Using association rules mining to characterize of loss of control eating in childhood

Alaina L. Pearce, Timothy R. Brick, Travis Masterson, Shana Adise, S. Nicole Fearnbach, Wendy Stein, Laural English, Marian Tanofsky-Kraff, Kathleen L. Keller

Research output: Contribution to journalArticlepeer-review

Abstract

Childhood loss of control (LOC)-eating, the perceived inability to stop or control eating, is associated with increased risk for binge-eating disorder and obesity. However, the correlates of LOC-eating in childhood remain unclear. A secondary analysis of 177, 7-12-year-old children from five laboratory feeding studies was performed to investigate potential family (e.g., frequency of meals together, feeding practices), parental (e.g., education, weight status), and child (e.g., weight status, appetite traits) correlates of LOC-eating. Association rules mining (ARM1), a data-driven approach, was used to examine all characteristics that were common across studies to identify which were associated with LOC-eating. Results showed LOC-eating was characterized by a combination of child appetitive behaviors and parental feeding practices. In particular, LOC-eating was associated with low parental pressure to eat in combination with a high propensity to want to eat all the time and frequent refusal or dislike of novel foods. This pattern of both food approach (i.e., wanting to eat all the time) and avoidant behaviors (i.e., food fussiness) highlights the need for more research to characterize the complex patterns of appetitive traits associated with LOC-eating. In contrast, the absence of LOC-eating was associated with a low propensity to want to eat all the time, greater family income, and infrequent emotional overeating. Therefore, propensity to want to eat all the time, a single question from the Children's Eating Behavior Questionnaire, characterized both the presence and absence of LOC-eating, highlighting the need for more research to determine if this question captures clinically relevant individual differences. Future studies addressing these questions will advance our understanding of pediatric LOC-eating and may lead to interventions to reduce risk for more severe eating disorder symptomology.

Original languageEnglish (US)
Article number105236
JournalAppetite
Volume163
DOIs
StatePublished - Aug 1 2021

All Science Journal Classification (ASJC) codes

  • General Psychology
  • Nutrition and Dietetics

Fingerprint

Dive into the research topics of 'Using association rules mining to characterize of loss of control eating in childhood'. Together they form a unique fingerprint.

Cite this