Eating disorders: recognition and treatment

Why this is important

There is little evidence on psychological treatments for people with binge eating disorder. The studies that have been published have not always provided remission outcomes or adequate definitions of remission. While there is some evidence for guided self-help and individual CBT‑ED, only 1 study was identified for individual CBT‑ED and no remission data were available. It is also unclear if individual CBT‑ED is more effective than guided self-help or group CBT‑ED (especially for people that find these treatments ineffective).

There is also no evidence on treatments for children and very little for young people. One study was found on individual CBT‑ED for young people, but only 26 participants were included in the data for remission. Randomised controlled trials should be carried out to compare the clinical and cost effectiveness of psychological treatments for adults, children and young people with binge eating disorder. In adults, the treatment should focus on the effectiveness of individual CBT‑ED compared with guided self-help and group CBT‑ED. For children and young people, the efficacy of eating disorder-focused family therapy could also be compared with individual CBT‑ED and different kinds of self-help (such as internet self-help or guided self-help). Primary outcome measures could include:

There should be at least a 1‑year follow up. Qualitative data could also be collected on the service user's and (if appropriate) their family members' or carers' experience of the treatment. Mediating and moderating factors that have an effect on treatment effectiveness should also be measured, so that treatment barriers can be addressed and positive factors can be promoted.

Why is this important

The psychological treatments currently recommended consist of a high number of sessions (typically between 20 and 40) delivered over a long period of time. Attending a high number of sessions is a major commitment for a person with an eating disorder and a large cost for services. People may be able to achieve remission with a smaller number of sessions or over a shorter period of time.

Randomised controlled trials of the psychological treatments recommended in this guideline should be carried out to compare whether a reduced number of sessions or a less intensive course is as effective as the recommended number. Primary outcome measures could include:

There should be at least a 1‑year follow up. Mediating and moderating factors that have an effect on treatment effectiveness should also be measured, so that treatment barriers can be addressed and positive factors can be promoted.

Why this is important

Medical conditions such as bradycardia, hypotension and hypothermia are common in people who are underweight because of an eating disorder. Key markers of medical instability due to underweight such as pulse rate, blood pressure, and degree of underweight are commonly used as indications of risk in people with eating disorders. A number of internationally used risk frameworks are based on these markers and are important in decision making for people with eating disorders (in particular when deciding whether to admit someone, whether to use compulsory care, and how to provide nutrition). The medical markers of acute risk are used throughout this guideline.

Despite their importance, almost all of the conventional risk frameworks are based on consensus with little validation. There is also a shortage of information on the physical factors most associated with mortality in eating disorders. Validated tools (such as Acute Physiology and Chronic Health Evaluation [APACHE] scores) are central to risk prediction in other areas of medical care, and it would be very useful to have a tool like this for eating disorders. Research is therefore needed to validate the range of individual clinical and biochemical markers, both individually and collectively, as predictors for physical harm (including death).

Why this is important

People with an eating disorder often have physical comorbidities (such as diabetes) or mental health comorbidities (such as substance abuse, self-harm or obsessive-compulsive disorder). However, there is little evidence on which treatments work best for people with an eating disorder and a comorbidity. A modified eating disorder therapy that addresses both conditions may avoid the need for different types of therapy (either in parallel or one after the other). Alternatively, a comorbidity may be severe enough that it needs addressing before treating the eating disorder, or treatment solely for the eating disorder may help with the comorbidity.

This is a complex area and likely to depend on the severity of the comorbidity and the eating disorder. There is limited evidence and randomised controlled trials are needed. For example, a trial could randomise people with an eating disorder and the same comorbidity (such as type 1 diabetes) to either a modified eating disorder therapy or a non-modified eating disorder therapy. Primary outcome measures may include:

There should be at least a 1‑year follow up. Mediating and moderating factors that have an effect on treatment effectiveness should also be measured, so that treatment barriers can be addressed and positive factors can be promoted.

Why this is important

There is a wide range of treatments available for anorexia nervosa. However, they are often ineffective, and even when they are successful there is a high risk of relapse. It is not clear which factors reduce the risk of relapse after successful treatment, or what benefit people receive from further treatment to prevent relapse. There is also little evidence on effective relapse prevention strategies for people in remission.

A series of studies should be done to identify the factors associated with an enduring response to treatment, and to test interventions specifically aimed at preventing relapse in people in remission. Primary outcome measures may include: