DSM-5 severity specifiers for anorexia nervosa and treatment outcomes in adult females

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.

Do risk factors differentiate DSM-5 and drive for thinness severity groups for anorexia nervosa?

BACKGROUND

The current study examined whether risk factors for anorexia nervosa (AN) were related to different levels of severity based on (a) the DSM-5/body mass index (BMI) and (b) drive for thinness (DT) severity ratings.

METHODS

The sample comprised 153 pairs of individuals with a lifetime diagnosis AN per DSM-IV criteria and their non-ED sisters (N = 306, mean age = 26.53; mean current BMI = 20.42 kg/m2). The Oxford risk factor interview was used to establish AN-related risk factors. Individuals were categorised into the DSM-5 severity groups based on their lowest BMI, while the DT subscale from the eating disorder inventory-2 was used to classify individuals with AN into low and high DT groups.

RESULTS

Multinominal regression models showed similar risk factors (e.g., perfectionism, having a history of being teased about weight and shape) contributed to the development of AN using the DSM-5 and DT severity ratings. Follow-up analyses across the severity groups for both indices revealed that only childhood perfectionism was found to be more common in the extreme severe DSM-5 BMI severity group compared to the severe DSM-5 group.

CONCLUSION

Overall, this study found little evidence for AN risk factors being related to the DSM-5 and DT severity ratings. However, given the novelty of this study, replication of the current results is warranted.

Cognitive flexibility and DSM-5 severity criteria for eating disorders: assessing drive for thinness and duration of illness as alternative severity variables

BACKGROUND

The severity criteria for eating disorders (EDs) proposed in the DSM-5 have been established without sufficient empirical support. Drive for thinness (DT) and duration of illness have been proposed as two alternative severity measures, however their empirical evidence is also limited. To date, no research has assessed the validity of current eating disorder (ED) severity criteria regarding cognitive flexibility factors. Cognitive flexibility is often impaired in EDs, becoming a possible severity symptom. The current study assessed for the first time (1) whether the severity indexes for EDs proposed in the DSM-5 were associated with deficits in cognitive flexibility and, (2) whether drive for thinness and illness duration, acted as an alternative, more meaningful severity indices for deficiencies in cognitive flexibility.

METHODS

Participants were 161 patients diagnosed with an ED, who were categorized according to DSM-5 severity categories, DT and duration of illness. Discriminative capacity of each classification was assessed for cognitive flexibility measured by Wisconsin card sorting test (WCST).

RESULTS

The findings for the DSM-5 classification comprised: (a) In the anorexia nervosa (AN) group, patients with moderate severity showed better scores in WCST than patients with mild and severe/extreme severity. Also, patients with moderate severity showed lower percentage of cognitive flexibility deficits than the other two severity categories; (b) For the binge spectrum disorders (BSD) group, the patients with mild severity showed a higher percentage of cognitive flexibility deficits than did the moderate and severe/extreme categories. When assessing the alternative severity index of DT, no differences were found in cognitive flexibility in any of the groups. Regarding illness duration, in the AN group the task performance of the patients with longer illness duration was worse than the performance of the short duration group and, in the BSD group, patients with longer duration also showed more deficits in cognitive flexibility than the patients with shorter duration of illness.

CONCLUSIONS

Our findings point out the limitations of the DSM-5 severity criteria to categorize cognitive flexibility in EDs and support illness duration as an alternative severity approach for EDs.

Assessing severity in anorexia nervosa: Do the DSM-5 and an alternative severity rating based on overvaluation of weight and shape severity differ in psychological and biological correlates?

OBJECTIVE

This study evaluated the severity ratings for anorexia nervosa (AN) in the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) and an alternative severity rating based on overvaluation of weight/shape, on a range of psychological and biological variables.

METHOD

A sample of 312 treatment-seeking patients with AN (mean age = 25.3, SD = 7.6; mean BMI = 16.8 kg/m² , SD = 2.4) were categorised using both DSM-5 severity levels (mild/moderate/severe/extreme) and weight/shape (low/high) overvaluation. The severity categories were compared on a range of psychological (e.g., eating psychopathology) and biological (e.g., sodium) variables.

RESULTS

Results showed that the overvaluation of weight/shape appeared better at indexing the level of severity in psychological variables among patients with AN compared to the DSM-5 severity rating with moderate to large effect sizes. Moreover, the DSM-5 mild and moderate severity groups experienced significantly higher eating and general psychopathology than the severe and extreme groups. Finally, neither the DSM-5 nor the weight/shape severity groups differed on any of the biological variables.

CONCLUSIONS

This study provided no support for the DSM-5 severity rating for AN, while initial support was found for the weight/shape overvaluation approach in indexing psychological but not biological correlates.

Body Mass Index Specifiers in Anorexia Nervosa: Anything below the "Extreme"?

The validity of body mass index (BMI) specifiers for anorexia nervosa (AN) has been questioned, but their applicability to inpatients with extremely low BMIs and their prognostic validity are currently unknown. Therefore, we designed this study: (a) to test current BMI specifiers in severe inpatients; (b) to explore a "very extreme" specifier (VE-AN; BMI ≤ 13.5); and (c) to verify inpatients' hospitalization outcome according to BMI severity. We enrolled 168 inpatients with AN completing the following: Eating disorder Examination-Questionnaire, Eating Disorder Inventory-2, State-Trait Anxiety Inventory, Beck Depression Inventory, Body Shape Questionnaire, and EQ-5D-VAS. According to the current BMI classification, those with a BMI < 15 versus those with non-extreme AN (NE-AN, BMI ≥ 15) differed on all measures but the quality of life with those with NE-AN reporting more impaired scores on all measures. Adopting an exploratory classification comparing VE-AN, extreme AN (E-AN, BMI = 13.6-14.99), and NE-AN, no differences emerged between VE-AN and E-AN, while those with NE-AN reported significantly more impaired scores on all variables while the quality of life again did not differ across groups. Hospitalization outcome improved for all groups, independently of BMI. Groups differed concerning the length of stay that mirrored BMI severity and impacted also hospitalization outcomes. Taken together, our data support the lack of validity of current BMI specifiers in AN, even in the acute setting. Moreover, the exploratory subgroup of patients with BMI ≤ 13.5 did not delineate a clinically different group.

The Mediator Effect of Personality on the Relationship Between Symptomatic Impairment and Treatment Outcome in Eating Disorders

Features of personality disorders (PDs) have been found to explain meaningful variance in the onset, maintenance, and symptomatic presentation of eating disorders (EDs), and a co-occurent personality pathology is commonly associated with poorer response to ED treatment. The "pathoplasty model" of the relationship between personality and EDs implies that, once both conditions are established, they are likely to interact in ways that modify therapy outcome; however, to date, no studies have explored overall personality functioning, and especially PD clusters, as a mediator of treatment outcome. The present study aimed at conjointly exploring the associations between personality functioning and PDs, respectively, with pre-treatment ED symptomatic impairment and therapy outcome; and the mediating role of personality variables. At treatment onset, a sample of 107 women with ED problems were evaluated using both the Structured Clinical Interview for DSM-5 (SCID-5-CV) and the Shedler-Westen Assessment Procedure-200 (SWAP-200)-a clinician-rated procedure to dimensionally assess personality. Participants were also asked to complete self-report questionnaires on overall ED symptomatology, symptoms of binge eating and purging behaviors, and therapy outcome. The findings showed that, over and above the categorical ED diagnosis, the SWAP-200 healthy personality functioning score mediated the relationship between baseline ED symptom severity and therapy outcome, as well as the association between baseline bulimic symptoms and treatment outcome; furthermore, SWAP-200 Cluster B PD scores mediated the link between baseline binge eating and purging symptoms and therapy outcome, whereas scores in Clusters A and C showed no significant effects. The findings suggest that personality-based outcome research may improve treatment effectiveness in this difficult-to-treat population.

Age dependency of body mass index distribution in childhood and adolescent inpatients with anorexia nervosa with a focus on DSM-5 and ICD-11 weight criteria and severity specifiers

Both DSM-5 and ICD-11 have provided weight cut-offs and severity specifiers for the diagnosis of anorexia nervosa (AN) in childhood, adolescence and adulthood. The aims of the current study focusing on inpatients aged < 19 years were to assess (1) the relationship between age and body mass index (BMI; kg/m²), BMI-centiles, BMI-standard deviation scores (BMI-SDS) and body height-SDS at referral, (2) the percentages of patients fulfilling the DSM-5 and ICD-11 weight criteria and severity categories for AN, and (3) the validity of the AN severity specifiers via analysis of both weight related data at discharge and inpatient treatment duration. The German Registry for Anorexia Nervosa encompassed complete data sets for 469 female patients (mean age = 15.2 years; range 8.9-18.9 years) with a diagnosis of AN (n = 404) or atypical AN (n = 65), who were ascertained at 16 German child and adolescent psychiatric hospitals. BMI at referral increased up to age 15 to subsequently plateau. Approximately one tenth of all patients with AN had a BMI above the fifth centile. The ICD-11 specifier based on a BMI-centile of 0.3 for childhood and adolescent AN entailed two equally sized groups of patients. Discharge data revealed limited validity of the specifiers. Height-SDS was not correlated with age thus stunting had no impact on our data. We corroborate the evidence to use the tenth instead of the fifth BMI-centile as the weight criterion in children and adolescents. Weight criteria should not entail major diagnostic shifts during the transition from adolescence to adulthood. The severity specifiers based on BMI or BMI-centiles do not seem to have substantial clinical validity.

Rehospitalization and "Revolving Door" in Anorexia Nervosa: Are There Any Predictors of Time to Readmission?

Objective: Anorexia nervosa (AN) is a severe psychiatric illness with multifactorial etiology and unsatisfactory treatment outcomes. Hospitalization is required for a substantial number of patients, and readmission (RA) commonly occurs. Some individuals need multiple hospitalizations sometimes over a short amount of time, thus, delineating the "revolving door" (RD) phenomenon. However, very little is known about readmissions and their frequency in AN. Therefore, we aimed to longitudinally investigate readmissions in AN in order to: (a) characterize patients with AN who need readmission (i.e., RA-AN), sometimes rapidly (RD-AN); (b) ascertain differences between RA-AN and non-RA-AN groups during baseline hospitalization; (c) investigate as to whether clinical or psychometric parameters worsened on RA; and (d) analyze predictors of time-to-readmission in AN. Methods: A total of 170 inpatients with AN were enrolled at their baseline hospitalization; all their subsequent rehospitalizations were recorded with a longitudinal design by which each patient has been observed for 3 years. Patients were classified as RD-AN if requiring a readmission <12 months since last discharge. Clinical characteristics were measured upon admission and discharge for each hospitalization, and at all time points, patients completed questionnaires assessing eating and general psychopathology, and body shape concerns. Results: Sixty-seven patients (39.4%) needed at least one readmission and 62 (92.5% of RA-AN) reported RD. Compared with non-RA-AN, those with RA-AN were younger, reported a shorter duration of illness, and were more frequently diagnosed with AN-BP. Also, greater severity of anxious and depressive symptoms and body shape concerns emerged in the RA-AN group. The outcome of baseline hospitalization did not differ between groups, and only depressive symptoms worsened at readmission. Shorter duration of AN and low weight gain during baseline hospitalization predicted early readmission but did not survive statistical control. In contrast, high scores on drive for thinness upon baseline hospital entry robustly predicted a shorter time to readmission even after statistical control. Discussion: Individuals with AN who require readmission do so over a short period notwithstanding a positive treatment outcome during their baseline hospitalization. Shorter time-to-readmission can be predicted mostly in case of marked drive for thinness and poor weight gain at baseline hospital admission.

Fertility and Reproduction after Recovery from Anorexia Nervosa: A Systematic Review and Meta-Analysis of Long-Term Follow-Up Studies

Reproductive health is compromised during anorexia nervosa (AN). However, it is still unclear whether this medical complication is reversible after recovery from AN. The purpose of this paper was to conduct a systematic review of the major reproductive health outcomes in females after recovery from AN. The review was conducted in adherence to preferred reporting items for systematic review and meta-analyses (PRISMA) guidelines. Data were collated using meta-analysis and a narrative approach. Of the 1186 articles retrieved, five studies met the inclusion criteria and were reviewed. These studies monitored weight-restored females who had recovered from AN for a follow-up period of between six and 18 years. Their narrative analysis revealed that appropriate treatment of AN leads to the normalization of reproductive function, especially in terms of fertility, pregnancy, and childbirth rates. The meta-analysis confirmed this finding, where the pooled odds of childbirth rates between the AN group and the general population was not statistically significant (OR = 0.75, 95% CI: 0.43–1.29, p = 0.41). We conclude that if patients undergo appropriate eating-disorder treatment and weight restoration, it appears to be unlikely that reproductive health is affected by AN. However, since this finding is derived from only a few studies, it requires replication and confirmation.

Drive for thinness provides an alternative, more meaningful, severity indicator than the DSM-5 severity indices for eating disorders

OBJECTIVE

To assess an alternative trans-diagnostic indicator for severity based on drive for thinness (DT) for anorexia nervosa (AN), bulimia nervosa (BN), binge-eating disorder (BED), and other specified feeding or eating disorder (OSFED), and to compare this new approach to the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) severity categories for EDs.

METHOD

A total of 2,811 ED [428 AN-restrictive (AN-R), 313 AN-binge purging (AN-BP), 1,340 BN, 329 BED, 154 OSFED/atypical AN (AT), and 223 OSFED/purging disorder (PD)] patients were classified using: (a) The DSM-5 severity categories and (b) a DT categorisation. These severity classifications were then compared based on ED symptoms, general psychopathology, personality, and impulsive behaviours.

RESULTS

For the DSM-5 categories, most ED patients fell into the 'mild' to 'moderate' categories. Using the DT categories, AN patients were mainly represented in the 'low' DT category, and BN, OSFED/AT, and PD in the 'high' DT category. The clinically significant findings were stronger for the DT than the DSM-5 severity approach (medium-to-large effect sizes). AN-BP and AN-R provided the most pronounced effects.

CONCLUSION

Our findings question the clinical value of the DSM-5 severity categorisation, and provide initial support for an alternative DT severity approach for AN. HIGHLIGHTS : This study assessed an alternative trans-diagnostic drive for thinness (DT) severity. Category for all eating disorder (ED) sub-types, and then compared this to the Diagnostic and Statistical Manual of Mental Disorders-5 (DSM-5) severity indices for EDs. ED symptoms, general psychopathology, personality, and impulsive behaviours were assessed using both classifications in a total of 2,811 female patients diagnosed with EDs. Clinically significant findings were stronger for the DT than the DSM-5 severity category (medium-to-large effect sizes); there was differentiation of the anorexia nervosa (AN) patients into mainly 'low' DT, and bulimia nervosa (BN) spectrum patients into mainly 'high' DT, vs. most patients were clustered in the 'mild-to-moderate' DSM-5 categories. Our findings provide initial support for an alternative trans-diagnostic DT severity category that may be more clinically meaningful than the DSM-5 severity indices for EDs.

A systematic review of enhanced cognitive behavioral therapy (CBT-E) for eating disorders

OBJECTIVE

To review the literature examining the efficacy and effectiveness of enhanced cognitive behavioral therapy (CBT-E) for adults and older adolescents with eating disorders.

METHOD

A systematic search of the literature (using PsycINFO and PubMed) was conducted in order to identify relevant publications (randomized controlled trials [RCTs] and uncontrolled trials) up to June 2019. Effect sizes were reported for outcomes including treatment attrition and remission rates, eating disorder behaviors, body mass index (BMI), and core eating disorder psychopathology. The Downs and Black checklist was used to assess the quality of included studies.

RESULTS

Twenty studies (10 RCTs and 10 uncontrolled trials) met criteria for inclusion. Support was found for the efficacy and effectiveness of CBT-E for the full spectrum of eating disorders, with respect to reducing eating disorder behaviors and core psychopathology. BMI also increased, with large effects, for individuals with AN. However, the majority of the randomized trials included in this review did not demonstrate superiority of CBT-E over comparison treatments, particularly in the longer-term. Furthermore, rates of attrition and remission for CBT-E among individuals without AN did not appear to differ from rates for CBT-BN.

DISCUSSION

There is evidence to support CBT-E as an efficacious and effective treatment for adults and older adolescents with a range of eating disorder diagnoses. Future research would benefit from directly comparing CBT-E to CBT-BN, expanding measured outcomes to include driven exercise and subjective binge eating, increasing consistency in the definition and measurement of outcomes, and exploring factors associated with treatment retention.

Mental, behavioral and neurodevelopmental disorders in the ICD-11: an international perspective on key changes and controversies

An update of the chapter on Mental, Behavioral and Neurodevelopmental Disorders in the International Classification of Diseases and Related Health Problems (ICD) is of great interest around the world. The recent approval of the 11th Revision of the ICD (ICD-11) by the World Health Organization (WHO) raises broad questions about the status of nosology of mental disorders as a whole as well as more focused questions regarding changes to the diagnostic guidelines for specific conditions and the implications of these changes for practice and research. This Forum brings together a broad range of experts to reflect on key changes and controversies in the ICD-11 classification of mental disorders. Taken together, there is consensus that the WHO's focus on global applicability and clinical utility in developing the diagnostic guidelines for this chapter will maximize the likelihood that it will be adopted by mental health professionals and administrators. This focus is also expected to enhance the application of the guidelines in non-specialist settings and their usefulness for scaling up evidence-based interventions. The new mental disorders classification in ICD-11 and its accompanying diagnostic guidelines therefore represent an important, albeit iterative, advance for the field.

Diagnostic Categories for Eating Disorders: Current Status and What Lies Ahead

The Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition) for 6 feeding and eating disorders were published in 2013 and were notable for officially recognizing binge-eating disorder and for articulating criteria for avoidant/restrictive food intake disorder. The criteria and the rationale for them are briefly described, and current and future challenges are discussed.