Impact of increasing the availability of healthier vs. less-healthy food on food selection: a randomised laboratory experiment

Table 2 Logistic regression predicting healthier option selection (socioeconomic status variable: highest educational qualification)

		Odds Ratio (95% CIs)	p-value
Availability condition (ref: Equal Healthier and Less-Healthy)	Increased Less-Healthy	0.34 (0.20, 0.60)	< 0.001
	Increased Healthier	2.49 (1.51, 4.10)	< 0.001
Education (ref: Lower: Up to GCSE)	Higher: Degree or above	1.10 (0.70, 1.70)	0.688
Age		1.01 (0.99, 1.03)	0.279
Gender (ref: Female)	Male	0.57 (0.37, 0.89)	0.013
Hunger^a		0.80 (0.65, 0.99)	0.036
Intercept		0.78 (0.38, 1.82)	0.572

Pseudo R-squared = 0.1267; Log-likelihood chi-square (degrees of freedom: 6) = 71.17 (p < 0.0001); Number of observations = 410 [5 participants did not report age; 2 did not report hunger – 2 of whom were allocated to the Increased Less-Healthy condition, 4 to the Equal Healthier and Less-Healthy condition, and 1 to the Increased Healthier condition]
^a 7-point rating scale: 0 was labelled “Neither hungry nor full”; 1 was labelled “A little hungry”

ISSN: 1471-2458