History of abuse is not predictive of eating disorders with binge-eating episodes (but posttraumatic stress disorder is).

In the article by Cabelguen and col­ leagues1, it is not surprising that a hist­ ory of abuse was not associated with recovery at 1 year, although the pres­ ence of a DSM­IV anxiety disorder was negatively associated with recovery. There are several notable points in ref­ erence to this paper. Traumatic events are ubiquitous in the general population, as well as in study populations of people with eat­ ing disorders. It is the experience and subsequent effects of traumatic events, not the events alone, that constitute the true meaning of trauma (the 3 Es).2 The authors of the related research looked only at the binary variable of whether there was any disclosure of trauma or not, regardless of type, and did not look at the sum of the different types of traumas endorsed. This is de­ spite their citation of the paper by Guillame and colleagues,3 which em­ phasized the association between addi­ tive trauma dose and severity of eating disorder. This has also been found in many other studies, including a major meta­analysis.4 Although the authors reported a rate of sexual and physical abuse of 35 %, they did not inquire about other potential traumas that may not only result in posttraumatic stress disorder (PTSD), but have also been associated with eating disorders.5 Nationally rep­ resentative data in France — derived from the World Health Organization’s World Mental Health Survey — showed that the rate of overall trauma exposure was 73 %, which is in line with studies in other countries.6 It is also notable that the authors’ numbers may be underestimates, given the low rate of study participa­ tion. Of 981 patients in the Evaluation of Behavioural Addictions eating dis­ orders cohort, fewer than half were eli­ gible for the study, and more than half (n = 219) of these dropped out. For many reasons, it may be that trauma and perhaps PTSD history figured prominently in why these patients dropped out, given that patients with both eating disorders and PTSD have greater severity of eating disorder, state–trait anxiety and depression, and poorer quality­of­life symptoms.7 Low rates of disclosure of sexual assault in France have also been described.6 Another limitation is the use of DSM­ IV criteria for anxiety disorders, which include PTSD and obsessive–compulsive disorder (OCD; both now in separate DSM­5 categories). Unfortunately, the authors do not specify exact ly how many patients met criteria for PTSD and OCD. Using validated assessment instruments, PTSD is known to occur in substantial proportions of patients with eating disor­ ders treated in higher levels of care, al­ though it is not clear how many patients in this study received inpatient or resi­ dential care. In addition, diagnosis of PTSD with the Mini International Neuro­ psychiatric Interview is limited, given that it diagnoses only current PTSD, and if a criterion A trauma is undisclosed, then other cluster questions are skipped. Other investigators have reported that a history of childhood abuse predicts more frequent binge eating at follow‐up, but only among those with lifetime PTSD.8 Furthermore, PTSD may serve as a mediator be­ tween trauma and eating disorder symptoms.9,10 Although the presence of PTSD does not preclude recovery, those with PTSD do not fare as well in treatment or upon follow­up, com­ pared with those without PTSD.7,11,12 Nevertheless, PTSD can be treated effectively in an integrated manner using trauma­focused approaches, such as cognitive processing therapy and prolonged exposure.7,13–16 Lastly, lifetime PTSD and the total number of victimization experiences endorsed (rather than a single trau­ matic experience) have been linked to measures of disinhibition and impul­ sivity in a national sample of women with binge­type eating disorders.17 Timothy D. Brewerton, MD