In recent years, sociocultural norms have promoted an aesthetic ideal which has produced an increase in anorexic and bulimic disorders, as well as considerable increase in pathological behavior in relation to eating habits. Several epidemiological studies have shown that although eating disorders in the adult population are frequent, the maximum incidence appears during the adolescent period (Hoek, 1991; Rastam, Gillberg, & Garton, 1989). According to these authors, there exists a high quantity of abnormal eating attitudes among adolescents who are neither anorexic nor bulimic patients. These symptomatologies range between 6 and 14% (Ledoux, Choquet, & Flament, 1991; Raich et al., 1991) and include subjects at risk of suffering eating disorders. It is important to note the clinical complexity of these pathologies and their consequences on the emotional level, as well as on the social and educational development of the adolescent.
Many authors have been interested in confirming the existence of a premorbid personality of anorexia nervosa and have found a high prevalence of personality disorder in populations with eating disorders (Kennedy, McKey, & Katz, 1990; Steiger, Liquornik, Chapman, & Hussein, 1991). Other studies have observed that DSM-III-R affective and anxiety disorders were the most common diagnoses in anorexic patients (Hendren, 1983; Piran, Kennedy, Garfinkel & Owens, 1985; Devlin & Walsh, 1989; Halmi et al., 1991; Fornari et al., 1992). Regarding these pathologies, low self-esteem is for many investigators an important risk factor in eating psychopathology (Button, 1990; Ledoux et al., 1991; Wertheim et al., 1992).
The aims of this study were to: (1) identify abnormal eating behavior in a sample of adolescents; (2) investigate their associated psychological, psychopathological, and somatic traits; and (3) elaborate a psychological pattern of risk eating disorders according to gender.
Participants in the study were schoolchildren from the Catalonia city of Reus (Spain), an urban commercial area of 96,000 inhabitants of above-average socioeconomic status. The boys selected were born between the 1st of January and the 30th of June, 1976, and the girls within the same period one year later.
The study was part of a wider longitudinal project to evaluate physical and psychological characteristics and psychopathological disorders. The age difference of one year between the sexes was chosen in order to achieve a close biological similarity level in puberty. The males were aged 14 (n = 290) and the females 13 (n = 225).
Eating Attitudes Test (EAT, Garner & Garfinkel, 1979). The EAT is widely used as a self-rating screening for abnormal eating attitudes and has proved to be efficient in detecting anorexia nervosa and bulimia nervosa. It contains 40 items with six possible answers for each statement ranging from "never" (0) to "always" (3). The cut-off score of 30 shows a potential risk of developing an eating disorder. A translated experimental version validated in Spanish (Toro et al., 1989) was used. Factorial analysis included dieting, bulimia, and oral control factors.
Children's Depression Inventory (CDI, Kovacs, 1983). The CDI is the most commonly used self-report depression inventory for children (ages 9-16) and its validity-reliability has been extensively demonstrated (Kovacs, 1983; Carey et al., 1987; Domenech & Polaino, 1990). It includes 27 items that represent the major symptoms of depression syndrome. Each item has a value between zero and two, according to the degree of current depressive symptomatology manifested. The cutoff score used by Kovacs to indicate probable depression was 19.
State-Trait Anxiety Inventory for Children (STAIC, Spielberger, 1973). The STAIC is a self-evaluation test for anxiety symptomatology which includes 40 items, 20 of which report trait anxiety (STAIC-T) and the other 20 report state anxiety (STAIC-S). …