Response: Anorexia nervosa,more than a body image disorder

In our recent work, we have proposed that the classification of anorexia nervosa (AN) as a ‘feeding and eating disorder’, as it is currently described in the Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5), is unsuitable (Phillipou et al., 2018). The criteria for AN and the gold standard measures for the disorder, such as the Eating Disorders Examination Questionnaire, emphasise disordered body image rather than disordered ‘feeding or eating’. This does not necessarily mean that AN is only a disorder of body image, but emphasises this is in fact what we are predominantly measuring with our current instruments.

Kaufmann and Russell (2018) responded with scepticism regarding our proposed re-classification. The authors cite cultural differences in the presentation of apparent AN, such that some individuals present primarily with somatic symptoms and do not manifest ‘fat phobia’. However, our current measures are not likely to lead to a clear diagnosis of AN based on somatic symptoms. This suggests either that our measures are not capturing the different forms that AN can take or that individuals who present with predominantly somatic symptoms without significant body image disturbance represent a different condition altogether. Kaufmann and Russell (2018) also argue that the re-classification of AN as a body image disorder may result in the general public believing the condition is related to vanity. We agree that vanity is not a feature of AN, and indeed that low self-esteem is more characteristic of the illness. Yet, we believe that educating the public on the idea that AN is a very serious disorder of body image perception, with neurobiological underpinnings, could help reduce the stigma associated with the illness. Arguably, its current conceptualisation as a ‘feeding or eating’ disorder could lead to a view that sufferers should simply ‘get over it and eat properly’ and that it is under their control.

Kaufmann and Russell (2018) also argue that any disturbance in body image experienced in AN is almost certainly a secondary phenomenon. We disagree with this statement for a number of reasons. First, as already described, body image disturbance is a key criterion for AN, whereas the subsequent excessive weight loss experienced in pursuit of an idealised body weight may be more appropriately considered secondary. Second, research suggests perceptual differences in estimating one’s body size in individuals with AN (Cornelissen et al., 2017; Phillipou et al., 2015). Furthermore, the neurobiological contributions to AN suggest a fundamental deficit in brain regions involved in body image perception (see Phillipou et al., 2014 for a review). Thus, we argue that those diagnosed with AN according to our current diagnostic criteria represent individuals with an illness driven by anxiety related to disturbed body image, not an illness of ‘feeding or eating’. Of course, more research is needed to gain a better understanding of the core features of AN, such as body image, rather than the secondary features of disturbed eating, and there indeed may be other features of AN that are important. It is, however, important to research the underlying mechanisms contributing to AN and to use this knowledge to help develop more effective treatments.

Kaufmann and Russell (2018) accurately point out that AN is a complex and multifaceted illness, whereby a number of features such as perfectionistic tendencies and obsessive-compulsive behaviours contribute to increased risk, in addition to a genetic predisposition. They also emphasise the importance of addressing negative emotions and improving cognitive flexibility in AN during treatment. It is of course necessary to address these broader issues of AN, which are important maintaining factors of the illness. However, as we have suggested, the disturbance of eating in AN must be addressed in conjunction with what is driving the disordered eating: distorted perception of self and one’s body size. Our current treatments focus on re-feeding, and issues around food and eating for this ‘eating disorder’, with often very little focus on body image. With increasing treatment advances in recent times, AN still has a mortality rate of 1 in 10, and a long-term recovery rate of under 50%. These statistics suggest that we are doing something wrong. If we are to improve the outcome of AN, we need to focus on what our patients are telling us, ‘I think I’m fat’.

See Commentary by Kaufmann and Russell 52: 286–287.