**Background:** The obesogenic environment, including easy access to hot food takeaway outlets, is thought to contribute to unhealthy diets and obesity. However, evidence linking takeaway outlet exposure to dietary and adiposity outcomes is mixed, possibly because individual psychological traits (eating behaviour traits) may moderate susceptibility to environmental food cues. This study investigated whether cognitive restraint, uncontrolled eating, and emotional eating (measured by the TFEQ-R18) moderate the associations between neighbourhood takeaway outlet exposure and both takeaway food consumption and body fat percentage.
**Methods:** Cross-sectional data were drawn from the Fenland Study, a population-based cohort in Cambridgeshire, UK. Participants (n=4791) were adults born between 1950 and 1975. Takeaway outlet exposure was defined as the count of outlets within a 1-mile Euclidean buffer around the home address, sourced from local council data (December 2011) and divided into quartiles (Q1: 0 outlets; Q2: 1–2; Q3: 3–12; Q4: 13–51). Eating behaviour traits were assessed using the TFEQ-R18, with raw scores transformed to a 0–100 scale. The primary outcomes were takeaway-like food consumption (g/day of pizza, burgers, fried fish, and French fries from a food frequency questionnaire) and body fat percentage (measured by dual-energy X-ray absorptiometry, DEXA). Secondary outcomes included fat mass index (FMI) and BMI. Covariates included age, sex, household income, age at completion of full-time education, occupational social class, and supermarket count in the home neighbourhood. Multiple linear regression was used, with multiplicative interaction terms (eating behaviour trait × takeaway outlet quartile) to test effect modification. Missing data were handled using multivariate imputation by chained equations (MICE). Statistical significance for interaction was set at α=0.10.
**Key Results:** Mean age was 51.0 years (SD 7.2), 53.9% were female, and median takeaway consumption was 30.0 g/day (IQR 17.5–47.5). Mean body fat percentage was 33.3% (SD 9.2). After adjustment, compared to Q1, living in Q3 was associated with 2.6 g/day higher takeaway consumption (95% CI 0.5, 4.8) and 1.8 percentage points higher body fat (95% CI 1.2, 2.5); Q4 was associated with 4.0 g/day higher takeaway consumption (95% CI 0.3, 7.6) and 1.8 percentage points higher body fat (95% CI 0.7, 2.9). Higher emotional eating and uncontrolled eating scores were positively associated with both outcomes (e.g., a 10-point higher uncontrolled eating score was associated with 1.1 percentage points higher body fat, 95% CI 0.9, 1.2). Cognitive restraint was negatively associated with takeaway consumption (β −2.2 per 10 points, 95% CI −2.6, −1.8) but positively associated with body fat percentage (β 0.2, 95% CI 0.1, 0.3). For interaction: cognitive restraint moderated the association with takeaway consumption (p=0.03); the association was strongest in those with high cognitive restraint scores. Emotional eating moderated the association with body fat percentage (p=0.05); the positive association was strongest in those with low emotional eating scores (e.g., low emotional eaters in Q4 had 2.8 percentage points higher body fat than those in Q1, 95% CI 1.6, 4.0, vs. 1.5 percentage points, 95% CI 0.3, 2.7 for high emotional eaters). Uncontrolled eating did not significantly moderate any association. The authors note that interaction effects were small and could be due to chance given multiple testing.
**Clinical Implications:** The findings suggest that while both environmental exposure and individual eating behaviour traits are independently associated with takeaway consumption and adiposity, the evidence that certain traits make individuals more susceptible to the food environment is weak. The small moderating effects observed indicate that intervening solely on individual eating behaviours is unlikely to fully mitigate the impact of the obesogenic environment, and vice versa. The authors recommend integrated strategies combining downstream approaches (e.g., mindfulness for high emotional eaters) with upstream population-level policies (e.g., restricting takeaway outlet density near schools). Limitations include the cross-sectional design, potential temporal mismatch between food outlet data (2011) and participant data (2005–2015), and reliance on self-reported dietary intake.