Differentiation between atypical anorexia nervosa and anorexia nervosa using machine learning.

OBJECTIVE: Body mass index (BMI) is the primary criterion differentiating anorexia nervosa (AN) and atypical anorexia nervosa despite prior literature indicating few differences between disorders. Machine learning (ML) classification provides us an efficient means of accurately distinguishing between two meaningful classes given any number of features. The aim of the present study was to determine if ML algorithms can accurately distinguish AN and atypical AN given an ensemble of features excluding BMI, and if not, if the inclusion of BMI enables ML to accurately classify between the two. METHODS: Using an aggregate sample from seven studies consisting of individuals with AN and atypical AN who completed baseline questionnaires (N = 448), we used logistic regression, decision tree, and random forest ML classification models each trained on two datasets, one containing demographic, eating disorder, and comorbid features without BMI, and one retaining all features and BMI. RESULTS: Model performance for all algorithms trained with BMI as a feature was deemed acceptable (mean accuracy = 74.98%, mean area under the receiving operating characteristics curve [AUC] = 74.75%), whereas model performance diminished without BMI (mean accuracy = 59.37%, mean AUC = 59.98%). DISCUSSION: Model performance was acceptable, but not strong, if BMI was included as a feature; no other features meaningfully improved classification. When BMI was excluded, ML algorithms performed poorly at classifying cases of AN and atypical AN when considering other demographic and clinical characteristics. Results suggest a reconceptualization of atypical AN should be considered. PUBLIC SIGNIFICANCE: There is a growing debate about the differences between anorexia nervosa and atypical anorexia nervosa as their diagnostic differentiation relies on BMI despite being similar otherwise. We aimed to see if machine learning could distinguish between the two disorders and found accurate classification only if BMI was used as a feature. This finding calls into question the need to differentiate between the two disorders.

Dynamic relationships among feeling fat, fear of weight gain, and eating disorder symptoms in an eating disorder sample.

Feeling fat and fear of weight gain are key cognitive-affective symptoms that are theorized to maintain eating disorders (EDs). Little research has examined the dynamic relationships among feeling fat, fear of weight gain, emotions, cognitions, and ED behaviors. Furthermore, it is unknown if these relations vary by ED diagnosis (e.g., anorexia nervosa (AN) vs other ED). The current study (N = 94 ED participants; AN n = 64) utilized ecological momentary assessments collected four times a day for 18 days (72 timepoints) asking about feeling fat, fear of weight gain, emotions (i.e., anxiety, guilt), cognitions (i.e., feelings of having overeaten, thoughts about dieting), and ED behaviors (i.e., vomiting, diuretic/laxative use, excessive exercise, body checking, self-weighing, binge-eating, restriction) at stressful timepoints (contemporaneous [mealtime], and prospective/temporal [next-meal]). Multilevel modeling was used to test for between and within-person associations. Higher feeling fat and fear of weight gain independently predicted higher next-meal emotions (i.e., anxiety, guilt), cognitions (i.e., feelings of having overeaten, thoughts about dieting, fear of weight gain, feeling fat), and ED behaviors (i.e., body checking, self-weighing [feeling fat]). There were relationships in the opposite direction, such that some emotions, cognitions, and ED behaviors prospectively predicted feeling fat and fear of weight gain, suggesting existence of a reciprocal cycle. Some differences were found via diagnosis. Findings pinpoint specific dynamic and cyclical relationships among feeling fat, fear of weight gain, and specific ED symptoms, and suggest the need for more research on how feeling fat, fear of weight gain and cognitive-affective-behavioral aspects of ED operate. Future research can test if treatment interventions targeted at feeling fat and fear of weight gain may disrupt these cycles.

Patient perspectives on mesh-related complications after hernia repair.

BACKGROUND: To explore the thoughts, feelings, and experiences of patients with mesh-related complications after hernia repair. The rate of long-term mesh-related complications requiring procedural intervention after abdominal core surgery, including hernia repair, is unknown. Determining this rate is challenging due to its anticipated low chance of occuring and historically poor systematic long-term follow-up in patients' hernia repair. The lived experience of these patients is also not well understood. METHODS: Purposive sampling was used to identify patients who have experienced mesh-related complications after hernia repair, and semistructured interviews were conducted. Descriptive thematic analysis was used to identify, analyze, and report common patterns across the data set related to the patient experience of mesh-related complications. RESULTS: Eight patients who had undergone a hernia repair with mesh and had at least 1 mesh-related complication after their repair requiring operation, an additional procedure, or medical treatment were included in the study and completed semistructured interviews over the phone. Five domains emerged from the interviews: indicators of mesh-related complications, knowledge of potential surgical complications, relationship/satisfcation with surgeon and/or surgical team, psychosocial impact of hernia repair and mesh-related complications, and function. CONCLUSION: Despite the widespread use of mesh in abdominal wall operations, little is known regarding the patient experience of mesh-related complications. The themes identified in the present study provide insight into the patient experience of mesh-related complications and can inform the future development of a patient-reported outcome measure to determine the true incidence of mesh-related complications and the impact of these complications on quality of life.

An eye-tracking study examining the relationship between males' eating disorder symptomatology, body mass index, and expectations about character behaviour in text.

Eating disorder prevalence is increasing in males, perhaps more rapidly than in females. Theorists have proposed that cognitive biases are important factors underpinning disordered eating, especially those related to food, body, and perfectionism. We investigated these factors in relation to males' eating disorder symptomatology in the general population by using eye-tracking during reading as a novel and implicit measure. 180 males' eye movements were monitored while they read scenarios (third-person in Experiment 1 (n = 90, 18-38(Mage = 21.50, SD = 3.65)); second-person in Experiment 2 (n = 90, 18-35(Mage = 20.50, SD = 2.22))) describing characters' emotional responses (e.g. upset) to food-, body image-, and perfectionism-related events. Participants' eating disorder symptomatology was then assessed, and body mass index (BMI) was calculated. Results showed processing of characters' emotional responses (detected via eye-tracking) to body- and perfectionism-related events for third-person scenarios was related to eating disorder symptomatology. Processing of characters' emotional responses to body-related events for second-person scenarios was related to males' BMI. The moment-to-moment processing of characters' emotional responses to food-related scenarios was not related to eating disorder symptomatology or BMI. Findings support theories that include body- and perfectionism-related cognitive biases as underlying mechanisms of eating disorder symptomatology and the use of implicit measures of cognitive processes underlying males' eating disorder symptomatology.