Skip to main content

Table 3 Relative risk of purchasing the consequent food category given the purchasing of the antecedent food category estimated by Association Rule Mining and confirmatory logistic regression models with random intercepts

From: Characterizing co-purchased food products with soda, fresh fruits, and fresh vegetables using loyalty card purchasing data in Montréal, Canada, 2015–2017

Antecedent

Consequent

Support

RR (95%CI) from

ARM

RR (95%CI) from

Regression

Soda

Salty snacks

3.20%

2.07 (2.06, 2.09)

1.54 (1.52, 1.57)

Soda

Sweet snacks/candies

3.17%

1.73 (1.72, 1.74)

1.20 (1.18, 1.22)

Soda

Juices/drinks

2.73%

1.71 (1.71, 1.73)

1.27 (1.25, 1.29)

Soda

Water

1.50%

1.98 (1.96, 2.01)

NA

Fresh vegetables

Pre-packaged salads/stir fries

6.85%

3.78 (3.74, 3.82)

2.20 (2.16, 2.24)

Fresh vegetables

Canned vegetables

6.08%

2.98 (2.94, 3.01)

1.63 (1.61, 1.66)

Fresh vegetables

Deli cheese

5.74%

3.00 (2.97, 3.04)

1.47 (1.44, 1.49)

Fresh vegetables

Fresh herbs

3.44%

6.56 (6.43, 6.70)

NA

Fresh fruits

Pre-packaged salads/stir fries

6.22%

2.79 (2.76, 2.81)

1.67 (1.64, 1.70)

Fresh fruits

Cereals

5.71%

2.56 (2.20, 2.58)

1.45 (1.42, 1.47)

Fresh fruits

Yogurt

11.47%

2.59 (2.58, 2.61)

1.54 (1.52, 1.56)

Fresh fruits

Nuts/seeds/dried fruits

3.64%

2.72 (2.69, 2.76)

NA

  1. Abbreviations; Relative Risk: RR, ARM: Association Rule Mining
  2. For the regression analysis, the outcome variable was the binary purchasing status of the consequent food category, and the exposure variable was the antecedent food category
  3. NA indicates the co-purchasing association not investigated by the regression analysis due to a smaller value of relative risk or support compared to other food categories