The Therapeutic Armamentarium in Eating Disorders
James E. Mitchell, Scott Crow, Tricia Cook Myers and Steve Wonderlich
In this paper we will briefly summarize the available literature on the empirically tested treatments for patients suffering from eating disorders. We will address the traditional eating disorders of anorexia nervosa and bulimia nervosa, and will also include binge-eating disorder, a condition included as an example of 'eating disorders—not otherwise specified' in the DSM-IV as a disorder for further study. We focus on treatments that have been shown to be effective in randomized trials, but include other clinical information when it appears relevant and necessary to the reader's understanding of that particular area.
In perusing this text, many readers will notice several trends that typify this literature. First and of greatest clinical concern, the reader will notice the relative paucity of literature on the treatment of patients with anorexia nervosa, despite the fact that anorexia nervosa was the first identified eating disorder and clearly is the most severe, with a significant well-documented risk for morbidity and mortality. There are a number of factors that have contributed to the relative lack of research in this area, including the following: (1) anorexia nervosa is a relatively rare condition and it is difficult for individual treatment centres to acquire the necessary number of subjects to complete randomized trials; (2) patients with anorexia nervosa, by virtue of their illness, many times are not particularly motivated to be cooperative with treatment and therefore it is difficult for them to be compliant with research treatment protocols; (3) when initially seen many patients with anorexia nervosa are critically ill, and require a multiplicity of interventions (e.g., medical stabilization, occasionally hospitalization, family involvement, individual counselling, medication management) which markedly complicates the ability of researchers to design clinical trials which adequately control for all of these variables; (4) anorexia nervosa patients are often quite difficult to treat, again given the nature of their illness, and this undoubtedly dissuades many potential investigators from pursuing work in this area. All of these factors have contributed to what currently is an apparent and quite worrisome lack of knowledge regarding the best treatments for patients with this disorder.
Second, readers of this text will probably also notice that although the treatment literature on bulimia nervosa is better developed, this entire literature has been published in the last 20 years. This is attributable to the fact that bulimia nervosa was only first identified as a discrete diagnostic entity in 1979 (Russell, 1979) and randomized clinical trials were first implemented several years after that.
Third, readers may notice that the literature on binge-eating disorder is also quite limited. This again is attributable to the fact that binge-eating disorder was only described in its current form in the DSM-IV, which was published in 1994. At that time there were little data regarding this group of patients, and our current understanding of this disorder has developed since then.
Fourth, an observation that may strike some readers as surprising is the finding that psychotherapy plays a clearly important and, in some cases, central role in the treatment of patients with eating disorders. The reader is therefore reminded that psychological interventions that result in psychological changes often result in corollary changes in the underlying biology, and therefore discussions of such therapies in a text on biological psychiatry is quite appropriate.
We turn now to the treatment literature on these three conditions.
The pharmacotherapy of bulimia nervosa can be conveniently divided into three areas: (1) antidepressant trials, which have employed a variety of types of agents using various experimental designs ranging from acute treatment studies to relapse prevention studies; (2) studies examining the utility of non-antidepressant pharmacological approaches —none of which have been definitive, but several of which are interesting and show promise; and (3) a handful of studies regarding the relative efficacy of pharmacotherapy and psychotherapy for patients with bulimia nervosa.
Early in the course of our understanding of bulimia nervosa, research groups studying this condition noted mat many of the patients were depressed (Russell, 1979). Based on this observation, it was hypothesized that patients might be better able to control their bulimic symptoms if their depression were treated. Some researchers even went so far as to hypothesize that bulimia nervosa might be a variant of affective disorders, although few would endorse such a model currently (Pope et at., 1983).
Given this background, a number of antidepressants were tried. Following the initial observation that tricyclic and MAO inhibitors both seemed beneficial for these patients, a series of randomized treatment trials were undertaken. These studies are summarized in Table XXIII-10.l, and as can be seen the list has grown rather lengthy. In examining these studies, several issues emerge. First, the number of compounds studied and the classes of compounds studied (e.g., MAOIs, tricyclics, SSRIs) is large but not exhaustive. Second, the sample size in most of the studies has been modest, me exceptions being the large multicentre studies involving fluoxetine that were funded by Eli Lilly, who subsequently sought and received FDA approval to market fluoxetine for bulimia nervosa in me US. Third, although not illustrated, there is a great deal of variability in response rates to placebo. This probably speaks to a number of issues, one of which is the lack of standardization of the protocols for administering agents in this population, a problem we have discussed previously in the literature (Mitchell et al., 2000). Fourth, the reductions in the frequencies of target eating behaviours