We share Gutierrez & Carrera's frustration about the difficulty in treating adults with anorexia nervosa. However, we disagree with their interpretation of our findings, and several other points they make.

First, in our trial specialist supportive clinical management (SSCM) was not superior to our new treatment, the Maudsley Model of Anorexia Treatment for Adults (MANTRA). In fact, outcomes for both interventions were similar. Moreover, in thesubgroup of patients with lower initial body mass index (<17.5 kg/m² at the beginning of treatment) there was some suggestion that patients receiving MANTRA showed greater weight gain than those receiving SSCM, but this was not statistically significant (P = 0.15) as the study was not powered to detect subgroup differences. Second, the original New Zealand trial - where SSCM compared well against cognitive-behavioural treatment and interpersonal therapy - included many patients who had a relatively mild, less chronic form of anorexia. In this earlier trial, SSCM effects seemed to wane in the long term. ^{Reference Carter, Jordan, McIntosh, Luty, McKenzie and Frampton1}

Second, contrary to Gutierrez & Carrera's assertion, there is plenty of evidence that the personality features, neuropsychological profile (thinking style) and aspects of altered socioemotional processing found in anorexia are not just an epiphenomenon of malnutrition but have trait characteristics which are accentuated in the starved state. ^{Reference Treasure and Schmidt2}

Taken together these findings suggest a definite place for SSCM, especially in the treatment of less severe cases of anorexia. It may be that a more complex treatment such as MANTRA, which is trait-focused and where patients are taught skills that help them to tackle a range of maintaining factors, is more effective in more severe cases. Our trial was too small to tease this out. However, a larger study is now under way that should be able to answer this question. ^{Reference Treasure and Schmidt2}

To suggest an ‘either/or’ dichotomy between a treatment focus on self or starvation seems remarkably simplistic to us. In fact, if an exclusive focus on reducing starvation was the key curative step in treatment, in-patient refeeding for anorexia should be used much more often, as this reverses poor nutrition most quickly. Yet, in-patient treatment has significant problems: it is unacceptable to many patients and has high relapse rates.

In a large-scale international survey of patients with eating disorders and their families, there was strong agreement between these stakeholders that specialist expertise and personal qualities of staff, expert psychological interventions and nutritional assistance (advice and intervention) combined are the key components of effective treatments and services. ^{Reference Nishizono-Maher, Escobar-Koch, Ringwood, Banker, van Furth and Schmidt3}

Clearly, we are a long way away from having a cure for adults with anorexia. Given the very limited evidence base, there is still much to learn about what works for whom and at which stage of illness. The past few years have seen the burgeoning of neuroscience data related to anorexia nervosa, which opens the way to treatments targeted at dysfunctional neurocircuitry. ^{Reference Van den Eynde, Guillaume, Broadbent, Campbell and Schmidt4,Reference Lipsman, Woodside, Giacobbe, Hamani, Carter and Norwood5} Ultimately, we predict that significant improvements in treatment outcomes in adults with anorexia are only going to be achieved through adding such ‘targeted brain-directed’ adjuncts to talking therapies and nutritional intervention.