Hal S. Currey

Development and validation of the Eating Behavior Inventory

The Eating Behavior Inventory (EBI) is a self-report instrument for assessing behaviors that have been theoretically implicated in weight loss, e.g., self-monitoring of food intake and of weight, refusing offers of food, eating at only one place, shopping from a list, eating in response to emotions. Thirty items were constructed in the form of first-person statements, e.g., “I eat in the middle of the night.” Each item was to be rated with a 5-point scale according to how often it was true for the respondent. Items were scored such that higher scores always reflected more “appropriate” (theoretically facilitative of weight control) eating patterns. Validity of individual items and total score was assessed in four studies. Twenty-six of the original items appeared valid and were retained. The resulting total score demonstrated validity in these studies and in two cross-validational comparisons. Internal consistency as measured by split-half reliability and correlations of item scores with total score was acceptable. One month test-retest reliability of item and total scores was satisfactory. Clinical and research applications of the EBI are discussed.

Effects of sex of subject and spouse involvement on weight loss in a behavioral treatment program: A retrospective investigation.

Abstract A subject × treatment model was used to examine the effects on weight loss of sex of subject and spouse involvement in behavioral treatment. Eleven males and 9 females who had completed a 12-week session of behavioral obesity treatment with their spouses, and individually matched male and female subjects who had completed treatment without their spouses, were assessed at follow-up 9–14 months after treatment. Males demonstrated a significantly greater weight,loss than did females on absolute body weight, weight change, percent change in body weight, and the Reduction Index. Males also appeared to maintain weight loss better than females. Spouse involvement had no effect on weight loss during treatment or over follow-up. Differences in weight loss between sexes could not be attributed to differences in age, initial percent overweight, age of onset of obesity, socioeconomic status, number of previous behavioral treatment sessions, self-reported activity level, or self-reported inappropriate eating patterns. Possible explanations of male-female differences in response to treatment are examined, and methodological implications of findings for obesity treatment outcome studies are discussed.

Experienced Control and Participation in Treatment of Obesity

Results of three studies are presented which explore the relation of experienced control to participation in treatment for obesity. Subjects in all studies were adult females participating in an outpatient, behaviorally-oriented obesity program. Experienced control was measured using the Experienced Control Scale (Tiffany, 1967). Study 1 was a naturally occurring experiment which compared two groups (ns = 7 and 8) that differed in extent of self-directed efforts to begin treatment. Study 2 compared 32 persons who completed the program and 18 persons who did not complete it. High levels of experienced control over internal forces were associated with increased efforts to initiate treatment and with completion of treatment. Study 3 assessed the temporal stability of the Experienced Control Scale with obese females tested at the beginning and end of the 12-wk. treatment program. Test-retest reliability coefficients were .64 (p < .01) for the internal control subscale and .29 (P < .10) for the external control subscale.

Restraint and age at onset of obesity

Abstract Dietary restraint in 30 obese females was examined in relation to age at onset of obesity. Juvenile-onset obese individuals were found to experience significantly more restraint in their eating behavior than adult-onset obese individuals, based on responses to Hermans revised Restraint Questionnaire. Restraint score was significantly correlated with both age at onset and duration of obesity, but not with weight, percent overweight, or present age. Possible biological and psychosocial contributors to the age at onset effect are discussed, as are implications of restraint differences for treatment.