Scholarlink

Background <p>Eating disorders (ED) are disabling conditions increasingly conceptualized as addictive disorders. While food addiction (FA) is well documented in bulimia nervosa (BN) and binge-eating disorder (BED), its mechanisms in anorexia nervosa (AN) remain unclear. The triadic model of addiction—encompassing reflexive, affective, and interoceptive systems—offers a relevant framework to investigate FA across EDs. The current study aimed to test to what extent these three components of addiction contribute to FA in patients with ED.</p> Methods <p>The sample included 115 patients (mean age = 28.41 years, <i>SD</i> = 11.64; 91% female) diagnosed with either AN (restrictive: <i>n</i> = 38; binge-eating/purging type: <i>n</i> = 32), BN (<i>n</i> = 29), or BED (<i>n</i> = 16). Participants completed the Emotional Appetite Questionnaire, the interoceptive subscale of the Eating Disorder Inventory-2, the Detail and Flexibility Questionnaire, and the Yale Food Addiction Scale 2.0. In the present analyses, FA was operationalized using the continuous total score reflecting FA symptom severity. Additional analyses using the symptom-count and diagnostic scoring procedures yielded comparable results. Linear and multinomial logistic regressions were used to test the associations between FA and the three systems, and their ability to distinguish diagnostic groups.</p> Results <p>When considered separately, interoceptive difficulties (<i>r</i> = .54), negative emotions (<i>r</i> = .47), and cognitive inflexibility (<i>r</i> = .33) were all associated with higher FA (all <i>p</i>s &lt; 0.001). However, in the joint model, only interoception (β = 0.39, <i>p</i> &lt; .001) and negative emotions (β = 0.33, <i>p</i> = .005) remained significant predictors, whereas cognitive flexibility had no significant effect. These effects did not differ significantly between diagnostic groups. The affective system predicted diagnostic (χ²(3) = 21.78, <i>p</i> &lt; .001), but the effect was inconsistent and the overall predictive value was modest (McFadden’s R² = 0.11, classification accuracy = 50%).</p> Conclusion <p>Interoceptive and affective systems appear to significantly predict FA across various ED; the reflexive system, as currently measured, seems to play a more limited role. These results highlight the need for more sensitive assessment tools and refined models of ED-related addictive dimensions.</p> <p><i>Trial Registration</i> NCT03160443, first post 2017-05-03.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Toward a triadic approach in eating disorders: investigating the contribution of the reflexive, affective and interoceptive systems in a clinical data cohort

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月15日

查看全文
  • Valentin Flaudias,
  • Corentin Gonthier,
  • Aurélien Cornil,
  • Maude Seneque,
  • Clara Lakritz,
  • Sylvain Iceta,
  • Sébastien Guillaume

摘要

Background

Eating disorders (ED) are disabling conditions increasingly conceptualized as addictive disorders. While food addiction (FA) is well documented in bulimia nervosa (BN) and binge-eating disorder (BED), its mechanisms in anorexia nervosa (AN) remain unclear. The triadic model of addiction—encompassing reflexive, affective, and interoceptive systems—offers a relevant framework to investigate FA across EDs. The current study aimed to test to what extent these three components of addiction contribute to FA in patients with ED.

Methods

The sample included 115 patients (mean age = 28.41 years, SD = 11.64; 91% female) diagnosed with either AN (restrictive: n = 38; binge-eating/purging type: n = 32), BN (n = 29), or BED (n = 16). Participants completed the Emotional Appetite Questionnaire, the interoceptive subscale of the Eating Disorder Inventory-2, the Detail and Flexibility Questionnaire, and the Yale Food Addiction Scale 2.0. In the present analyses, FA was operationalized using the continuous total score reflecting FA symptom severity. Additional analyses using the symptom-count and diagnostic scoring procedures yielded comparable results. Linear and multinomial logistic regressions were used to test the associations between FA and the three systems, and their ability to distinguish diagnostic groups.

Results

When considered separately, interoceptive difficulties (r = .54), negative emotions (r = .47), and cognitive inflexibility (r = .33) were all associated with higher FA (all ps < 0.001). However, in the joint model, only interoception (β = 0.39, p < .001) and negative emotions (β = 0.33, p = .005) remained significant predictors, whereas cognitive flexibility had no significant effect. These effects did not differ significantly between diagnostic groups. The affective system predicted diagnostic (χ²(3) = 21.78, p < .001), but the effect was inconsistent and the overall predictive value was modest (McFadden’s R² = 0.11, classification accuracy = 50%).

Conclusion

Interoceptive and affective systems appear to significantly predict FA across various ED; the reflexive system, as currently measured, seems to play a more limited role. These results highlight the need for more sensitive assessment tools and refined models of ED-related addictive dimensions.

Trial Registration NCT03160443, first post 2017-05-03.