Department of Psychology, University of Kansas, Lawrence, Kansas, USA.
Department of Psychiatry, University of Wisconsin-Madison, Madison, Wisconsin, USA.
Int J Eat Disord. 2020 Dec;53(12):1983-1992. doi: 10.1002/eat.23383. Epub 2020 Sep 18.
The Criterion B binge-eating symptoms represent five symptoms associated with binge eating. Any three out of five symptoms can be used to meet Criterion B. However, Criterion B symptoms may not be interchangeable in terms of how binge-eating severity is associated with each symptom. Item response theory (IRT) can test how endorsing each symptom relates to the overall level (i.e., severity) of binge-eating measured by Criterion B. We used IRT to identify (a) how each Criterion B symptom corresponded with binge-eating severity in a transdiagnostic binge-eating sample and (b) how well each symptom differentiated individuals with differing levels of severity.
Participants (N = 219) were adults (80.8% female) with a current ED that included objective binge-eating episodes assessed via semi-structured interview. A two-parameter logistic IRT model evaluated how endorsement of each Criterion B symptom corresponded with the level of latent binge-eating severity.
"Eating large amounts when not hungry" and "eating alone" reflected the highest binge-eating severity. "Eating alone" was the best discriminator across different binge-eating severity levels, whereas "uncomfortably full" was the poorest discriminator across binge-eating severity levels.
Criterion B symptoms were not interchangeable in terms of what level of binge-eating severity corresponded with symptom endorsement. "Eating large amounts when not hungry" or "eating alone" may signify elevated binge-eating severity, whereas "uncomfortably full" and "feeling disgusted/depressed/guilty" were not necessarily indicative of elevated severity. Results suggested that Criterion B may need to be revised to eliminate symptoms that are redundant with other binge-eating diagnostic criteria.
标准 B 暴食症状代表与暴食相关的五个症状。任何五个症状中的三个可以用来满足标准 B。然而，就暴食严重程度与每个症状的关系而言，标准 B 症状可能不是可互换的。项目反应理论 (IRT) 可以测试每个症状对标准 B 衡量的暴食整体水平（即严重程度）的关联程度。我们使用 IRT 来确定：（a）在跨诊断暴食样本中，每个标准 B 症状与暴食严重程度的对应关系；（b）每个症状区分不同严重程度个体的效果。
参与者（N = 219）为患有当前饮食障碍的成年人（80.8%为女性），通过半结构化访谈评估存在客观的暴食发作。双参数逻辑 IRT 模型评估了每个标准 B 症状对潜在暴食严重程度的对应关系。
“当不饿时吃大量食物”和“独自进食”反映了最高的暴食严重程度。“独自进食”是不同暴食严重程度水平的最佳区分指标，而“感到非常饱”是暴食严重程度水平的最差区分指标。
就与症状确认相对应的暴食严重程度水平而言，标准 B 症状在概念上并不相同。“当不饿时吃大量食物”或“独自进食”可能表明暴食严重程度较高，而“感到非常饱”和“感到厌恶/沮丧/内疚”并不一定表明严重程度较高。结果表明，标准 B 可能需要修订，以消除与其他暴食诊断标准冗余的症状。
