Academic journal article Journal of Counseling and Development : JCD

The Continuum versus Categorical Debate on Eating Disorders: Implications for Counselors

Academic journal article Journal of Counseling and Development : JCD

The Continuum versus Categorical Debate on Eating Disorders: Implications for Counselors

Article excerpt

For many years a debate has ensued over two views of mental illness: (a) the conceptualization of mental illnesses as a set of categorical (i.e., qualitatively different) disorders, distinct from normal development and from each other, and (b) a perspective that views mental disorders as dimensions occurring along a continuum on which individuals vary in degree but not in kind. From its inception, the Diagnostic and Statistical Manual of Mental Disorders (4th ed., DSM-IV; American Psychiatric Association [APA], 1994) has been based on the categorical approach for its classification system (Adams & Cassidy, 1993). For example, the DSM-IV delineates two primary diagnostic categories for eating disorders: anorexia nervosa, restricting type and binge-eating/purging type; and bulimia nervosa, purging type and non-purging type. The system also includes a residual category, eating disorders not otherwise specified, which encompasses subthreshold diagnoses of anorexia and bulimia nervosa and a new category called binge eating disorder.

Two questions have guided researchers attempting to determine which view is correct for the eating disorders: Do eating disorders fall at the extreme end of a continuum of eating behaviors differing in degree but not in kind from normal and problematic eating behaviors (e.g., Stice, Killen, Hayward, & Taylor, 1998) or are the eating disorders as a group qualitatively different (i.e., differing in kind) from normative eating, dieting, and weight concerns? Similarly, is each eating disorder separate and discrete (i.e., qualitatively different) from the others or are they variations of a single disorder occurring on a continuum of pathology?


In support of the continuum hypothesis, Scarano and Kalodner-Martin (1994) noted that the groups along the continuum (e.g., people who eat normally, people preoccupied with weight, individuals who chronically diet, people who purge, people with subthreshold bulimia, and people with bulimia) share similar psychological characteristics (e.g., dissatisfaction with body image, low self-esteem, fear of becoming fat) and differ only in the frequency or severity of these characteristics. For example, significant differences in interoceptive awareness and feelings of ineffectiveness were found between clinical, subclinical, and asymptomatic women as expected for a linear model. Tylka and Subich (2002) found that many psychological risk factors (e.g., ineffectiveness, interpersonal distrust, and perceptions of the safety of weight control techniques) were related in a linear fashion to eating disorder continuum group placement for high school and college women. Lowe et al. (1996) found support for continuity on measures of general psychopathology for individuals who eat in an unrestrained or restrained manner and individuals with bulimia.

As for the discontinuity perspective, Lowe et al. (1996) found evidence for discontinuity on binge eating severity, and Ruderman and Besbeas (1992) found evidence of discontinuity on various measures of psychopathology. In general, the results of empirical studies conducted to date have been unable to resolve the debate.


Beyond and apart from the diversity of findings on this question is the issue of the appropriateness of the statistical procedures used in these studies. Discriminant function analysis, trend analysis, and even cluster analysis do not adequately address the question of distinguishing types from continua (Meehl, 1995). Only taxometric statistical procedures can provide a more definitive answer to the question of whether a given disorder represents a qualitatively distinct class (i.e., a taxon) or an extreme endpoint on a continuously distributed dimension. The difference between the taxometric and the traditional approaches to studying classification questions involves the kinds of comparisons that are made between the two (or more) entities under consideration. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.