Banana Therapy: Case Study of Anorexia Nervosa and Obsessive-Compulsive Disorder
Fall 2005, Vol 4, Issue 4
Kenneth W. Littlefield, PsyD
Department of Psychological Services
Remuda Ranch Programs for Eating Disorders
There is a strong relationship between anxiety and eating disorders (ED) and their treatments have similarities. We examine the overlap specifically between anorexia nervosa and obsessive compulsive disorder by considering Lara, a 15 year-old girl with a history of treatment failures.
Lara admitted to Remuda at 74% of her medically expected weight. Like most adolescents entering treatment she was functionally a mandated patient, since her parents sent her to treatment against her will.
Sitting sideways in a chair and hugging her knees, she hung her hair in front of her face and avoided all eye contact. She was severely depressed. Conversations between us consisted of my talking, and her shrugging her shoulders in response. Lara refused all group and individual therapy. She refused meals and supplemental feedings. She even refused to look at food and pushed away water.
Lara also announced her atheism when she arrived for treatment. She did not want to hear about God or spirituality. She believed that life was meaningless, expressed no hope for her future and saw no reason to try recovery. She wanted to die. Her perfectionism and consequent sense of worthlessness seemed to penetrate to her bones. She indicated that she was not worth the labor to produce food that would be better utilized by someone else. Lara believed that there was nothing good about her and she was worthy only of punishment.
Not surprisingly, Lara soon needed regular trips to the hospital to be re-hydrated. Lara’s treatment team began asking if she needed a higher level of care where compulsory feedings could be administered.
Needing to help this patient progress, I employed the “door in the face” technique (Tusing & Dillard, 2000). In this technique, the therapist makes a large and unreasonable request of the patient, knowing it will probably be refused. The objective is to get the patient to agree to a smaller request that follows, because by comparison it will seem reasonable. The large request gets you the "door in the face" when you ask it, but the smaller request gets an “OK”.
So I offered Lara a minimal reward if she ate an entire meal—a large request with little incentive. Of course, she immediately refused this unreasonable recommendation. I then continued by offering increasingly larger rewards for decreasingly less consumption, until we reached a balance that she believed to be reasonable and agreed to do. Because Lara loved to play cards, we ended our first negotiation with Lara eating two baby carrots and my teaching her a new card game that she wanted to learn.
For dinner we settled on her drinking an eight-ounce glass of water and my taking her out to the horse corral. She fed a snack to a horse that would be assigned to her when she started eating what was recommended by her dietician. Lara loved horses, and this motivated her. Interestingly, as much as she loved the experience she washed herself thoroughly immediately after petting the horse. I soon learned the extent of Lara’s obsessions and compulsions.
After weeks of this approach, Lara still talked minimally, but she was allowing nasogastric tube feedings which supplied full nutrition, drinking the minimum required water, and eating selected fruits and vegetables. We were moving along.
I then introduced a variable reinforcement schedule rewarding Lara when she ate solid foods. Finding rewards she wanted and foods she believed she could at least try eating was the hardest part. I added the expectation that she talk during individual therapy.
Table 1.
Experience OCD Eating Disorder
Feared Stimulus Doorknob Food
Specific Fear Germs Calories/Fat Grams
Fear Category Contamination Contamination
Body Location Hands Stomach/thighs/etc.
Means of Avoidance Sterile Environment Restricting
Eliminating Consequences of Contact Washing Purging/Exercise
Feared Outcome Sickness Weight Gain/Fat
Ultimate Feared Outcome Dying Rejection by Others
For certain patients, EDs, particularly anorexia, can be effectively viewed as a hybrid or subset of OCD. EDs and OCD have high comorbidity rates. Many more ED patients have histories of obsessions and compulsions. Others have no obsessions or compulsions except those related to their ED, but they usually identify with OCD through general knowledge or popular figures such as Jack Nicholson in the movie As Good As It Gets (Brooks & Andrus, 1997). For some patients, previously existing OCD concerns have become redirected to food and weight concerns. Patients with OCD histories notice a clear decline in their previous obsessions and compulsions when ED-related obsessions and compulsions begin.
By definition, OCD is an anxiety disorder. Generally, obsessions result in anxiety and compulsions relieve the anxiety. Like Lara, patients with EDs often find that no matter what they are anxious about, their ED relieves their anxiety. In the long run the ED increases their anxiety, but the fact that it provides immediate relief reinforces ED behavior and increases the likelihood that the person will repeat the behavior. In essence, the ED behaviors may serve the same function of managing anxiety and are analogous to their stereotypic OCD counterparts. See Table 1, above.
Patients with OCD and ED are conditioned over time to react to the feared stimulus as though it were the ultimate feared outcome. Over time they may lose sight of the relationship of the fear to the feared outcome or ultimate outcome, if they were ever consciously aware of it. Consequently, becoming aware of the fact that their obsessions are based on a
greater fear often helps patients tolerate the obsessions.
More detailed information about the diagnosis and treatment of OCD is available on the internet at http://www.ocfoundation.org or in the following references: Kozak & Foa, 1997; March & Mulle, 1998; Steketee & White, 1990.
Lara was anxious. She likely exhibited a genetically/biologically-based proneness to anxiety. Superadded to her pre-existing anxiety about germs was anxiety about weight gain, a developmentally-appropriate fear for a 15 year-old girl in our culture. For both fears, she had a solution in her OCD and ED symptoms. Lara had already become familiar with her OCD at Remuda, speaking about her “OCDs” and recognizing that she was a captive of her anxiety. So I showed her the table above. Quickly she saw similarities between her OCD and ED that she had previously missed. She saw that the ED had a concrete function in her life—to reduce anxiety.
Treatment for OCD includes exposure therapy with response prevention (ERP). Similarly, treatment for certain EDs that resemble OCD can benefit by incorporating ERP. ERP involves exposure to the fear without giving in to it or engaging in compulsive behavior. Over time the fear subsides when the feared outcome is not realized. ERP is thus based on an extinction of behaviors by breaking the bonds formed by previous conditioning processes.
Lara was able to understand this, but she did not like it. I needed to enlist her cooperation. There are many resources available for information on how to educate and enlist children’s cooperation with ERP (e.g., Jablonsky, Moritz, & Geary, 2001; Niner & Swearingen, 2004). Cultivating hope had also been a goal written into Lara’s treatment plan. By watching and listening to others in Remuda’s Christian atmosphere, she was beginning to understand the fundamental Judeo-Christian principles of grace, forgiveness, and intrinsic worth. She was experiencing a restoration of hope and self-esteem and an incipient belief in a higher power offering assistance for her recovery. With her new-found hope, she cooperated and agreed to try ERP. She wanted to get well.
Once she agreed to ERP, we created a hierarchy of fears that paralleled each other. One for the OCD and one for the ED. I repeated that we would challenge only fears that she believed she could resist. Giving in to fears and engaging in the compulsion or avoiding the stimulus reinforce the behavior and increase the conditioned fear. We would not challenge any fear that she knew she would end up giving into. Note that we look for the “sweet spot”—between the fears that are too powerful to resist and those that can be easily resisted. That is the area where we concentrate our efforts.
We never challenge fears of things that are actually perilous. ERP only addresses exaggerated fears of objects or situations. The exposure task places patients in such situations to raise their level of anxiety. Patients endure the anxiety and the conditioned responses are extinguished over time when the fears are not realized.
Lara had an exaggerated fear of weight gain from a variety of foods. Eventually she disclosed that she had an irrational fear that she would gain weight from the smell of food. She also believed that she could absorb calories through her skin by touching food.
We created a hierarchy list of fears. Lara first listed examples of foods that caused various degrees of anxiety. Then we ranked what sensations were most aversive for the foods. For example, tasting them was the scariest, followed by smell, then touch and sight. Hearing the foods, such as bacon frying, or simply hearing about food in conversation raised no anxiety by Lara’s report.
Lara suffered from several other OCD fears, involving counting, numbers, perfectionism, and superstitions. This complicated the application of ERP. Normally patients are asked to rate their subjective anxiety throughout the exposure task on a traditional 1 to 10 scale. This helps the therapist judge when to change the task by adjusting the intensity or bringing the session to an end. This was not possible with Lara: she would always rate her anxiety as a “six” because of her preoccupation with this number. Even if she could overcome the obsession with the number six, she would avoid assigning another number because she feared that doing so would permanently lock her anxiety at that level.
My eventual solution was to draw a picture of a cylinder on a white sheet of paper. The cylinder was exactly ten centimeters tall. I asked Lara to choose a colored marker and fill the cylinder with ink to represent her rising anxiety. She was able to manage this style of rating without interference from her obsessions and compulsions about numbers. Following our sessions, I would take a ruler and measure the height of her markings to document objectively her anxiety on the traditional 1 to 10 scale. This allowed me to track Lara’s anxiety and progress.
The task we agreed to begin working on with Lara was exposure to a food relatively low on her fear hierarchy. We wanted a food that would raise her anxiety based on smell and touch first, since eating it would be too anxiety provoking. It was unnecessary to have her eat it because she was already receiving sufficient caloric intake through Remuda’s nutrition schedule and because contact with the food through the smell and touch would effectively raise her anxiety to the target range.
Finally, after much educating, negotiating, and contemplating, we began what Lara eventually dubbed “Banana Therapy.” Two or three times each week we blocked two to three hours in our day to engage in ERP. The times were open-ended because the task needs to continue until there is a significant reduction in anxiety. ERP should not be ended prior to the reduction of anxiety.
Lara was initially petrified that she was gaining weight from the smell of the bananas. Later in the course of therapy she no longer worried about the olfactory transmission of calories, but she still feared weight gain by touching the bananas. This too waned by the end of treatment, such that one day she actually ate the bananas in session in lieu of her normally prescribed snack. We thus progressed from seeing to smelling to touching to eating the bananas. We used ERP with increasingly anxiety-provoking sensations instead of the traditional method of progressing to increasingly anxiety-provoking foods. To work on Lara’s smell and touch fears, we engaged in activities such as smashing bananas into paste and having Lara smear the paste onto her hands. Activities included much food play.
“Banana Therapy” lasted for two months. Meanwhile, the normal modalities of therapy at Remuda continued.
Lara learned many of the principles of ERP through our sessions. She was able to generalize those skills to other obsessions and compulsions that were not related to specific food fears. For instance, her concerns about germs diminished and her focus on numbers no longer impaired her functioning. She left treatment after reaching a reasonable weight through a combination of normal eating for maintenance calories and nocturnal tube feedings for weight gain calories.
Lara also developed a good sense of humor about ERP as her anxiety decreased and she began to feel powerful over her OCD and ED. Her developing sense of humor prompted me to arrive for our final session wearing a cowboy hat, boots, and “six-shooter” holsters slung across my hips, each holster armed with a banana. Drawing the bananas gunslinger-style, I evoked uproarious laughter from Lara. This was quite a change for a girl who made no eye contact and spoke no words during the first week of treatment.
References
Brooks, J.L. (Producer/Writer/Director) & Andrus, M. (Writer). (1997). As good as it gets. [Motion picture].
Jablonsky, J., Moritz, E.K., & Geary, R. (2001). Blink, blink, clop, clop. Plainview, NY: Childswork/Childsplay.
Kozak, M.J. & Foa, E.B. (1997). Mastery of obsessive-compulsive disorder: A cognitive-behavioral approach. Oxford, England: Oxford University Press.
March, J.S. & Mulle, K. (1998). OCD in children and adolescents: A cognitive-behavioral treatment manual. New York, NY: Guilford Press.
Niner, H.L. & Swearingen, G. (2004). Mr. Worry: A story about OCD. Morton Grove, IL: Albert Whitman & Company.
Steketee, G. & White, K. (1990). When Once is not enough: Help for obsessive compulsives. Oakland, CA: New Harbinger Publications.
Tusing, K.J. & Dillard, J.P. (2000). The psychological reality of the door-in-the-face: It's helping, not bargaining. Journal of Language and Social Psychology, 19, 5-25.
