A description of behaviour that may indicate crossover from weight‐restored anorexia nervosa to bulimia nervosa
Introduction: The course and outcome of eating disorders can be characterised by the degree of diagnostic crossover. Crossover is relatively common, with the crossover from Anorexia Nervosa (AN) to Bulimia Nervosa (BN) being the most prevalent. Crossover commonly occurs within the first 5 years of illness and is often observed when patients are progressing to partial or full recovery. No information regarding crossover in South African persons with eating disorders has been published, hence the purpose of this study. Main objective: The main objective of the study was to describe the behaviour that may indicate crossover from weight‐restored AN to BN in South African young adults. In order to achieve the main objective, anthropometric measurements and descriptive information regarding disordered eating patterns were obtained. Information regarding behaviour that may be associated with crossover from AN to BN or within AN sub‐types was collected. In addition BN patients were assessed to determine whether they have a previous history of AN, which may further indicate crossover. Subjects and methods: Participants were recruited from the student population of the University of the Free State and Bloemcare Psychiatric Clinic. Anthropometric measurements were taken by the researcher and one of two questionnaires (compiled by the researcher), depending on diagnosis, was completed during a semi‐structured, one‐to‐one interview between the researcher and each participant. Questionnaires were coded by the researcher and analysed by the Department of Biostatistics (UFS). Results:: Nine participants were recruited and included in the study. Five out of the nine participants were diagnosed with Anorexia Nervosa Restrictive type (ANR). These five participants had all crossed over to bulimic tendencies during and after the process of weight restoration. One of the five participants has crossed over to a current diagnosis of Anorexia Nervosa Binging and Purging type (ANBP). The five participants indicated that they engaged in inappropriate compensatory behaviour after a binge episode in order to prevent further weight gain or to lose weight. The most common inappropriate compensatory behaviour reported was self‐induced vomiting. Two of the five participants indicated that they could currently be diagnosed with EDNOS because they had not completely recovered, whereas the other two participants indicated that they have fully recovered. The remaining four of the nine participants were diagnosed with BN. Two were currently diagnosed and the other two had previously been diagnosed with BN. Of the previously diagnosed BN participants, one participant had a history of ANR. The particular participant never fully recovered from the initial diagnosis and therefore crossed over from ANR to BN. The two previously diagnosed BN participants also indicated that they could be diagnosed with EDNOS at the time of the interview because they had not completely recovered. Overall the nine participants reported that they were still preoccupied with their weight at the time that the study was conducted. Seven of the nine participants indicated that they were more comfortable at a lower weight, whereas two participants indicated that they could not identify a weight at which they felt most comfortable. Conclusions: The course and outcome of eating disorders is partially determined by the occurrence of crossover. Comparable to reviewed literature, despite the small sample crossover was observed from AN to bulimic tendencies. In addition, crossover occured more commonly during the progression to partial or full recovery. With this in mind, further research should focus on whether crossover occurs as a result of the weight gain associated with recovery and whether the fear or anxiety thereof acts as a trigger. This knowledge may enable the multidiscliplinary health care team to prevent crossover from occurring in patients during the recovery period.