Background We previously demonstrated a relationship between takeaway outlet exposure and consumption of hot takeaway food and body mass. Little research has investigated whether individual differences in eating behaviour traits represent differences in susceptibility to these environmental influences on food consumption and body weight. The current study aimed to investigate the moderating role of eating behaviour traits (i.e. cognitive restraint, uncontrolled eating and emotional eating) in the relation between neighbourhood exposure to hot food takeaways, and takeaway consumption and body fat.

Methods This cross-sectional study included data from a cohort study of the environmental and genetic determinants of obesity and type 2 diabetes in Cambridgeshire, UK (i.e. phase 1 of the Fenland study with n=12,325). Takeaway food outlet exposure was derived using data from local authorities and divided into quarters (Q1 being the least exposed and Q4 being most exposed). Primary outcomes were consumption of energy dense foods commonly obtained from takeaway food outlets (derived from food frequency questionnaire), and body fat percentage (BF%; measured using dual-energy X-ray absorptiometry). The Three Factor Eating questionnaire (TFEQ-R18) was used to measure eating behaviour traits. Multivariable regression models were used to estimate the associations of takeaway exposure and eating behaviour traits, independently, with measured takeaway consumption and BF%. Effect modification was investigated by adding interaction terms (takeaway exposure x eating behaviour trait) to the models.

Results Mean age of participants was 48.6 (SD= 7.5) and 53.8% was female. More exposure to takeaway outlets around the residential neighbourhood was associated with increased consumption of takeaway foods (e.g. B_Q4= 4.7 g/day, 95%CI= 2.7; 6.8) and BF% (e.g. B_Q4= 1.2, 95%CI= 0.6; 1.8) compared to those least exposed. Uncontrolled eating and emotional eating were positively associated with takeaway consumption, whereas cognitive restraint was negatively associated with takeaway consumption (B= -0.1, 95%CI= -0.2; -0.1). All three eating behaviour traits were positively associated with BF%. The only statistically significant two-way interaction indicated that the association between takeaway exposure quartiles and BF% was less strong in those with higher emotional eating scores.

Conclusion Individual differences in eating behaviour traits and exposure to takeaways were both associated with increased takeaway consumption and body fat. There was no evidence that the effect of takeaway exposure was stronger in those with greater susceptibility to overeat. There was some evidence to suggest that in areas of high takeaway exposure, individual differences in emotional eating may be a less important determinant of body fat.