skip to main content

Explore Research Products in the NSF-PAR

This content will become publicly available on February 14, 2025

Title: Differentiation between atypical anorexia nervosa and anorexia nervosa using machine learning
Abstract 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 twomeaningfulclasses 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.

  more » « less
NSF-PAR ID:
10490838
Author(s) / Creator(s):
 ;  ;  ;  ;  ;  ;  ;  ;  
Publisher / Repository:
Wiley Blackwell (John Wiley & Sons)
Date Published:
Journal Name:
International Journal of Eating Disorders
Volume:
57
Issue:
4
ISSN:
0276-3478
Format(s):
Medium: X Size: p. 937-950
Size(s):
["p. 937-950"]
Sponsoring Org:
National Science Foundation
More Like this
  1. ; ; ; ; ; ; ; ; ; ; et al ( , International Journal of Eating Disorders)
    Abstract Objective

    This study examined a hypothesized pathway by which interoceptive dysfunction accounted for associations between personality features (harm avoidance, self‐directedness, and perfectionism) and anorexia nervosa (AN) severity (indicated by drive for thinness, eating disorder‐related preoccupations and rituals, and body mass index).

     Method

    The study sample (n = 270, mean age = 28.47, 95.2% female, 98% White/Caucasian) consisted of probands and biological relatives who met DSM‐IV criteria for lifetime diagnoses of AN (omitting criterion D, amenorrhea) drawn from the Price Foundation Anorexia Nervosa Affected Relative Pairs Study (AN‐ARP). Participants completed measures assessing personality, interoceptive dysfunction, and eating pathology.

     Results

    Associations between personality features of low self‐directedness and high perfectionism and indicators of AN severity (drive for thinness and eating disorder‐related preoccupations and rituals) were significant, as were the hypothesized indirect pathways through interoceptive dysfunction. Neither harm avoidance nor body mass index was significantly related to other study variables, and the proposed indirect pathways involving these variables were not significant.

     Discussion

    Findings suggest that certain personality features may relate to AN severity, in part, through their associations with interoceptive dysfunction. Future research should examine prospective associations and the value of interventions targeting interoceptive dysfunction for interrupting the link between personality and AN severity.

     
    more » « less
  2. ; ; ; ; ; ; ; ( , International Journal of Eating Disorders)
    Abstract Objective

    Avoidant/restrictive food intake disorder (ARFID) and anorexia nervosa (AN) are restrictive eating disorders. There is a proposal before the American Psychiatric Association to broaden the currentDSM‐5criteria for ARFID, which currently require dietary intake that is inadequate to support energy or nutritional needs. We compared the clinical presentations of ARFID and AN in an outpatient sample to determine how a more inclusive definition of ARFID, heterogeneous for age and weight status, is distinct from AN.

     Methods

    As part of standard care, 138 individuals with AN or ARFID completed an online assessment battery and agreed to include their responses in research.

     Results

    Individuals with ARFID were younger, reported earlier age of onset, and had higher percent median BMI (%mBMI) than those with AN (allps < .001). Individuals with ARFID scored lower on measures of eating pathology, depression, anxiety, and clinical impairment (allps < .05), butdid notdiffer from those with AN on restrictive eating (p = .52), and scored higher on food neophobia (p < .001).

     Discussion

    Allowing psychosocial impairment to be sufficient for an ARFID diagnosis resulted in a clinical picture of ARFID such that %mBMI was higher (and in the normal range) compared with AN. Differences in gender distribution, age, and age of onset remained consistent with previous research. Both groups reported similar levels of dietary restriction, although ARFID can be distinguished by relatively higher levels of food neophobia. Currently available measures of eating pathology may capture certain ARFID symptoms, but highlight the need for measures of impairment relative to ARFID.

     
    more » « less
  3. ; ; ; ; ; ; ; ; ( , International Journal of Eating Disorders)
    Abstract Objective

    To compare individuals who have experienced binge‐eating disorder (BED) and anorexia nervosa (AN) (BED AN+) to those who have experienced BED and not AN (BED AN–).

     Method

    Participants (N = 898) met criteria for lifetime BED and reported current binge eating. Approximately 14% had a lifetime diagnosis of AN. Analyses compared BED AN+ and BED AN– on sociodemographic variables and clinical history.

     Results

    The presence of lifetimeANwas associated with more severe eating disorder symptoms, including earlier onset, more frequent, more chronic, and more types of eating disorder behaviors over the lifetime, as well as a higher lifetime prevalence of bulimia nervosa (BN). Participants with lifetimeANreported being more likely to have received treatments forBEDorBN, had significantly lower minimum, current, and maximumBMIs, had more severe general anxiety, and were significantly more likely to be younger and female. In the full sample, the lifetime prevalence of unhealthy weight control behaviors was high and treatment utilization was low, despite an average 15‐year history since symptom onset. Gastrointestinal disorders and comorbid anxiety, depression, and attention‐deficit/hyperactivity disorder symptoms were prevalent.

     Discussion

    Individuals fared poorly on a wide array of domains, yet those with lifetimeANfared considerably more poorly. All patients withBEDshould be screened for mental health and gastrointestinal comorbidities and offered referral and treatment options.

     Public Significance

    Individuals experiencing binge‐eating disorder have severe symptomology, but those who have experienced binge‐eating disorder and anorexia nervosa fare even more poorly. Our study emphasizes that patients with binge‐eating disorder would benefit from being screened for mental health and gastrointestinal comorbidities, and clinicians should consider history of unhealthy weight control behaviors to inform treatment and relapse prevention.

     
    more » « less
  4. ; ; ; ; ; ; ; ; ; ; et al ( , International Journal of Eating Disorders)
    Abstract Objective

    This study examined whether patterns of eating‐disorder (ED) psychopathology differed by gender acrossDSM‐5severity specifiers in anorexia nervosa (AN) and bulimia nervosa (BN).

     Method

    We tested whether ED psychopathology differed acrossDSM‐5 severity specifiers among 532 adults (76% female) in a residential treatment center with AN or BN. We hypothesized that severity of ED psychopathology would increase in tandem with increasing severity classifications for both males and females with AN and BN.

     Results

    Among females with BN,DSM‐5severity categories were significantly associated with increasing ED psychopathology, including Eating Disorder Examination‐Questionnaire dietary restraint, eating concern, shape concern, and weight concern; and Eating Disorder Inventory drive for thinness and bulimia. ED psychopathology did not differ acrossDSM‐5severity levels for males with BN. For both males and females with AN, there were no differences in ED psychopathology across severity levels.

     Discussion

    Results demonstrate thatDSM‐5severity specifiers may function differently for males versus females with BN. Taken together, data suggestDSM‐5severity specifiers may not adequately capture severity, as intended, for males with BN and all with AN. Future research should evaluate additional clinical validators ofDSM‐5severity categories (e.g., chronicity, treatment non‐response), and consider alternate classification schemes.

     
    more » « less
  5. ; ; ( , Synapse)
    Abstract

    Anorexia nervosa (AN) is a mental illness with the highest rates of mortality and relapse, and no approved pharmacological treatment. Using an animal model of AN, called activity‐based anorexia (ABA), we showed earlier that a single intraperitoneal injection of ketamine at a dose of 30 mg/kg (30mgKET), but not 3 mg/kg (3mgKET), has a long‐lasting effect upon adolescent females of ameliorating anorexia‐like symptoms through the following changes: enhanced food consumption and body weight; reduced running and anxiety‐like behavior. However, there were also individual differences in the drug's efficacy. We hypothesized that individual differences in ketamine's ameliorative effects involve drebrin A, an F‐actin‐binding protein known to be required for the activity‐dependent trafficking of NMDA receptors (NMDARs). We tested this hypothesis by electron microscopic quantifications of drebrin A immunoreactivity at excitatory synapses of pyramidal neurons (PN) and GABAergic interneurons (GABA‐IN) in deep layer 1 of prefrontal cortex (PFC) of these mice. Results reveal that (1) the areal density of excitatory synapses on GABA‐IN is greater for the 30mgKET group than the 3mgKET group; (2) the proportion of drebrin A+ excitatory synapses is greater for both PN and GABA‐IN of 30mgKET than 3mgKET group. Correlation analyses with behavioral measurements revealed that (3) 30mgKET's protection is associated with reduced levels of drebrin A in the cytoplasm of GABA‐IN and higher levels at extrasynaptic membranous sites of PN and GABA‐IN; (5) altogether pointing to 30mgKET‐induced homeostatic plasticity that engages drebrin A at excitatory synapses of both PN and GABA‐IN.

     
    more » « less
  • Citation Formats
  • MLA
  • APA
  • Chicago
  • BibTeX
  • Save / Share this Record
  • Send to Email