Although comprehensive theories of addiction recognize the etiological importance of environmental and cognitive factors, it has been widely accepted for many years that addiction is also a brain disease and that individuals differ in their susceptibility to this condition (Leshner, 1997; Wise and Bozarth, 1987). Explanations of the eating disorders have tended to eschew biological models in favor of those that focus on psychosocial and family influences-the most prominent models arising from psychoanalytic, feminist and cultural theory. It is not surprising, therefore, that although clear parallels exist between the abuse of substances and disturbances in eating, there has been a reluctance to accept that the two may share a common etiology. It is also probable that their similarities were obscured by dramatic differences in the social profile of the stereotypic drug addict and the patient with an eating disorder-the former typically associated with male criminality and social deviance and the latter with female submissiveness and social conformity.
In the past decade, however, there has been a growing paradigmatic shift in eating disorder research, with a movement away from explanations that rely solely on psychosocial factors, to a belief that disturbances in the function of brain neurotransmitter pathways are also highly relevant (Kaye, 1999). One outcome of this change in orientation has been an emerging and increasing interest in the links between eating disorders and substance abuse disorders.
Clinical and Biological Traits
It is generally agreed that the commencement of addictive behaviors can take two motivational routes: either the seeking of positive sensations or the self-medicating of painful affective states. While current research documents a substantial lifetime comorbidity between the eating disorders and other forms of addiction, there is less agreement on the reasons for this link (Holderness et al., 1994; Wiederman and Pryor, 1996). Some researchers have suggested that a common set of personality traits predispose an individual to a range of behaviors that have the potential to become excessive (Koob and Le Moal, 1997; Leshner, 1997). Support for this idea comes from evidence that anxiety and depression are frequent premorbid characteristics both of addicts (Grant and Harford, 1995; Kessler et al., 1997) and of patients with eating disorders (Deep et al., 1995; Vitousek and Manke, 1994). Our own research has also found that among eating-disordered patients, irrespective of diagnostic category, scores on a measure of addictive personality characteristics were comparable to those reported for drug addicts and alcoholics (Davis and Claridge, 1998). Complementary to this viewpoint, an addiction to one behavior reinforces a certain style of coping pattern that leaves the individual vulnerable to developing another type of addiction (Holderness et al., 1994).
Others have suggested that the eating disorders are, themselves, a form of drug addiction since their characteristics satisfy all the clinical and biological criteria for conventional addictions such as smoking, alcoholism and cocaine abuse (Davis and Claridge, 1998; Davis et al., 1999; Marrazzi and Luby, 1986). Foremost among these is the progressively compulsive nature of the behavior, even in the face of adverse consequences to health and safety (Heyman, 1996; Robinson and Berridge, 1993). Moreover, with continual exposure, individuals typically require more of the behavior to produce the same reinforcing effect (Berridge and Robinson, 1995). They also tend to experience an obsessively increasing craving for the behavior that can persist even after a long period of abstinence. Presumably that accounts, at least in part, for the fact that addicts have a strong tendency to resume the addictive behavior after treatment and for the chronic relapsing nature of addiction (Robinson and Berridge, 1993). These characteristics find direct parallels in the core eating-disorder behaviors such as dieting, over-exercising and binge eating, all of which tend to become increasingly excessive over time. Patients also report a strong compulsion to continue these behaviors despite serious medical complications, which is reflected in their prolonged morbidity and the high rate of relapse (Herzog et al., 1999; Strober et al., 1999).
At the biological level, similarities are also evident. We know, for instance, that strenuous exercise and starvation activate the dopaminergic (DA) reward pathway of the brain (Bergh and Sodersten, 1996; Casper, 1998). The resulting biological events underlie the auto-addiction opioid theory, which proposes that a chronic eating disorder is an addiction to the body’s production of endogenous opioids and therefore is identical to the physiology and psychology of substance abuse in general (Huebner, 1993; Marrazzi and Luby, 1986). In other words, starving, bingeing and exercise all serve as drug delivery devices since they increase circulating levels of -endorphins that are chemically identical to exogenous opiates, and these endorphins are as potentially addictive because of their ability to stimulate DA in the brain’s mesolimbic reward centers.
Via a different route, self-starving may have other biologically rewarding properties, albeit as a negative reinforcer. For example, in certain individuals, food restriction is reported to reduce anxiety. It has been suggested this might occur because of reduced serotonin activity in those with overactivity in this neurotransmitter system (Kaye, 1999).