School-based gardening, cooking and nutrition intervention increased vegetable intake but did not reduce BMI: Texas sprouts - a cluster randomized controlled trial

Table 1 Differences in child demographic, clinical, and dietary characteristics between Texas Sprouts intervention and control groups at baseline

	Total Mean (SE)	Intervention Mean (SE)^a	Control Mean (SE)^a	P-value**
Sample size
CCA	2721	1212	1509	NA
MI	3135	1412	1723	NA
Age (y)
CCA	9.24 (0.07)	9.28 (0.07)	9.20 (0.07)	0.33
MI	9.23 (0.02)	9.26 (0.06)	9.20 (0.06)	0.47
Male (%) (SE)
CCA	47.16 (3.88)	46.42 (4.08)	47.89 (3.67)	0.45
MI	47.35 (3.61)	46.99 (3.79)	47.71 (3.44)	0.59
Race/ethnicity % (SE)
White
CCA	19.82 (2.75)	21.28 (2.76)	18.37 (2.73)	0.86
MI	19.69 (2.58)	20.90 (2.67)	18.48 (2.51)	0.87
Black
CCA	8.91 (2.22)	8.61 (2.23)	9.21 (2.20)	0.91
MI	11.10 (2.23)	11.19 (2.32)	11.02 (2.16)	0.85
Hispanic
CCA	65.80 (3.46)	64.75 (3.46)	66.86 (3.45)	0.92
MI	63.67 (3.25)	62.23 (3.29)	64.42 (3.21)	0.89
Nat.Amer/Asian/Pac.Island/Other
CCA	5.47 (1.72)	5.37 (1.79)	5.57 (1.65)	0.89
MI	5.88 (1.64)	5.68 (1.69)	6.08 (1.61)	0.75
Eligible FRL %(SE)
CCA	68.91 (3.55)	66.62 (3.69)	71.21 (3.41)	0.68
MI	68.25 (3.21)	66.47 (3.38)	69.43 (3.07)	0.78

CCA complete case analyses; MI multiple imputation; FRL free and reduced school lunch; Nat. Amer Native American; Pac Island Pacific Islander
MI analyses is required for all clinical trials and CCA is what is more commonly reported, therefore, both MI and CCA are reported here
^a Weighted mean and standard error (SE) of each variable. Weighted refers to the total number of children within each school
**Generalized linear mixed models (GLMM) with linear link was used to compute p-values of the continuous variables, GLLM with cumulative logit link was used to compute p-values of the categorical variables

ISSN: 1479-5868