Skip to main content
. 2016 Nov 8;20(4):660–670. doi: 10.1017/S1368980016002640

Table 4.

Multiple logistic regression models examining the associations between perinatal data, parental anthropometrics, family sociodemographic and children’s dietary and physical activity indices (independent variables) and increased total and visceral fat mass levels (dependent variables) among 9–13-year-old children from four regions of Greece; Healthy Growth Study

Dependent variable
Model 1: total fat mass (third tertile) Model 2: visceral fat mass (third tertile)
Independent variable OR 95 % CI OR 95 % CI
Age category
9–11 years 1·00 1·00
11–13 years 1·31 0·95, 1·78 1·18 0·86, 1·62
Mother’s pre-pregnancy weight status
Normal weight 1·00 1·00
Underweight 0·59 0·30, 1·19 0·62 0·30, 1·27
Overweight 1·31 0·83, 2·08 1·53 0·96, 2·44
Obese 1·39 0·58, 3·33 2·46 1·02, 5·90
Maternal smoking during pregnancy
No smoking 1·00 1·00
Smoking at any one or two trimesters 1·52 0·85, 2·71 0·81 0·43, 1·53
All trimesters 1·72 1·06, 2·80 1·93 1·19, 3·13
Weight gain in the first 6 months
Average (Z-score change=−1 to +1) 1·00 1·00
Poor (Z-score change<−1) 0·98 0·60, 1·59 1·01 0·61, 1·67
Rapid (Z-score change >+1) 1·42 1·03, 1·97 1·96 1·41, 2·72
Timing of solid food initiation
≤4 months 1·00 1·00
5–6 months 1·18 0·78, 1·79 1·43 0·92, 2·22
>6 months 1·41 0·82, 2·42 1·72 0·98, 3·03
Father’s age
<42 years 1·00 1·00
42–46 years 0·83 0·58, 1·18 0·83 0·57, 1·20
>46 years 0·57 0·39, 0·83 0·81 0·55, 1·19
Mother’s BMI
Normal weight 1·00 1·00
Overweight 1·31 0·91, 1·88 1·28 0·88, 1·86
Obese 3·03 1·76, 5·23 1·77 1·01, 3·10
Father’s BMI
Normal weight 1·00 1·00
Overweight 1·24 0·84, 1·83 1·32 0·88, 1·98
Obese 1·62 1·04, 2·54 1·78 1·12, 2·83
Paternal education
<9 years 1·00 1·00
9–12 years 1·31 0·87, 1·96 1·08 0·72, 1·62
12–16 years 1·19 0·73, 1·94 1·09 0·66, 1·78
>16 years 1·83 0·96, 3·49 1·24 0·63, 2·44
Maternal education
<9 years 1·00 1·00
9–12 years 0·62 0·41, 0·94 0·67 0·44, 1·03
12–16 years 0·53 0·33, 0·86 0·58 0·36, 0·94
>16 years 0·58 0·29, 1·16 0·45 0·22, 0·95
Television time
<2 h/d 1·00 1·00
>2 h/d 1·40 1·03, 1·90 1·34 0·98, 1·82
Physical activity level
<12 000 steps/d (girls), <13 000 steps/d (boys) 1·00 1·00
≥12 000 steps/d (girls), ≥13 000 steps/d (boys) 0·66 0·49, 0·89 0·47 0·34, 0·64
Breakfast consumption
No 1·00 1·00
Yes 0·83 0·60, 1·14 0·81 0·59, 1·12

Bold font indicates statistically significant OR (P<0·05).

Adjustment was made for dietary energy intake.

Adjusted OR and 95 % CI were derived from multivariate logistic regression models, i.e. with its model including one perinatal, parental anthropometric, family sociodemographic or children’s dietary or physical activity variable as independent variable and highest total or visceral fat mass level as dependent variable.

As no specific and widely used cut-off points for total and visceral fat mass are available, the third tertiles of children with the highest levels with total and visceral fat mass were used as independent variables.

Based on recommendations by the Institute of Medicine( 13 ).