Bulimia nervosa appears to be a 'modern' disorder, reaching psychiatric awareness in the 1970s. It has been treated as a 'depressive equivalent' with antidepressant medication and therapies modified from antidepressant strategies. High-dose antidepressant medication (such as fluoxetine 60 mg daily) offers proven but often short-lived anti-bulimic benefit (Walsh et al, 1991) and benefits are inferior to the best psychological therapies (Agras et al., 1992). Extensive evidence supports cognitive behaviour therapy (CBT), in individual or group format, as the first-line treatment, with interpersonal therapy (IPT) a close second. Disappointingly, most individuals treated for BN do not receive evidence-based treatments (Crow et al., 1999). This may partially explain the finding (Ben-Tovim etal., 2001) that five-year outcome for BN (and other eating disorders) was independent of receiving treatment.
Whatever the therapy, lower BMI, greater frequency of binge-purge episodes, (Fairburn et al., 1995) and poor self-esteem (Fairburn et al., 1997) are associated with poorer prognosis. Those with co-morbid personality disorders, particularly the 'multi-impulsive' picture (Fichter, Quadlif & Rief, 1994) are hardest of all to treat successfully. Surprisingly, longer duration of symptoms is associated with better outcome (Turnbull et al., 1996). In practice bulimia is a secret disorder with an average delay of six years from onset to presentation. Early intervention is rarely an option. This long delay also accounts for the belief that bulimia has a later onset than anorexia. Adolescents who are brought to help early may be harder to engage. At present little research addresses bulimia in the younger age group. Subjects in research trials generally meet DSM or ICD criteria:
• by definition, subjects with BN are at least normal weight range, body mass index (BMI) 19 or over;
• the core symptom is of binges -rapid consumption of a large amount of food, substantially more than a normal meal, in a finite amount of time (not 'grazing');
• binges are followed by purging behaviour, such as vomiting, abuse of laxatives or diuretics and by renewed attempts to fast or restrain eating;
• there have been at least three binge-purge episodes a week for at least six months;
• abnormally high value is placed on body image and slimness.
Cognitive Behaviour Therapy (CBT-BN): The Evidence Base
In 1979, the same year that Russell alerted the world to BN as 'an ominous variant of anorexia nervosa', Beck and colleagues published the classic Cognitive Therapy of Depression. The convenient marriage between the worrying young disorder and hopeful new therapy has spawned more than 50 trials. Cognitive behaviour therapy remains the leading treatment for normal weight BN. About 50 % of subjects improve substantially, with an overall 50 % decline in bingeing and purging, for a disorder that shows minimal placebo response and continues chronically without treatment (Keller et al., 1992). Cognitive behaviour therapy works faster than IPT and outcome can be predicted as early as the sixth session, when those who show no improvement are unlikely to respond.
Most trials use the modification of CBT manualised as 'CBT-BN' by Fairburn (1993) or similar adaptations, such as that of Cooper (1995). Manuals can be administered by minimally trained professionals or offered in workbook form as self-help. The model appears robust with regard to setting and therapist characteristics. Treatment is effective in individual or group format and even in self-help, or guided self-help format (Cooper, 1996). There have recently been pilot studies of telephone delivery (Hugo, 1999; Palmer et al., 2002) and Internet delivery (Robinson, 2001).
Benefit from self-help is strongly correlated with outcome in formal CBT. Self-help approaches rely greatly on patients' capacity for self-monitoring and the mainstay of CBT-BN is the 'food diary', listing food intake, binge-purge episodes and associated feelings and thoughts. Therapists need tact and creativity to help patients who are disorganised or bored by the task, or show the common reluctance to detail embarrassing behaviour.
After the professional's usual assessment, patients complete baseline ratings such as the BITE (Henderson & Freeman, 1987) and measures of depression. These facilitate audit and research and are discussed collaboratively with the patient to monitor progress. Patients are encouraged to let the therapist take over weekly weighing. Full physical examination is rarely needed but blood tests may highlight low potassium, anaemia and (in drinkers) abnormal liver function.
Patients are offered CBT-BN as a course of about 20 one-hour sessions. Fairburn (1985) suggests two or three sessions a week early in treatment with later sessions spaced fortnightly. Mitchell et al. (1993) found high intensity approaches more effective. There are three main stages:
• a psychoeducational and monitoring phase, introducing regular eating;
• a more cognitive phase, teaching strategies to eliminate binge-purging and challenge obstacles to normal eating behaviour;
• final sessions to address relapse management, as this is common.
Belinda, a 22-year-old music student, had suffered from bulimia nervosa for six years. She binged and vomited most evenings, and took laxatives. She entered a research trial and was randomised to receive 19 weekly sessions of CBT.
Belinda was given reading material and food diaries to complete. These were reviewed as part of the agenda of each session. She agreed to the therapist weighing her weekly instead of weighing herself several times a day. She borrowed books that discussed the health implications of bulimia and helped her to distinguish scientifically investigated information from her own assumptions about 'fattening foods'. She scheduled three mealtimes and three snacks each day and practised stimulus control techniques to structure these. At first distraction strategies failed to reduce binge-purge episodes. However, she stopped taking laxatives when she learned they did not get rid of any calories.
By session eight, a planned review, Belinda's BITE score was lower, particularly on the severity subscale. Her depression inventory showed improvement too. She told the therapist she could now eat socially with friends.
Therapy next focused on eliminating eating disordered behaviour and reintroducing 'forbidden' foods. Belinda used her diaries to monitor the automatic thoughts that occurred when she tried to eat normally or resist binge-purging. These included 'normal meals will make me fat' and 'I must fast today'. She learned techniques to challenge these thoughts and to generate and test out more helpful alternatives, such as 'eating regularly helps keep hunger under control'.
The last three sessions were spaced fortnightly. Belinda feared she would slip up and go 'back to square one'. The therapist reminded her that lapses were common but need not be catastrophic. They prepared a written relapse management plan and arranged three- and six-month follow-up appointments. Belinda now binged less than once a month and ate a range of foods at normal mealtimes. She was spending time with a new boyfriend and their music group. Progress continued at six-month follow up.
Interpersonal Psychotherapy (IPT-BN): The Evidence Base
Klerman et al. (1984) drew on key skills of experienced therapists to design a set of techniques as a condition for inclusion in clinical trials of the management of depression. Their package developed into a versatile therapy in its own right. It was particularly attractive to Fairburn's group, as a structured, manualised, time-limited therapy against which to test out the specific effectiveness of CBT. At the end of active therapy CBT is consistently ahead of both behaviour therapy and of IPT. Remarkably, though, at one-year follow-up, IPT subjects catch up with the CBT cohort on all measures of functioning, including binge-purges (Fairburn et al., 1995). Interpersonal psychotherapy when available, is a close second to CBT. Unfortunately there are no indicators of differential suitability.
The Fairburn model of IPT for bulimia nervosa differs in spirit from the original IPT model in that it does not 'prescribe the sick role' to the patient and does not begin each session with a review of the symptoms of bulimia. This was to avoid overlap with CBT in research trials. Including review of bulimic symptoms has not been evaluated. Interpersonal psychotherapy is a three-part course of therapy.
• During the assessment phase (four sessions) the therapist formulates the onset and progression of the patient's eating disorder in relation to interpersonal events.
• The longest part of therapy is devoted to the patient's active work on the selected focus area - interpersonal role transition, role dispute, grief (interpersonal loss) or interpersonal deficits. No homework is formally set, but the patient improvises and practises new interpersonal skills between sessions.
• The final few sessions specifically address termination issues.
Suppose that Belinda (described above) was allocated to the IPT arm of the research study:
The therapist spent the early sessions mapping out the important people in Belinda's interpersonal network. He charted events, changes in her mood, and the onset and deterioration of her eating disorder, in a series of columns on a 'life chart'.
Belinda's father had died eight years earlier and she became withdrawn, starving herself. Two years later, having lost so much weight that her periods stopped, she found exam revision interrupted by food binges. Her weight increased so she began to vomit. She took a year off and things improved. Her weight was normal, which reassured her mother. When she finally left home, though, Belinda's bingeing increased and she started to use laxatives. She made no friends at university, but spent weekends travelling home to her widowed mother.
The therapist acknowledged the importance of grief in the onset of her symptoms and also a role dispute with her mother, which perpetuated the disorder. However, he proposed that they devote the middle part of therapy to the focus of the role transition from home to university life. (The other focus is interpersonal deficits, but Belinda's isolation was attributable to her move rather than to longstanding difficulties.)
Belinda now worked hard to build a new network of friends and deal with personality clashes, rejections and overwhelming demands. The therapist used word for word 'replays' of incidents and role-play to examine and prepare for interpersonal problems. When Belinda spoke of bingeing or dieting, the therapist would refocus on interpersonal concerns.
Late sessions were explicitly devoted to termination issues. Belinda recalled the guilt and homesickness she no longer felt. Her mother seemed more cheerful, although Belinda spent fewer weekends at home. Her bingeing and vomiting were somewhat reduced. She was busy with friends and music.
At one year follow-up, symptoms continued to improve with less than monthly binge-purge episodes. She was playing in several bands and had a steady boyfriend. Her mother had remarried.
Some centres practise a sequenced approach to therapy. At first referral patients might be offered guided self-help. Those who fail to improve would enter individual therapy using an evidence-based model. Fifty per cent of patients are not helped by CBT or IPT. If these were unsuccessful, different models of therapy might then be offered. Day-patient programmes or admission are a last resort.
In practice, few patients comply strenuously with further therapeutic efforts after 'failing,' unless there is a built-in expectation that therapy will occur in a series of blocks addressing different skills. Nevonen et al. (1999) have designed a group treatment for BN starting with CBT, then proceeding to group IPT. Preliminary evaluations are encouraging.
Other options, particularly where there is co-morbidity, include cognitive analytic therapy (CAT - Denman, 1995), guided imagery (Esplen et al., 1998) and dialectical behavioural therapy (DBT) (Linehan, 1993; Safer et al., 2000).
Was this article helpful?
I already know two things about you. You are an intelligent person who has a weighty problem. I know that you are intelligent because you are seeking help to solve your problem and that is always the second step to solving a problem. The first one is acknowledging that there is, in fact, a problem that needs to be solved.