Anorexia Nervosa: Predicting If Patients Will Accept Treatment

Anorexia nervosa is still one of the most difficult psychiatric disorders to treat, as evidenced by the fact that despite many advances, treatment is about as successful today as it was 50 years ago.

According to Dr. Katherine Halmi and colleagues (Arch Gen Psychiatry 2005; 62:776), few controlled studies of the course of treatment have been done because of the nature of AN itself. That is, the disorder is relatively rare, so it is difficult to get large enough numbers of cases at any one center. In addition, patients with AN resist treatment and are at best ambivalent about it. Finally, medical complications often lead to withdrawal from treatment protocols.

Dr. Halmi, at Cornell University, and colleagues at Stanford University and the University of Minnesota, designed a randomized prospective study of 122 patients from 14 to 50 years of age with DSM-IV criteria for AN to evaluate the factors that can lead to nonacceptance and noncompletion of treatment. The patients were randomly assigned to one of three treatments: cognitive behavioral therapy (CBT), fluoxetine, or a combination of the two, for one year.

CBT sessions were scheduled twice weekly for the first month, once weekly during months 2-6, twice monthly for months 7-9, and once a month for months 10 to 12. The CBT was a manual-based intervention developed specially for this protocol. Fifteen-minute medical management sessions were conducted by a psychiatrist after the therapy session weekly for the first month. These sessions were then scheduled twice a month through the fourth month and then monthly.

Patients in the fluoxetine treatment goup received the medication during the medical management session. Within 6 weeks, all these patients were receiving the maximum dose, 60 mg/day. After 6 weeks, they continued taking the maximum dosage they could tolerate.

About a third of the participants had previously been hospitalized and two-thirds had previous treatment as outpatients. More than half engaged in purging. The mean body mass index (BMI) was 17.8 kg/m2.

Who dropped out or withdrew?

The dropout and withdrawal rates were substantial, according to the authors. Twenty-one participants, or 17%, were withdrawn from the study, mainly due to treatment failure. Other reasons included pregnancy (2 patients) and intolerable side effects of the medication (3 patients).

Among the 101 remaining patients, 56, or 55%, dropped out. The main reason given was dissatisfaction with some aspect of the treatment (68%). An additional 11 subjects, or 20%, dropped out very early in treatment; 4 were withdrawn by their families, and 2 (4%) dropped out later in treatment—all without giving a reason.

The strongest predictor of treatment acceptance was the type of treatment. Eighty-nine of the original 122 study participants were treatment acceptors. Of these, 45 or 51% completed treatment. The only predictor of treatment completion was self-esteem. A low score on the Rosenberg Self-Esteem Scale led to a 40% completion rate, whereas a high self-esteem score was associated with an 80% treatment completion rate. The type of treatment was not a significant predictor of treatment completion among those who accepted treatment.

The very low rate of acceptance in the medication treatment group does not seem to pertain when medication is given together with some form of psychotherapy. As Dr. Halmi noted, this suggests that medication alone cannot be an effective treatment for anorexia nervosa if only because most anorectic patients will not accept such treatment.

Overcoming high dropout rates

Is there any way to overcome these high dropout rates? The authors note that it became clear that medication alone is not a viable treatment for such patients. At one of the three study sites, for example, 165 potential subjects were screened, 16 were randomized to medication and only 1 completed that treatment. Also, one major reason for dropping out of the treatment group was medical difficulties. Therefore, it is necessary to develop new protocols that include dealing with such difficulties without dropping patients from the protocol to which they have been assigned.

Finally, according to the authors, in the groups that received psychotherapy, patients with high obsessive preoccupation tended to have relatively high acceptance rates for treatment involving psychotherapy. Those with high self-esteem were more likely to complete treatment than those with low self-esteem.

It might be possible that devising different treatment protocols for patients with AN that take into consideration such baseline characteristics might alleviate the problem of having a patient initially agree to treatment and then drop out. According to the authors, remedies need to be identified that will improve acceptance of treatment and reduce dropout among patients with AN who have low obsessive preoccupation and low self-esteem.