Table 3 Empirical weights derived from multivariate linear regression analysis using all selected eating habits as independent variables: from the responses to FFQ2.

From: Development and preliminary validation of a prediction formula of sodium and sodium-to-potassium ratio based on multiple regression using 24-h urines

 

Prediction expression with significant variables†

FFQ2-all (n = 244)

FFQ2-Developping group (n = 126)

Intake behavior items only

With characteristics

Intake behavior items only

With characteristics

Sodium

Sodium-to-potassiumratio (mmol)

Sodium

Sodium-to-potassiumratio (mmol)

Sodium

Sodium-to-potassiumratio (mmol)

Sodium

Sodium-to-potassiumratio (mmol)

Intercept

2523

2.89

865

1.73

2133

2.39

620

1.14

Taste preference

47

0.18

145

0.21

185

0.20

278

0.25

Soy source use at the table

187

0.32

87

0.29

210

0.47

63

0.39

Noodle soup

199

–

104

–

98

–

6

–

Pickled vegetables

101

–

163

–

140

–

170

–

Number of bowls (miso soup)

161

–

170

–

198

–

194

–

Vegetables (quartile)

–

 − 0.18

–

 − 0.11

–

 − 0.18

–

 − 0.12

Fruits (quartile)

–

 − 0.18

–

 − 0.15

–

 − 0.18

–

 − 0.20

Milk Products (quartile)

–

 − 0.13

–

 − 0.13

–

 − 0.07

–

 − 0.06

Sex (1:male 2:female)

–

–

 − 529

 − 0.26

–

–

 − 509

 − 0.31

Age (continuous)

–

–

3

 − 0.01

–

–

3

0.00

BMI (continuous)

–

–

102

0.08

–

–

102

0.07

Use of hypertension medication

–

–

30

0.06

–

–

195

0.03

  1. In a regression analysis, independent variables was treated as follows: taste preference (1: very mild 2: mild 3: common 4: strong 5: very strong), soy source use at the table (1: unused 2: rarely 3: sometimes 4: almost always 5: always), noodle soup (1: drink little 2: drink 1/3 of a bowl 3: drink half of a bowl 4: drink 2/3 of a bowl 5: drink almost all), pickled vegetables (1: < 3/day 2: ≥ 3, < 7/day 3: ≥ 7, < 14/day 4: ≥ 14/day), number of bowls (miso soup) (1: 0.5/day 2: ≥ 0.5, < 1/day 3: ≥ 1, < 2/day 4: ≥ 2/day), vegetables, fruits, and milk products quartile (1: 1st quartile 2: 2nd quartile 3: 3rd quartile 4: 4th quartile), sex (1: male 2: female), age (continuous), BMI (continuous), use of hypertension medication (0: no 1: yes).
  2. The individual responses for each item were transformed accordingly into ordinal variables (or directly into continuous or nominal variables), then multiplied by the regression coefficients for each item. The sum of these and the value of the intercept was the predicted value.
  3. †Variables using prediction equations are significant variables (p < 0.05) at regression analysis.
Back to article page