(P15-051-24) Functional Near Infrared Spectroscopy (fNIRS) Assessment of State Vs. Trait Eating Behavior

Objectives: Disordered Eating Behaviors (DEBs) are state-dependent (short-term) behaviors that can lead to trait-dependent (long-term) characteristics over time. The Eating Disorder Examination-Questionnaire (EDE-Q) Global Score (GS)^© does not allow for identification of DEBs to verify non-ED status for those seeking weight loss.

The overarching objective is to use a state-dependent neuroimaging method, functional Near Infrared Spectroscopy (fNIRS), to assess a modified EDE-Q GS for identifying DEBs in a non-ED population. Specifically, this work aims to (1) explore relationships between fNIRS variables, EDE-Q GS, and waist circumference (WC), (2) statistically assess the construction of the EDE-Q GS, and (3) modify the EDE-Q GS to embed the Behavioral Features (BF) subscale into it. Permission to modify the EDE-Q was obtained from the original author, Dr. Christopher Fairburn.

Methods: Secondary data analysis of fNIRS variables and self-reported eating behaviors (GS+BF) was conducted on data from a study examining relationships between regional prefrontal cortex activation and amount of food eaten in a test meal. A modified GS, the DEBS-EDE-Q, was developed to include and address BF. EDE-Q GS and DEBS-EDE-Q were statistically assessed using multiple regression.

Results: Data from 97 participants (M=45, W=52) were analyzed. HbR added significantly to the model predicting EDE-Q GS, >F(3, 76) = 3.107, p < 0.05, adj. R² = 0.109. WC predicted EDE-Q GS in models with Oxy and HbO, F (3, 78) = 1.895, p < 0.05, adj. R² = 0.068 and F(3, 78) = 1.565, p < 0.05, adj. R² = 0.057, respectively. WC significantly predicted DEBS-EDE-Q in all the multiple regression models and identified 18 participants who endorsed DEBs (as per the BF subscale).

Conclusions: Use of fNIRS revealed that using both state- and trait-dependent questions may provide a more comprehensive assessment of eating behavior for individuals not diagnosed with ED. DEBS-EDE-Q has potential to motivate a contemporary approach to identifying DEBs sooner, given that all antecedent EDE constructs (including the BF subscale) are embedded within it. As such, it may also encourage extension of its use in non-ED populations. Additional studies in non-clinical and clinical samples are warranted.

Funding Sources: No funding sources to be disclosed. This work was a secondary data analyses and part of the author's PhD dissertation.