Identifying developmental trajectories of body mass index in childhood using latent class growth (mixture) modelling: associations with dietary, sedentary and physical activity behaviors: a longitudinal study

Table 3 Health related behaviors in the BMI SDS trajectories

Variables of interest [%]	BMI trajectories			P-value^a
Variables of interest [%]	Increasing (N = 197)	Decreasing (N = 416)	Total sample (N = 613)
Dietary behaviors
Vegetable intake < 5 days a week
2006	60.8	55.5	57.1
2012	52.8	59.8	57.4
Fruit < 2 portions a day
2006	72.5	72.9 ***^b	72.8 ***
2012	76.0	82.2 ***	80.4 ***
Sugared drinks > 3 glasses a day
2006	54.0 ***	46.0 ***	48.6 ***
2012	40.2 ***	33.8 ***	35.9 ***
Physical activity behaviors
Outside play < 1 h a day
2006	53.6 (4.2)	53.3 (5.8)	53.4 (5.4)
2012	71.8 (8.1)	62.8 (5.1)	65.7 (7.5)
Organized sports < 1 h a week
2006	69.1 ***	66.5 ***	67.3 ***
2012	20.9 ***	14.7 ***	16.7 ***	*
Sedentary behaviors
TV viewing > 2 h a day
2006	4.1 ***	2.8 ***	3.2 ***
2012	14.9 ***	9.9 ***	11.5 ***
Screentime > 2 h a day
2006	6.2 ***	5.6 ***	5.8 ***
2012	65.0 ***	57.2 ***	59.7 ***

Abbreviations: BMI body mass index (calculated as weight in kilograms divided by height in metres squared), SDS standard deviation scores
^aP-values determined using X² test : *p < 0.05 for difference between trajectories
^bP-values determined using X² test: ***p <0.001 for differences in health related behaviors between baseline and follow-up

ISSN: 1471-2458