Characteristics of Monozygotic Twins Discordant for Anorexia Nervosa: Comprehensive Risk Evaluation for Anorexia Nervosa in Twins (CREAT) Study
Characteristics of Monozygotic Twins Discordant for Anorexia Nervosa: Comprehensive Risk Evaluation for Anorexia Nervosa in Twins (CREAT) Study

Characteristics of Monozygotic Twins Discordant for Anorexia Nervosa: Comprehensive Risk Evaluation for Anorexia Nervosa in Twins (CREAT) Study

Eur Eat Disord Rev. 2026 Sep 2. doi: 10.1002/erv.70165. Online ahead of print.

ABSTRACT

OBJECTIVE: To characterise the clinical and phenotypic profile of the Comprehensive Risk Evaluation for Anorexia nervosa in Twins (CREAT) cohort, identify candidate risk and illness-related correlates of AN, and establish a foundation for forthcoming biological, neuroimaging, endocrinological, and microbiota studies.

METHODS: MZ twins discordant for lifetime AN (44 individuals) and control MZ twin pairs (42 individuals) were recruited. Analyses included total-sample associations with AN, between-group comparisons (affected, unaffected co-twins, controls), and within-pair analyses of discordant twins.

RESULTS: Affected twins, most of whom were weight-restored and non-acute, differed markedly from unaffected co-twins and controls. Lifetime AN was associated with higher perfectionism, goal-directed drive, neuroticism, behavioural inhibition, impulsivity, broader psychiatric symptoms, teasing history, autism symptoms, and lower quality of life. Within-pair analyses implicated perfectionism, goal-directed drive, and competency-related teasing as candidate individual-specific risk factors for AN, with evidence of additional associations with impulsivity, behavioural inhibition, neuroticism, broader psychiatric symptoms, and autism symptoms.

CONCLUSION: Findings support a multifactorial model of AN involving individual-specific influences and shared familial liability. Perfectionism, goal-directed drive, and competency-related teasing emerged as candidate individual-specific risk factors, while the pattern of elevations, with unaffected co-twins often falling between affected twins and healthy controls, suggests that several clinical and phenotypic features may reflect familial liability for AN.

PMID:42683660 | DOI:10.1002/erv.70165