The relationship of gender and family environment to eating disorder risk in adolescents

Adolescence, Winter, 1994 by Kenneth R. Felker, Cathie Stivers

INTRODUCTION

The success rate for treating anorexia nervosa is abysmal. Only one-half of all persons diagnosed with anorexia nervosa fully recover, and one out of every thirty die as a result (Patton, 1989). Although eating disorders were once considered rare, reports of a steady increase have appeared with relative frequency in both professional and popular literature (Shainess, 1979; Schwartz, Thompson, & Johnson, 1982; Kagan & Squires, 1984). Research not only supports this finding of an increase in incidence, but eating disorders are now seen as occurring in all socioeconomic classes (Brownell & Foreyt, 1986).

Anorexia nervosa and bulimia are primarily a problem of young women, with bimodal risk ages of onset determined to be 14 and 18 years of age (Halmi, Casper, Eckert, Goldberg, & Davis, 1979). A prevalence study of bulimia in adolescents (Gross & Rose, 1988) examined 1,373 high school students and identified bulimia in 9.6% of the girls and 1.2% of the boys. This finding relative to prevalence of bulimia is similar to findings previously reported (Gray & Ford, 1985; Pyle et al., 1983). Halmi, Frank and Schwartz (1981) studied 539 college students and identified 13% who met the clinical criteria for bulimia while 3% were diagnosed as anorexic. Incidence of anorexia nervosa has increased from 0.55 per 100,000 during 1960-1969 to 3.26 per 100,000 during the 1970-1976 period (Jones, Fox, Babegan, & Hutton, 1980). There is also an increase in the prevalence of anorexia nervosa in western cultures in comparison to eastern cultures, which would suggest that sociocultural factors play a significant role in the development of this eating disorder (Buhrich, 1981). Increases in incidence and prevalence of eating disorders can be attributed to both a rise in the number of cases and improved methods of diagnosis. Even with improved eating disorder identification, two factors hinder accurate reporting within a symptomatic population. First, anorexics do not view themselves as abnormal and therefore do not seek treatment. Second, bulimics are usually aware that a problem exists, but in many cases are too embarrassed to seek help (Theander, 1970).

A combination of factors influence the development of anorexia nervosa and bulimia. Anderson (1983) identified perfectionism, a society that emphasizes thinness, a family that manifests depression, and biological predisposition as primary predisposing factors. Clinical studies indicate that prolonged restrictive dieting, loss or separation, difficulties in handling particular emotions, impulsivity, fear of heterosexual relations, and depression usually precede the onset of an eating disorder (Johnson, Stuckey Lewis, & Schwartz, 1983). Although research supports a correlation between family environment and the development of an eating disorder in adolescents, many parents do not understand eating disorders or know what they can do with regard to family environment to decrease the likelihood (Moye, 1985; Wirth, 1986; Stern et al., 1989).

Family environment also appears to play a significant role in the prognosis of eating-disordered patients. Several studies have identified conditions associated with a poorer outcome. These conditions include an older age of onset, longer duration of illness, lower body weight at presentation, poorer adjustment in childhood, disturbed family relationships, and a history of previous psychiatric treatment (Morgan & Russell, 1975; Hsu, Crisp, & Harding, 1979). An examination of conditions associated with both the risk of developing an eating disorder and the prognosis reveals the importance of family environment when developing prevention strategies.

Several instruments for identifying persons with eating disorders have been developed during a ten-year period beginning in 1979. The Eating Attitudes Test (EAT; Garner & Garfinkel, 1979) and the Eating Disorder Inventory (EDI; Garner, Olmstead, & Polevy, 1983) are two standardized measures that are used to assess attitudes and behaviors characteristic of eating-disordered individuals (Silberg, Gross, & Rosen, 1988). The EAT was developed in 1979 and has been used extensively to identify symptoms of anorexia nervosa in both clinical and community samples. In 1983 the EDI was developed to measure anorexic and bulimic symptomology as well as to identify attitudinal and behavioral traits relevant to these eating disorders (Slade et al., 1986).

The Setting Conditions for Anorexia Nervosa Scale (SCANS; Slade et al., 1986) was developed in 1986 and was designed to identify individuals who are at risk of developing anorexia nervosa or bulimia. The SCANS is nontransparent and thus addresses the denial component commonly associated with eating disorders. Prior to the development of this instrument, clinicians used the EAT and EDI to measure symptoms and attitudes only after recognizable symptoms became observable. The development of the SCANS enables the health professional to work toward prevention of eating disorders and to answer some of the following questions: Do the same correlates that apply to eating-disordered individuals hold true for those at risk of developing an eating disorder? Are there additional correlates present in at-risk individuals that are not present in those who have been diagnosed as eating disordered, and what causes one adolescent to be at risk when another is not? The SCANS questionnaire, by its ability to measure risk of developing an eating disorder, opens the door to prevention.