Table 6 Significant variables in regression analysis

Estimated spending (outcome) and local area prescription policy (main explanatory variable) (n = 1074)

R²_adj 0.11, p < 0.001

Spending on GFSF was higher

(i) in participants living in non-prescribing areas (p < 0.001)

(ii) in participants who were just able to afford to buy GFSF (p < 0.001)

(iii) in males (p < 0.001)

(iv) where there were additional household members with coeliac disease (p < 0.001)

Estimated spending (outcome) and self-report prescription status (main explanatory variable) (n = 1074)

R²_adj 0.11, p < 0.001

Spending on GFSF was higher

(i) in participants living in non-prescribing areas (p < 0.001)

(ii) in participants who were just able to afford to buy GFSF (p < 0.001)

(iii) in males (p < 0.001)

(iv) where there were additional household members with coeliac disease (p < 0.001)

Spending on GFSF was lower

(i) in participants who are separated/divorced (p = 0.028)

Self-reported spending (outcome) and local area prescription policy (main explanatory variable) (n = 1074)

R²_adj 0.10, p < 0.001

Spending on GFSF was higher

(i) in participants living in non-prescribing areas (p < 0.001)

(ii) in participants who were just able to afford to buy GFSF (p < 0.001)

(iii) in males (p < 0.001)

(iv) where there were additional household members with coeliac disease (p = 0.003)

Spending on GFSF was lower

(i) as age increase (p < 0.001)

(ii) in socioeconomic domain of supervisory or clerical (p = 0.017) and skilled manual work (p = 0.042)

Self-reported spending (outcome) and self-reported prescription status (main explanatory variable) (n = 1074)

R²_adj 0.10, p < 0.001

Spending on GFSF was higher

(i) in participants living in non-prescribing areas (p < 0.001)

(ii) in participants who were just able to afford to buy GFSF (p < 0.001)

(iii) in males (p < 0.001)

(iv) where there were additional household members with coeliac disease (p = 0.002)

Spending on GFSF was lower

(i) as age increase (p < 0.001)

(ii) in socioeconomic domain of supervisory or clerical (p = 0.016) and skilled manual work (p = 0.046)

Estimated spending (outcome) and local area prescription policy (main explanatory variable). Sample restricted to participants whose self-report prescription status matched with the local area prescription policy (n = 784)

R²_adj 0.16, p < 0.001

Spending on GFSF was higher

(i) in participants living in non-prescribing areas (p < 0.001)

(ii) in participants who were just able to afford to buy GFSF (p < 0.001)

(iii) in males (p < 0.001)

(iv) where there were additional household members with coeliac disease (p < 0.001)