The aim of the current study was to examine the relationship between each subtype of EDs and ASD trait regarding the presence or absence of self-induced vomiting. We first compared the illness duration, BMI, EDE-Q scores, and AQ scores in the four groups (BN, AN-BP, AN-R, and BED). Next, the four subtypes were divided as per the presence or absence of self-induced vomiting; the illness duration, BMI, age, EDE-Q, scores and AQ scores were compared. Assuming that BMI is a confounding factor, we compared EDE-Q and AQ scores between anorexic group (AN-BP and AN-R) and bulimic group (BN and BED). Furthermore, illness duration could be a confounding factor; therefore, we performed additional analysis to confirm it. Finally, we examined the ratio of the subtypes and the number of patients who exceeded the cut-off value of the AQ score.
We had hypothesised that patients without self-induced vomiting had a higher tendency for ASD; however, there was no difference in the AQ between AN-BP and AN-R. The significant difference in the AQ score between BN and BED was clear. Of the four ED subtypes, BED had the highest ASD trait. Even after the statistical analyses were adjusted to determine confounding factors, such as low BMI and duration of illness, the significant AQ differences in our results can be explained by the difference between BN and BED. However, this finding is not completely aligned with our hypothesis that ASD traits are correlated with a lack of vomiting. That is, for EDs of patients within or above normal weight, ASD traits were associated with a lack of vomiting in this study. In addition, when the illness duration was divided into <4 years and ≥4 years, non-vomiting was common in the <4 year group and vomiting was common in the ≥4 year group. This result supports the rationale for a 4 year cut-off to designate the chronic and acute phases. Notably, early intervention in ED (before the transition from onset to vomiting) may be necessary.
In addition, the AQ communication score was significantly higher in the group without vomiting and with <4 years of illness duration than in the group with vomiting ≥ 4 years of illness duration. Second, the non-vomiting group ≥4 years of illness duration scored higher than the vomiting groups. This result suggested that the non-vomiting groups tended to have a higher AQ communication score. A previous study had reported that all-or-none thinking about food and dieting was typical of patients with BED . Some patients say that they ‘would never want to vomit’ and are tolerant of weight gain even if they start overeating. In these BED patients, the absence of self-induced vomiting means that they tend to be obese , and it is unclear why they do not vomit in the face of weight gain . If considering the characteristics of ASD, it may be possible that some patients with BED do not vomit because another obsessive compulsion arising from ASD is stronger than the core psychopathology of the ED, i.e. fear of being fat and the failure of severe restriction leads to acceptance of weight gain. This is seen clinically in our practise with comments from patients such as ‘I am scared to vomit,’ ‘My life is over when I am vomiting,’ or ‘Looking at vomit disgusts me.’ In such instances, the fear or aversion to vomiting might be stronger than the desire not to gain weight. In addition, patients who were absorbed in dietary restrictions were able to postpone the desire to lose weight due to the disgust of vomiting along with the failure of restrictions, and to endure the weight gain somehow could not be explained by the psychopathology of EDs. In some cases, from the experience of vomiting once in the past, people have a strong sense to visceral sensations and/or disgust of vomiting; they are unable to forget the trauma and find it difficult to eat food because they do not want to vomit again. People with hyperaesthesia within the autism spectrum are reluctant to induce vomiting. In addition, some patients cannot eat because they are afraid of vomiting.
If there is an ASD trait, it is easy to fall into maladaptation to environment because of impairment of social skills, communication and lack of flexibility. In this study, social skills and communication scores of AQ in patients without self-induced vomiting were higher than those in patients with self-induced vomiting.
The maladaptation to environment may lead to routine behaviour, because certain routine behaviour, such as routine dietary patterns of patients, tends to reassure individuals. Environmental adjustment is the first requirement for such individuals.
For patients with EDs, it may be necessary to prioritise the identification of characteristics over the diagnosis or types. When assessing patients with EDs who never vomit, it is important to clarify the reason; i.e. determine what they are afraid of as a consequence of vomiting. Therefore, detailed assessments for appropriate recognition of each patient who does not vomit and the application of empirically derived treatments are required. Based on the above, for patients with BED with a high tendency towards ASD, environmental adjustment and psychoeducation regarding ASD may be necessary. In addition, patients should not be corrected in terms of their diet choices that are derived from sensory sensitivity (i.e. sense of smell, taste), and their feeling of disgust should not be ignored. It is also essential to take into consideration in their abnormal eating behaviour caused by the stress of their poor communication skills. For patients (AN-R and BED) who have not vomited and have illness duration of <4 years, it is important for therapists to identify these patients’ attitude towards self -induced vomiting for the evaluation of their ASD traits and to perform an early intervention before the condition becomes chronic. To the best of our knowledge, this is the first study to compare the predisposition for ASD by ED subtype and the presence or absence of self-induced vomiting.
This study had some limitations. Of note, the sample size was small, there were differences in the number of participants in each subtype. There were no data for healthy subjects to compare patient data with standard values. Besides the above, there was a large age spread. In general, it is known that AN-R is much more common in the younger ages, and self-induced vomiting usually appear later on in adolescence/young adulthood. The differences of subjects' age, BMI and illness duration may always be a confounder and should be controlled it with bigger sample size.
Symptoms of depression and anxiety disorders, irritability, emotional lability and obsessional features are frequent accompaniments in ED. Typically, these features worsen with weight loss and improve with weight regain . Interest in the outside world also declines as patients become underweight, with the result that most patients become socially withdrawn and isolated. Since we did not measure anxiety or depression in this study, it is unclear how these were associated with ED and AQ scores. Since ED has a high incidence of anxiety and depression, it should be added to the evaluation index in the future study. There are still problems to be examined in the future regarding coexistence of autism trait and measurement of individual differences.
This study used only the AQ for evaluation of ASD traits. We should have used the AQ-10 in addition to the AQ because a few adolescent patients were included in our study. These are good tools for assessing the presence or absence of ASD traits in a busy clinical setting. However, the AQ is a self-completed scale and is not used to diagnose ASD. Using the Autism Diagnostic Observation Schedule (ADOS) or the Autism Diagnostic Interview (ADI) for ASD evaluation, excluding patients with extremely low body weights, and including control subjects in a larger overall sample should be done in future research [9,21]. A future issue is to determine the clinical usefulness of using the ASD evaluation tool, which is called the gold standard such as ADOS and ADI, for ED patients whose ASD characteristics are considered to be a factor for maintaining the symptoms.