Moderation: an alternative to restraint as a mode of weight self-regulation.

This study considered two types of eating and weight self-regulation, in five groups, including four types of weight controllers and one non-dieting group. New scales were developed to measure eating moderation and restraint. Moderation was largely uncorrelated with restraint in 4 groups and had a fairly strong positive relation in 1 group. The moderation scale was unrelated to the Dutch Eating Behavior Questionnaire (DEBQ) restraint scale and the Three Factor Eating Questionnaire (TFEQ) rigid restraint subscale and weakly positively related to TFEQ flexible restraint. The restraint scale was strongly correlated to the DEBQ restraint scale, and to both flexible and rigid restraint subscales of the TFEQ. Across the five groups, moderation had exclusively positive relationships with attitude, behavior and emotion variables, while restraint had primarily negative relationships. The study supports moderation as a new dimension of weight self-regulation, independent of restraint. The new measures of moderation and restraint can be used together in research on the processes of change in weight management.

Positive and negative dimensions of weight control motivation.

This study examined weight control motivation among patients (N=5460 females and 547 males) who sought weight loss treatment with family physicians. An eight-item measure assessed the frequency of thoughts and feelings related to weight control "outcome" (e.g. expected physical and psychological benefits) and "process" (e.g. resentment and doubt). Factor analysis supported the existence of two factors, labeled Positive and Negative motivation. Positive motivation was high (average frequency of thoughts about benefits was 'every day') and stable throughout treatment, while Negative motivation declined rapidly and then stabilized. The determinants of changes in the Positive and Negative dimensions during treatment were examined within 3 time frames: first month, months 2-6, and 6-12. Maintenance of high scores on Positive motivation was associated with higher BMI and more disturbed eating habits. Early reductions in Negative motivation were greater for those starting treatment with higher weight and more disturbed eating habits, but less depression and stress, while later reductions in Negative motivation were predicted by improvements in eating habits, weight, stress and perfectionism. Clinicians treating obesity should be sensitive to fluctuations in both motivational dimensions, as they are likely to play a central role in determining long-term behavior and weight change.

Early treatment response as a predictor of ongoing weight loss in obesity treatment.

OBJECTIVES: This study examined early treatment response in obesity treatment, defined as early change in body mass index (BMI) and early change in eating behaviour, as a predictor of ongoing weight loss in obese patients. METHODS: We conducted a repeated measures analysis of eating behaviour, emotional factors (depression, stress, perfectionism) and BMI, over a 9 month period. Subjects were 344 females, aged 18-65 (mean = 41.8), with a BMI of at least 25 (mean BMI = 33.7), engaged in very-low calorie (VLCD) or low-calorie (LCD) diets. RESULTS: Multi-level modelling identified four significant predictors of ongoing weight loss (weight loss occurring between 5 weeks and 9 months after the start of treatment). These included: type of diet, early BMI change (start to 5 weeks), number of weigh-ins and the early change in uncontrolled eating (start to 5 weeks). Estimates based on multi-level modelling indicate that patients with strong versus weak early treatment responses would be expected to show large differences in ongoing weight loss. CONCLUSIONS: Early improvements in eating behaviour and weight appear to have additive effects in the prediction of ongoing weight change. Future research is required to identify the optimal rate of weight loss, whether there are critical periods for behaviour change, and factors which influence the likelihood of early behaviour change.

Convergent validity of the Larocque Obesity Questionnaire and self-reported behavior during obesity treatment.

A validation study of the Larocque Obesity Questionnaire designed for use in obesity treatment was performed. Unlike other measures of obesity, this questionnaire includes scales measuring general emotional state as well as eating behavior. Subscales measured uncontrolled eating, physical stress responses, depression, and perfectionism. Subjects were 458 women and 79 men in treatment for obesity by general practitioners. The subscales showed acceptable internal consistency and related in predictable ways to measures of eating behavior, depression, self-criticism, stress, physical complaints and weight-control motivation. Subjects in the heaviest weight category (Body Mass Index > or = 40) showed higher scores on Stress Response and Depression subscales. There were no significant sex differences, after controlling for weight. All four subscales showed significant improvement after 5 wk., which indicates their sensitivity to changes during treatment.

Hypersensitivity to social interactions in bulimic syndromes: relationship to binge eating.

This study used a 6- to 22-day experience-sampling procedure to test for hypersensitivity to social interactions in bulimic individuals. Ratings on daily social interactions, self-concepts, moods, and eating behaviors from 55 actively bulimic, 18 formerly bulimic, and 31 noneating-disordered women were obtained. Hierarchical linear modeling analyses showed negative social interactions to be associated with significant increases in self-criticism (SC) and deteriorations in mood in all participants. However, bulimic participants (active or in remission) showed larger increases in SC following negative social interactions than did normal eaters (and thus seemed to be hypersensitive to interpersonal experiences). Additional analyses indicated that binge episodes tended to be preceded by poorer than average social experiences, self-concepts, and moods and followed by deteriorations in self-concept, mood, and social perception. The authors discuss possible perpetuating influences of interpersonal sensitivity in bulimic syndromes.

A controlled study of trait narcissism in anorexia and bulimia nervosa.

INTRODUCTION: Many theories attribute anorexia and bulimia nervosa to "pathological narcissism," but this conception has not been adequately evaluated. METHOD: We compared the scores of 90 eating disorder (ED) sufferers (23 anorexic restricters, 14 anorexic bingers, and 53 bulimics) with those of 36 psychiatric-control (PC) and 54 normal-control (NC) females on validated self-report scales measuring Narcissism, Affective Instability, Stimulus Seeking, Compulsivity, and Restricted Expression. RESULTS: Narcissism scores of ED patients (whether restricters or bingers/purgers) consistently exceeded those of the PC and NC cases, suggesting that Narcissism does indeed load more heavily in the EDs than in other psychiatric disturbances. Conversely, Affective Instability was characteristic of all clinical cases (i.e., of EDs and PCs), Restricted Expression and Compulsivity were characteristic of restricters, and Stimulus Seeking was characteristic of bingers/purgers. DISCUSSION: Results are consistent with the notion that different ED variants may reflect subtype-specific temperaments and/or adaptive styles acting-upon shared underlying narcissistic disturbances.

Prospective study of outcome in bulimics as a function of Axis-II comorbidity: long-term responses on eating and psychiatric symptoms.

We assessed prognostic implications of categorical personality disorder (PD) diagnoses at selected points during and following treatment for bulimic syndromes. Seventy-six bulimic cases were organized into Borderline PD, Other PD, or No PD groups, and then assessed at pretreatment, after 3 months of therapy, and at termination (after a modal 8 months). Results at 3- and 12-month posttreatment follow-ups (available in 43 and 38 cases, respectively) provided a reflection of status after therapy termination. On comorbid symptoms, borderline/ nonborderline differences were very striking: Borderlines showed more initial psychiatric symptoms than did nonborderlines, and despite improvements, retained disturbances of clinical magnitude to completion of therapy and into follow-up. On eating symptoms, borderline/ nonborderline differences were less dramatic, the overall pattern implying that borderlines showed only marginally poorer response in selected areas of disturbance. Results corroborate others linking Axis-II comorbidity to unfavorable outcome in the eating disorders, but suggest that character disturbance is more strongly linked to course of general psychiatric symptoms than it is to course of eating disturbances. We discuss clinical implications of differing strengths of association observed between eating and psychiatric symptoms, on the one hand, and Axis-II pathology, on the other.

Familial eating concerns and psychopathological traits: causal implications of transgenerational effects.

OBJECTIVE: This study extends an earlier investigation on the link between familial traits and eating disorders (EDs), and examines the extent to which selected eating attitudes and psychopathological traits are (a) familial in nature and (b) specific to anorexia- and bulimia-spectrum disorders. METHODS: We measured various ED-relevant dimensions (eating and body image attitudes) and psychopathological traits (e.g., affective instability, narcissism, compulsivity, restricted expression) in the mothers, fathers, and sisters of probands displaying an ED (n = 88), another psychiatric disturbance (n = 42), or neither disturbance (n = 59). Total sample, including relatives, was 553. RESULTS: A principal components analysis (PCA), used to reduce variables and to characterize main sources of variation, yielded three interpretable factors: eating concerns and symptoms (grouping all eating-related dimensions), dramatic-erratic traits (grouping affective instability, narcissism, and conceptually related dimensions), and obsessive-compulsive traits (grouping compulsivity and restricted expression). Correlations among subjects' factor scores (derived from the PCA) showed two types of transgenerational effects: correspondences between daughters' and parents' psychopathological traits, and between daughters' and mothers' eating concerns. Despite these, relatives of ED probands were not discriminable on any factor score from relatives of controls. DISCUSSION: These results imply that transgenerational effects exist on eating attitudes and psychopathological traits, but do not uniquely identify families in which clinical ED syndromes occur.

Dependency, self-criticism and social interactions.

We investigated the social interactions of college students varying in dependency and self-criticism. Forty-eight college students used a modified version of the Rochester Interaction Record to record quantitative and qualitative features of every 10-minute or longer interaction during a seven-day period. Daily measures of mood were also collected. Dependency was related to more frequent and more intimate interactions, and self-criticism was negatively related to pleasantness of social interactions. Although dependency and self-criticism were both associated with daily dysphoria, the social interaction findings could not be attributed to the effects of mood. The social environments associated with dependency and self-criticism may influence the aetiology and course of depressive episodes.

Controlled study of eating concerns and psychopathological traits in relatives of eating-disordered probands: do familial traits exist?

To examine the extent to which first-degree relatives of eating-disordered (ED) probands endorse maladaptive eating attitudes and personality/affective traits, we compared self-reported eating concerns (Restraint, Emotional Eating, Body Dissatisfaction, and maladaptive eating attitudes) and psychopathological traits (Affective Instability, Anxiousness, Compulsivity, and Narcissism) across groups of restricter (n = 19), binger (n = 56), psychiatric control (PC, n = 38), normal dieter (ND, n = 29), and nondieter control (NC, n = 28) probands, and then across participating nuclear family members. Results among probands were as anticipated: ED probands showed expected elevations in both areas, and predicted restricter/binger differences were obtained. However, corresponding differences were not obtained on measures of mothers', fathers', or siblings' eating concerns and traits. Our findings corroborate the notion that EDs represent a convergence of eating, affective, and personality disturbances, but not that such a clustering of features exists as a familial trait. We discuss normal trait and attitudinal variations observed in ED probands' relatives in light of findings showing EDs and other psychiatric syndromes to aggregate within families.

Prognostic implications of stable versus transient "borderline features" in bulimic patients.

BACKGROUND: This study examined the association between "borderline features" and treatment response in bulimic patients. METHOD: Treatment response was assessed in 69 bulimic patients over 6 months of treatment (and 1-year response in 44 of the patients). Patients were classified as a function of whether "borderline features" were (1) "stably" present (at pretreatment and after 3 months), (2) "transiently" present (at pretreatment only), or (3) "absent" (at both time points). RESULTS: The stably borderline profile coincided systematically with Axis II comorbidity (not with mood disorders) and was generally predictive of poorer 6- and 12-month response of eating and comorbid symptoms. Hierarchical regressions showed predictive effects to have a partial independence from effects caused by concurrent eating symptoms and depression. CONCLUSION: Results highlight the importance, with respect to prognosis in bulimia nervosa, of establishing the temporal stability of features that imply character disturbance.

Situational dieting self-efficacy and short-term regulation of eating.

The Situation-Based Dieting Self-Efficacy Scale (SDS) measures dieters' beliefs in their abilities to adhere to a diet in eating situations. The scale has high internal consistency and test-retest reliability. Female dieters were recruited who were either high or low on SDS, and assigned either to consume a high calorie preload or to a no preload condition. Subjects with high SDS ate less in a subsequent "taste rating" task than subjects with low SDS. Amount eaten was not predicted by hunger, weight, palatability, mood, or restrained eating style. This finding challenges the tendency to view all dieters as equally likely to lapse in response to dietary challenges.

A new measure of weight locus of control: the Dieting Beliefs Scale.

This article describes the construction and preliminary validation of a new scale of weight locus of control, the Dieting Beliefs Scale. The 156-item scale demonstrated moderate internal consistency and high test-retest reliability in a sample of undergraduate women. Principal-components analysis suggested three factors. The three factors were interpretable and had distinct relations with a variety of weight-related and psychological variables. The results suggest that weight locus of control is a multidimensional construct, and they provide a possible explanation for the inconsistent findings concerning the relation between weight locus of control and dieting success. Implications for the study of dieting relapse and for the construction of treatment programs are discussed.