Clustering of eating disorder symptoms in a general population female twin sample: A latent class analysis

Research output: Contribution to journalArticlepeer-review

37 Scopus citations

Abstract

Background. Previous studies have reported that the current DSM-IV eating disorder (ED) criteria do not adequately describe ED symptomatology. The objective of the current study was to examine the clustering of ED symptoms in a general population sample using latent class analysis (LCA). Method. ED symptoms from 3723 female young adult twins (mean age 22) were analyzed using LCA, and resulting classes were compared on external validators reflecting ED and other comorbid psychiatric diagnoses, substance use disorders (SUDs), and suicidality. Results. The optimal solution consisted of five latent classes characterized as: (1) Unaffected; (2) Low Weight Gain; (3) Weight Concerned; (4) Dieters; and (5) ED. Members of the ED class had significantly higher prevalence of co-morbid psychiatric disorders, SUDs, and suicidality than the Unaffected and Low Weight Gain classes, and elevated rates of suicidality and major depression compared to the Weight Concerned and Dieter classes, which differed from each other primarily in terms of current body mass index (BMI). Dieter class members were more likely to be overweight and obese and less likely to be underweight than Weight Concerned class members. The majority of women with an ED diagnosis were assigned to the ED class, and few differences were found between ED class members with and without an ED diagnosis. Conclusions. The results add to the evidence that many women with significant ED psychopathology are not being identified by the DSM-IV ED categories.

Original languageEnglish
Pages (from-to)1097-1107
Number of pages11
JournalPsychological medicine
Volume37
Issue number8
DOIs
StatePublished - Aug 2007

Fingerprint

Dive into the research topics of 'Clustering of eating disorder symptoms in a general population female twin sample: A latent class analysis'. Together they form a unique fingerprint.

Cite this