Review
Impairments among DSM-5 eating disorders: A systematic review and multilevel meta-analysis

https://doi.org/10.1016/j.cpr.2023.102267Get rights and content

Highlights

  • Multilevel meta-analyses compared impairments among DSM-5 eating disorders (EDs).

  • Other specified EDs have high physical and psychiatric morbidity.

  • DSM-5 does not differentiate between EDs in terms of impairment.

  • Diagnostic criteria need refining to facilitate proper ED diagnosis and treatment.

Abstract

Previous research revealed that people who did not meet full DSM-IV criteria for anorexia nervosa (AN), bulimia nervosa (BN), or binge-eating disorder (BED) but met criteria for eating disorder not otherwise specified (EDNOS) display high levels of psychiatric and physical morbidity commensurate with full criteria eating disorders. The DSM-5 introduced significant changes to eating disorder diagnostic criteria, so the present study aimed to determine whether the revised diagnostic criteria better distinguish between full criteria eating disorders, and other specified feeding or eating disorder (OSFED) and unspecified feeding or eating disorder (UFED). We present a series of meta-analyses comparing eating pathology, general psychopathology, and physical health impairments among those with AN, BN, and BED, compared to those with OSFED or UFED (n = 69 eligible studies). Results showed significantly more eating pathology in OSFED compared to AN, no difference in general psychopathology, and greater physical health impairments in AN. BN had greater eating pathology and general psychopathology than OSFED, but OSFED showed more physical health impairments. No differences were found between BN and purging disorder or low-frequency BN, or between BED and OSFED. Findings highlight the clinical severity of OSFED and suggest the DSM-5 criteria may not appropriately account for these presentations.

Introduction

The use of a universal classification system is important for psychiatric disorders. It helps to inform research on specific disorders, facilitates consistency in diagnoses and communication among health professionals, and aids in the development of targeted treatments. Eating disorders come in a variety of presentations, which may be linked with access to services, with different presentations responding to different therapies. Consequently, categorising them according to their specific symptoms is useful to ensure the appropriate interventions can be implemented. The DSM-5 outlines three specified eating disorders; anorexia nervosa (AN), bulimia nervosa (BN) and binge-eating disorder (BED)1(American Psychiatric Association [APA], 2013). These diagnoses are defined by specific, universally recognised criteria. Other specified feeding or eating disorder (OSFED) and unspecified feeding or eating disorder (UFED) are DSM-5 diagnoses given to those who present with a clinically significant eating disorder, but do not meet full criteria for AN, BN or BED (APA, 2013). An update to the DSM-5 has been published (DSM-5-TR) but no notable changes were made to AN, BN, BED, OSFED, or UFED criteria (APA, 2022).

Previous studies have demonstrated that most people who present with an eating disorder do not meet criteria for the three specified eating disorders (Fairburn & Cooper, 2011; Mitchison et al., 2020; Murray & Anderson, 2015; Stice, Marti, & Rohde, 2013). For example, a recent study by Mitchison et al. (2020) reported that the prevalence of DSM-5 eating disorders among Australian adolescent girls was 1.3%, 7.7% and 1.8% for AN, BN and BED respectively, and 14.5% for OSFED. The criteria for OSFED and UFED are broad and cater for a range of eating disorder symptoms. There are five presentations listed under OSFED (atypical anorexia nervosa, low-frequency or limited duration bulimia nervosa, low-frequency or limited duration binge-eating disorder, purging disorder, and night-eating syndrome) and UFED represents eating disorders that are not described under any AN, BN, BED or OSFED presentation. This heterogeneity means that an umbrella diagnosis of OSFED or UFED is not very informative, unless a specific OSFED presentation is described. This can often make treatment planning and communication among health professionals difficult (Murray & Anderson, 2015).

Evidence for effective OSFED and UFED interventions is also limited in comparison to AN, BN and BED (Riesco et al., 2018). This is largely because OSFED and UFED were introduced in 2013, so these disorders have existed as researchable entities for less time than AN, BN and BED. Although BED was officially recognised as a specified eating disorder in 2013, diagnostic criteria for BED was first proposed by Spitzer et al. in 1991 and the disorder has received considerable research attention since (Dingemans, Bruna, & Van Furth, 2002; Spitzer et al., 1991). Despite the limited research on OSFED and UFED to date, available evidence suggests that people with these disorders display similar levels of impairment to those with AN, BN, and BED (Mustelin, Lehtokari, & Keski-Rahkonen, 2016; Wade & O'Shea, 2015; Withnell, Kinnear, Masson, & Bodell, 2022), however this research has not been consolidated since the introduction of these diagnoses to the DSM-5, representing a significant gap in the literature. We define impairment as a reduction in psychological, psychosocial, or physical functioning.

Prior to the publication of the DSM-5, eating disorder not otherwise specified (EDNOS) was the diagnosis given to those who presented with a clinically significant eating disorder but did not meet criteria for AN or BN (American Psychiatric Association, 2013). Thomas, Vartanian, and Brownell (2009) conducted a systematic review and meta-analysis investigating the relationship between EDNOS and AN, BN, and BED. The latter disorder was listed under EDNOS in the DSM-IV, however due to the distinct criteria for BED, Thomas et al. (2009) investigated it separately alongside AN and BN. Thomas et al. (2009) conducted a meta-analysis of 125 studies, including 84 that compared EDNOS to AN, 99 comparing EDNOS to BN, and 30 comparing EDNOS to BED. The authors compared the eating pathology, general psychopathology, and physical health impairments experienced by those with DSM-IV eating disorders with the aim of informing DSM revisions, including those made in the DSM-5 (Thomas et al., 2009). Thomas et al. (2009) reported that there were no significant differences between EDNOS and AN with respect to eating pathology, general psychopathology, or physical health. BN showed higher general psychopathology than EDNOS, and EDNOS showed poorer physical health than BN. BN also showed higher eating pathology than EDNOS, but when those with EDNOS group met all criteria for BN except binge-eating frequency, there was no significant difference. No differences in eating pathology or general psychopathology were found between those with EDNOS and those with BED. Overall, those with EDNOS demonstrated high levels of psychiatric and physical morbidity that were in line with the specified DSM-IV eating disorders (Thomas et al., 2009).

Since the publication of the Thomas et al. (2009) meta-analysis, the DSM-5 was published which broadened the criteria for feeding and eating disorders. Changes included relaxing the weight criterion and removing the amenorrhea requirement for AN, reducing the frequency of binge-eating and compensatory behaviour requirement for BN, recognition of BED as a distinct disorder, and replacing EDNOS with OSFED and UFED (Attia et al., 2013). Criteria were developed for five additional eating disorder presentations now listed under OSFED, including atypical anorexia nervosa (AAN), low-frequency or limited duration bulimia nervosa, low-frequency or limited duration binge-eating disorder, purging disorder, and night-eating syndrome. The UFED diagnosis was also included to recognise significant eating disorders that do not meet criteria for AN, BN, BED, or any OSFED presentation, or for situations where there is insufficient information to make a more specific diagnosis (American Psychiatric Association, 2013; Attia et al., 2013). These changes to DSM-5 criteria were introduced with the intention of reducing the number of eating disorder diagnoses assigned to the ‘other specified’ categories (Call, Walsh, & Attia, 2013). In turn, it was expected this would improve the clinical utility of eating disorder diagnoses by providing a more specific representation of a person's symptoms and better differentiation between diagnostic categories in terms of symptom severity (Call et al., 2013; Thomas et al., 2009). The broadening of the AN, BN and BED criteria has been successful in reducing the number of people diagnosed with OSFED or UFED by accommodating a large number who previously fell short of meeting all criteria, however the majority of eating disorder presentations still fall under the OSFED and UFED categories (Ernst, Bürger, & Hammerle, 2017; Keel, Brown, Holm-Denoma, & Bodell, 2011; Mitchison et al., 2020; Murray & Anderson, 2015; Stice et al., 2013). Research on whether these changes have succeeded in improving the differentiation between symptom severity across eating disorder diagnoses has yet to be consolidated.

The aim of the present study was to provide an update on the meta-analysis by Thomas et al. (2009) by investigating the physical and psychological impairments experienced by those with DSM-5 eating disorders. Specifically, we aimed to conduct a series of meta-analyses to determine whether those diagnosed with OSFED or UFED experience significantly different levels of impairment to those diagnosed with AN, BN or BED. This will build on the Thomas et al. (2009) findings by investigating whether the changes made to the DSM-5 have improved the distinction between specified and unspecified eating disorder presentations in terms of clinical impairment severity. To allow the most direct comparison of our findings to the Thomas et al. (2009) meta-analysis, we analysed impairment measures in the same three categories: eating pathology, general psychopathology, and physical health.

Section snippets

Search strategy

Four electronic databases were searched, including Medline, PubMed, CINAHL, and PsychINFO. The primary search included studies published from May 2013 (when the DSM-5 was published) onwards. Search terms and criteria were adapted from the previous meta-analysis by Thomas et al. (2009). We searched for studies containing the following terms in the title or abstract: (other specified feeding or eating disorder* OR OSFED OR unspecified feeding or eating disorder* OR UFED). For the databases that

Study characteristics

We identified 69 studies eligible for inclusion in this meta-analysis (see Supplemental Appendix D for full list). Of these, 80% reported effect sizes comparing OSFED to AN, 68% compared OSFED to BN, and 41% compared OSFED to BED. There were a total of 20,651 participants across all studies, including 6623 participants with AN, 5532 participants with BN, 1692 participants with BED, and 6804 participants with OSFED or UFED. The average mean participant age was 25.2 (SD 7.4, range 14.0–40.5) and

Discussion

The present study sought to compare eating pathology, general psychology, and physical health impairments among those with DSM-5 eating disorders, with the specific aim of determining whether there were any differences in impairment levels between those with AN, BN or BED, compared to those with OSFED or UFED. Our study was designed as a follow-up to the meta-analysis by Thomas et al. (2009) who investigated impairment among DSM-IV eating disorders. Based on their findings, Thomas et al. (2009)

Role of funding sources

The first author is supported by a Flinders University Research Scholarship. Flinders University had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the decision to submit the paper for publication. The third author is supported by an Australian Government Research Training Program Scholarship. The Australian Government had no role in the study design, collection, analysis or interpretation of the data, writing the manuscript, or the

CRediT authorship contribution statement

Madeleine Wilkop: Conceptualization, Methodology, Investigation, Formal analysis, Data curation, Writing – original draft, Writing – review & editing, Visualization, Project administration. Tracey Wade: Conceptualization, Methodology, Resources, Writing – review & editing, Supervision. Ella Keegan: Investigation, Data curation. Sarah Cohen-Woods: Conceptualization, Investigation, Methodology, Resources, Writing – review & editing, Supervision.

Declaration of Competing Interest

None.

References (67)

  • E. Attia et al.

    Feeding and eating disorders in DSM-5

    American Journal of Psychiatry

    (2013)
  • A. Austin et al.

    Duration of untreated eating disorder and relationship to outcomes: A systematic review of the literature

    European Eating Disorders Review

    (2021)
  • A.T. Beck et al.

    An inventory for measuring depression

    Archives of General Psychiatry

    (1961)
  • N.D. Berkman et al.

    Outcomes of eating disorders: A systematic review of the literature

    The International Journal of Eating Disorders

    (2007)
  • K. Bohn et al.

    The clinical impairment assessment questionnaire (CIA)

    Cognitive Behavioral Therapy for Eating Disorders

    (2008)
  • M. Borenstein et al.

    Introduction to meta-analysis

    (2009)
  • A. Burt et al.

    Prevalence, features and health impacts of eating disorders amongst first-Australian Yiramarang (adolescents) and in comparison with other Australian adolescents

    Journal of Eating Disorders

    (2020)
  • C. Call et al.

    From DSM-IV to DSM-5: Changes to eating disorder diagnoses

    Current Opinion in Psychiatry

    (2013)
  • J. Cohen

    Statistical power analysis

    Current Directions in Psychological Science: a Journal of the American Psychological Society

    (1992)
  • R.D. Cook et al.

    Residuals and influence in regression

    (1982)
  • M. Cyr et al.

    Abnormal fronto-striatal activation as a marker of threshold and subthreshold bulimia nervosa

    Human Brain Mapping

    (2018)
  • A.M. Darcy et al.

    The eating disorders examination in adolescent males with anorexia nervosa: How does it compare to adolescent females?

    The International Journal of Eating Disorders

    (2012)
  • L.R. Derogatis et al.

    Symptom checklist-90-revised

    The Corsini encyclopedia of psychology

    (2010)
  • A. Dingemans et al.

    Binge eating disorder: A review

    International Journal of Obesity

    (2002)
  • M.E. Duffy et al.

    Presence and severity of suicidal thoughts and behaviors across the eating disorder diagnostic spectrum: A pilot study

    Journal of Clinical Psychology

    (2021)
  • M. Egger et al.

    Bias in meta-analysis detected by a simple, graphical test

    BMJ

    (1997)
  • V. Ernst et al.

    Prevalence and severity of eating disorders: A comparison of DSM-IV and DSM-5 among German adolescents

    International Journal of Eating Disorders

    (2017)
  • C.G. Fairburn et al.

    Eating disorder examination questionnaire

    Cognitive Behavior Therapy and Eating Disorders

    (2008)
  • C.G. Fairburn et al.

    Eating disorders, DSM-5 and clinical reality

    British Journal of Psychiatry

    (2011)
  • M.F. Flament et al.

    Comparative distribution and validity of DSM-IV and DSM-5 diagnoses of eating disorders in adolescents from the community

    European Eating Disorders Review

    (2015)
  • K.J. Forney et al.

    A naturalistic, long-term follow-up of purging disorder

    Psychological Medicine

    (2021)
  • D.M. Garner et al.

    Development and validation of a multidimensional eating disorder inventory for anorexia nervosa and bulimia

    International Journal of Eating Disorders

    (1983)
  • C.W. Habeck et al.

    Community-level impacts of white-tailed deer on understorey plants in north American forests: A meta-analysis

    AoB Plants

    (2015)
  • View full text