0.86); and bulimia = 0.33 (SD = 0.42). There were significant, positive correlations of all three EDI scales with EBQ social and EBQ appearance subscales. In addition, YSQ Defectiveness/Shame beliefs predicted EBQ social scores, while YSQ Unrelenting Standards predicted EBQ appearance scores.DiscussionThose women with relatively unhealthy eating attitudes are likely to believe that exercise will prevent negative social consequences, and are likely to be motivated to exercise in order to preserve or enhance their physical appearance. These same types of exercise belief (regarding social consequences and appearance) are predicted by feelings of defectiveness and shame and by unrelenting high personal standards.     

The stability of the eating disorder inventory

This research assessed the stability of the Eating Disorder Inventory (EDI; Gamer, Olmstead, & Polivy [1983]. International Journal of Eating Disorders, 2, 75–34). over a 7-year period in a female, non-clinical population. Of 401 women who initially completed the EDI, 282 (70.3%) completed the EDI again 1 year later. Pearson correlations calculated between the EDI total score and its eight subscale scores indicate considerable stability in the total score and the Drive for Thinness, Body Dissatisfaction, Ineffectiveness, Perfectionism, and Interpersonal Distrust subscales of the EDI. There was less stability in the Bulimia, Interoceptive Awareness, and Maturity Fears subscales. Several explanations of these findings are discussed.     

Eating attitudes and behaviors in adolescent women: Discrimination of normals,dieters, and suspected bulimics using the eating attitudes test and eating disorder inventory

The prevalence of eating disorders has increased in teenage and young adult women. Since early treatment of eating disorders is thought to lead to improved outcomes, the development of an effective self-report screening device for eating disorders in adolescents could help reduce morbidity from these illnesses. In order to examine the feasibility of developing such an instrument, eating attitudes and behaviors were studied in a sample of 72 female adolescents, using the Eating Attitudes Test (EAT), the Eating Disorder Inventory (EDI), and blinded structured clinical interviews focused on eating disorders. Scores on the EAT and EDI were similar to scores of older normal female comparison groups in other studies. Based upon interview information, 54 subjects were classified as normal, 9 as dieters, 8 as suspected bulimics, and 1 as bulimic. Subjects, on average, accurately reported their actual weight and height. The results of the interviews indicated that 25% of the 72 subjects were currently dieting, 60% regularly skipped meals, 35% said they were overweight or very overweight, 22% were binge eaters, 8% vomited after eating, and 4% used drugs to lose weight. Linear discriminant analysis was used to compare the results of the questionnaires with that of the interviews in order to identify a composite set of questionnaire items to be used as a screening device for individuals with clinical eating disorders and for those at high risk of developing such disorders. A preliminary equation was obtained that correctly classified 86% of subjects into categories of normal, dieter, or suspected bulimic that were determined from the interviews. These preliminary data suggest that adolescents' scores on the EAT and the EDI are similar to young adult women's scores, and also that further refinement of a discriminant equation might lead to the development of an effective screening device for eating disorders in adolescents.     

Patterns of menstrual disturbance in eating disorders

OBJECTIVE: To describe menstrual disturbance in eating disorders (ED). METHOD: We describe menstrual history in 1,705 women and compare eating, weight, and psychopathological traits across menstrual groups. RESULTS: Menstrual dysfunction occurred across all eating disorder subtypes. Individuals with normal menstrual history and primary amenorrhea reported the highest and lowest lifetime body mass index (BMI), respectively. Normal menstruation and oligomenorrhea groups reported greater binge eating, vomiting, and appetite suppressant use. Amenorrhea was associated with lower caloric intake and higher exercise. Harm avoidance, novelty seeking, perfectionism, and obsessionality discriminated among menstrual status groups. No differences in comorbid Axis I and II disorders were observed. CONCLUSION: Menstrual dysfunction is not limited to any eating disorder subtype. BMI, caloric intake, and exercise were strongly associated with menstrual function. Menstrual status is not associated with comorbidity. Menstrual irregularity is an associated feature of all ED rather than being restricted to AN only.     

Six-month open trial of haloperidol as an adjunctive treatment for anorexia nervosa: a preliminary report

OBJECTIVE: To evaluate the effectiveness of haloperidoll as an adjunctive treatment for resistant anorexia nervosa restricting subtype (AN-R). METHOD: Thirteen outpatients with treatment-resistant AN-R were treated for 6 months with haloperidol in addition to standard treatment. Treatment resistance was defined as persistent and resistant anorectic symptoms despite multiple standard therapies. Assessments were carried out at baseline and after 1, 3, and 6 months with the Eating Disorder Inventory (EAT), the Eating Attitude Test (EAT), and the Clinical Global Impression and Improvement Scale (CGI-I). RESULTS: Significant change from baseline to end point was observed on EDI total score (p =.02) and on the subscales Drive for Thinness (p =.009), Bulimia (p =.01), and Interoceptive Awareness (p =.02), as well as on the EAT (p =.009) and CGI scores (p =.001). Body mass index changed significantly from baseline (15.7 +/- 1.9) to end point (18.1 +/- 2.5; p =.03). DISCUSSION: These preliminary data suggest that low doses of haloperidol might be effective as an adjunctive treatment for patients with severe AN-R. Larger controlled studies are warranted to confirm these data.     

Disorders of the female athlete triad among collegiate athletes

This study examined the prevalence of and relationship between the disorders of the female athlete triad in collegiate athletes participating in aesthetic, endurance, or team/anaerobic sports. Participants were 425 female collegiate athletes from 7 universities across the United States. Disordered eating, menstrual dysfunction, and musculoskeletal injuries were assessed by a health/medical, dieting and menstrual history questionnaire, the Eating Attitudes Test (EAT-26), and the Eating Disorder Inventory Body Dissatisfaction Subscale (EDI-BD). The percentage of athletes reporting a clinical diagnosis of anorexia and bulimia nervosa was 3.3% and 2.3%, respectively; mean ( SD) EAT and EDI-BD scores were 10.6 9.6 and 9.8 7.6, respectively. The percentage of athletes with scores indicating "at-risk" behavior for an eating disorder were 15.2% using the EAT-26 and 32.4% using the EDI-BD. A similar percentage of athletes in aesthetic, endurance, and team/anaerobic sports reported a clinical diagnosis of anorexia or bulimia. However, athletes in aesthetic sports scored higher on the EAT-26 (13.5 10.9) than athletes in endurance (10.0 9.3) or team/anaerobic sports (9.9 9.0, p <.02); and more athletes in aesthetic versus endurance or team/anaerobic sports scored above the EAT-26 cut-off score of 20 (p <.01). Menstrual irregularity was reported by 31% of the athletes not using oral contraceptives, and there were no group differences in the prevalence of self-reported menstrual irregularity. Muscle and bone injuries sustained during the collegiate career were reported by 65.9% and 34.3% of athletes, respectively, and more athletes in aesthetic versus endurance and team/anaerobic sports reported muscle (p =.005) and/or bone injuries (p <.001). Athletes "at risk" for eating disorders more frequently reported menstrual irregularity (p =.004) and sustained more bone injuries (p =.003) during their collegiate career. These data indicate that while the prevalence of clinical eating disorders is low in female collegiate athletes, many are "at risk" for an eating disorder, which places them at increased risk for menstrual irregularity and bone injuries.     

Application of the eating disorders inventory to a sample of black,white, and mixed race schoolgirls in zimbabwe

This paper describes the application of Garner's Eating Disorders Inventory (EDI) to 399 black, white, and mixed race schoolgirls attending private schools in Zimbabwe. To the authors' knowledge the instrument has not been applied previously on the African continent. The results obtained indicate concern with eating and weight amongst the black and mixed race groups when compared with previous studies of indigenous subjects in de- veloping countries (Buhrich, 198 1). High scorers are isolated on the basis of obtaining scores on or above the 90th percentile (normative sample com- prised of North American female college students N = 770)on the “drive for thinness” subscale of the ED/. There are 80 high scorers of which 12.5% are black and 17.5% of mixed race, the remainder are white. There is a bulimic tendency amongst the high-Scoring black and mixed race groups. The results of the high scores are further evaluated on the remaining dimensions of the EDI, namely, body dissatisfaction, ineffectiveness, interpersonal distrust, per- fectionism, interoceptive awareness and maturity fears. Finally, a small sample of black, white, and mixed race are engaged in a semistructured interview—-EAT. (The EAT was constructed by Dr. C. Szmukler, Maudsley Hospital, London). This semistructured interview assesses attitudes towards (inter alia) increased interest in certain foods and avoidance of others, de- tailed calorie counting, bingeing, dieting, fear of loss of control over eating, overactivity, etc. The results are reviewed against a background of traditional and evolving cultural values which are juxtaposed within this developing so- ciety.     

The significance of self-induced vomiting as a weight control method among nonclinical samples

The aim of the current study is to determine the viability of employing self-induced vomiting as an index of psychopathology in nonclinical samples. Eating Disorder Inventory (EDI) profiles were obtained for 77 college and ballet students who reported engaging in self-induced vomiting as a method of weight control; these profiles were compared with those of 366 women who reported never having engaged in self-induced vomiting. Cluster analyses of individuals were performed to examine the structure of EDI subscale scores for both groups of women. Three clusters of vomiting women were derived. The first consisted of approximately 10% of the vomiting sample, and they had EDI subscale scores that were at or above the mean for bulimia nervosa patients. The second cluster contained women who had elevated EDI subscales scores related to attitudes about eating and body shape but were in the normal range on other scales of psychological functioning. The third cluster was composed of 47% of the vomiting women, and they had normal to high-normal mean EDI scores. The mean frequency of vomiting did not distinguish the three clusters. Similar clusters were obtained for the nonvomiting sample of women. It is concluded that relying upon self-induced vomiting or simple measures of weight preoccupation without considering other dimensions of psychological functioning is of limited value in identifying the presence of a clinically significant eating disorder.     

Correlation between the EAT-26 and the EAT-40, the Eating Disorders Inventory,and the Restrained Eating Inventory

The 40-item Eating Attitudes Jest, the 26-item Eating Attitudes Test, the Restrained Eating Inventory (El), and the Eating Disorders Inventory (EDI) were given to 81 women who were participating in an eating disorders study. The EAT-26 totals and the scores of the three factors were intercorrelated. The scores of the EAT-26 were also correlated with the scores of the EAT-40, and El, and the EDI. The EAT-26 correlated significantly with all three of its factors as well as with the EAT-40. These results suggest that the EAT-26 is a reliable and economical substitute for the EAT-40. The EAT-26 correlates highly with the eight subscales of the EDI and with the El, suggesting concurrent validity.     

A transcultural comparison of the eating disorder inventory in former east and West Berlin

The Eating Disorder Inventory (EDI) was used in a transcultural study of East and West Berlin clinical and nonclinical samples at a time when the two parts of the city were still strictly separated. East Berlin patients with anorexia nervosa scored significantly lower than their Western counterparts with regard to the majority of EDI scales. Furthermore, there was a lack of discriminant validity for the -East Berlin clinical and nonclinical samples. It was speculated that this deficit might reflect a lack of transcultural validity in the psychological constructs upon which the questionnaire is based. However, the study revealed some evidence of meaningful correlations with clinical parameters. Finally, evidence was presented supporting internal consistency of the subscales and convergent validity of the EDI with the Eating Attitudes Jest (EAT). © 1992 John Wiley & Sons, Inc.