ARS | Publication request: Usual energy intake mediated the relationship between food reinforcement and BMI

Submitted to: Obesity
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: December 22, 2011
Publication Date: January 13, 2012
Citation: Epstein, L.H., Carr, K.A., Lin, H., Fletcher, K.D., Roemmich, J.N. 2012. Usual energy intake mediated the relationship between food reinforcement and BMI. Obesity. DOI:10.1038/oby.2012.2.

Interpretive Summary: The motivation to eat is correlated with weighing more and the amount of energy consumed during a meal. One hypothesis relating these three variables is that the motivation to eat is related to BMI through usual energy intake. Using a database of two hundred sixty-four adults of varying weight and body mass index (BMI) levels, we found that the motivation to eat was correlated with both BMI (r = 0.15, p = 0.016), energy intake during a laboratory meal (r = 0.31, p < 0.001) and usual energy intake (r = 0.34, p < 0.001). We also found that BMI was correlated with laboratory (r = 0.12, p = 0.04) and usual energy intake (r = 0.18, p = 0.003). Laboratory and usual energy intake were also correlated (r = 0.22, p < 0.001). Tests for mediation showed that both laboratory and usual energy intake mediated the relationship between the motivation to eat and BMI. These data confirm that the motivation to eat may influence BMI through effects on energy intake.

Technical Abstract: The relative reinforcing value of food (RRVfood) is correlated with overweight status and energy consumed, as those who find food more reinforcing are heavier and consume more energy. One hypothesis relating these variables is that food reinforcement is related to BMI through usual energy intake. Using a database of two hundred sixty-four adults of varying weight and body mass index (BMI) levels, results showed RRVfood was correlated with BMI (r = 0.15, p = 0.016), RRVfood was correlated with laboratory energy intake (r = 0.31, p < 0.001) and usual energy intake (r = 0.34, p < 0.001), and BMI was correlated with laboratory (r = 0.12, p = 0.04) and usual energy intake (r = 0.18, p = 0.003). Laboratory and usual energy intake were also correlated (r = 0.22, p < 0.001). Tests for mediation showed that both laboratory and usual energy intake mediated the relationship between RRVfood and BMI. These data confirm that RRVfood influences BMI through effects on energy intake. Implications of laboratory or usual energy intake as mediators is discussed.