Table 4 Additive interaction between quartiles of fast-food outlet proportion and household income on the odds of being obese (BMI ≥ 30 kg/m²), modelled using a binomial logistic regression model in the UK Biobank sample, UK (n = 51,361)

>  100,000

ref^b

1.30 (1.07, 1.59);

P = 0.008^c

1.60 (1.27, 2.00);

P < 0.001^c

1.73 (1.34, 2.23);

P < 0.001^c

1.53 (1.12, 2.08)

P = 0.008^d

52,000–100,000

1.27 (1.08, 1.50);

P = 0.004^c

1.44 (1.22, 1.69);

P < 0.001^c

1.81 (1.53, 2.13);

P < 0.001^c

2.26 (1.91, 2.66);

P < 0.001^c

1.69 (1.43, 2.02)

P < 0.001^d

31,000–51,999

1.42 (1.20, 1.69);

P < 0.001^c

1.64 (1.39, 1.93);

P < 0.001^c

1.95 (1.66, 2.29);

P < 0.001^c

2.20 (1.87, 2.58);

P < 0.001^c

1.51 (1.27, 1.80)

P < 0.001^d

<  31,000

1.67 (1.43, 1.95);

P < 0.001^c

2.05 (1.76, 2.39);

P < 0.001^c

2.30 (1.98, 2.68);

P < 0.001^c

2.43 (2.09, 2.84);

P < 0.001^c

1.56 (1.37, 1.78)

P < 0.001^d

1.58 (1.34, 1.88)

P < 0.001^e

1.36 (1.15, 1.65)

P = 0.001^e

1.49 (1.21, 1.82)

P < 0.001^e

1.53 (1.22, 1.93)

P < 0.001^e