Table 2 Meal parameters, baseline weight and growth from two to five years (n = 1939).

From: Meal size is a critical driver of weight gain in early childhood

MEAL PARAMETER

 

Two year weight a

Two year weight SDS a

Growth rate (g/wk) b (coefficients of interactions with age)

ß (SE)

p-value c

ß (SE)

p-value d

ß (SE)

% growth increase e

p-value f

Meal size (10 kcals per eating occasion) g

Separate models

21 (7)

0.002

0.016 (0.005)

0.002

1.5 b (0.5)

4.0c

0.005

Mutual adjustment models

33 (8)

<0.001

0.024 (0.062)

<0.001

2.6 b (0.6)

7.3c

<0.001

Meal frequency (meals per day)

Separate models

3 (35)

0.93

0.001 (0.026)

0.967

0.3 (0.3)

0.9d

0.20

Mutual adjustment models

95 (41)

0.02

0.067 (0.03)

0.03

1.0 (0.3)

2.9d

0.001

  1. Abbreviations: ß = unstandardized coefficient, SE = standard error, SDS = Standard Deviation Score.
  2. aAnalyses have been adjusted for sex, gestational age, birth weight, difference in age between diet diary completion and weight measurement as potential confounders.
  3. bAnalyses have been adjusted for sex, gestational age, birth weight and weight at 24 months of age as potential confounders. The intra-class correlations at the family and twin levels were 0.39 and 0.47 respectively. At the family level this value represents the between family variance in weight as a proportion of the total variance in weight. At the twin level the value represents the between child variance in weight as a proportion of the total variance in weight over repeated measurement occasions. The random portions included in the longitudinal model were the intercept at the family and intercept and slope of age at the twin levels.
  4. cp-value for significance of coefficient: associations between 2 ycoefficient: associations between 2 year weight and meal parameters.
  5. dp-value for significance of coefficient: associations between 2 ycoefficient: associations between 2 year weight SDS and meal parameters.
  6. e% growth increase in addition to the mean base growth rate (36 g/wk) was calculated by dividing the B coefficient by the mean growth rate (36 g/wk) and multiplying by 100.
  7. fp-value for significance of B coefficient: interactions between meal parameters and age.
  8. gcoefficient: associations between 2 y coefficient has been re-scaled by multiplying by 10 (per 10 kcals); for each 10 kcals increase in meal size a child’s weight at 2 years would be 21 g higher and growth rate would increase by 1.5 g/week in addition to the mean base growth rate (36 kg/wk).
Back to article page