Table 3 Effect of intervention on growth at different follow-up times.

From: A comprehensive intervention package improves the linear growth of children under 2-years-old in rural Bangladesh: a community-based cluster randomized controlled trial

 

Treatment

Control

Unadjusted

Adjustedc

ICC

Coefa/ORb (95% CI)

P-value

Coef/OR (95% CI)

P-value

8-month follow-up (treatment n = 191, control n = 175)

Length (cm)

78.79 (3.43)

77.17 (3.14)

1.62 (0.7, 2.5)

0.001

0.95 (0.5, 1.4)

 < 0.001

0.023

LAZ

− 1.01 (0.96)

− 1.17 (0.89)

0.16 (− 0.1, 0.4)

0.225

− 0.01(− 0.1, 0.1)

0.886

0.000

Weight (kg)

10.32 (1.75)

9.84 (1.41)

0.47 (0.1, 0.9)

0.023

0.53 (0.5,0.7)

 < 0.001

0.000

WAZ

− 0.36 (1.27)

− 0.51 (1.10)

0.15 (− 0.1, 0.4)

0.290

0.41 (0.3, 0.6)

 < 0.001

0.000

Stunting n (%)

30 (15.71)

30 (17.14)

0.90 (0.5, 1.6)

0.712

0.85 (0.5, 1.6)

0.624

0.008

10-month follow-up (treatment n = 197, control n = 184)

Length (cm)

81.14 (3.53)

79.15 (3.14)

1.98 (1.1, 2.9)

 < 0.001

1.42 (1.0, 1.9)

 < 0.001

0.027

LAZ

− 0.84 (1.03)

− 1.15 (0.90)

0.32 (− 0.4, 0.7)

0.024

0.21 (0.1, 0.4)

0.003

0.000

Weight (kg)

10.72 (1.87)

10.25 (1.45)

0.47 (0.1, 0.9)

0.020

0.45 (0.2,0.7)

 < 0.001

0.019

WAZ

− 0.36 (1.30)

− 0.52 (1.07)

0.15 (− 0.1, 0.4)

0.252

0.32 (0.1, 0.5)

0.001

0.017

Stunting n (%)

24 (12.18)

27 (14.67)

0.80 (0.4, 1.5)

0.490

0.79 (0.4, 1. 6)

0.498

0.012

12-month follow-up (treatment n = 191, control n = 183)

Length (cm)

83.52 (3.40)

80.89 (3.19)

2.61 (1.9, 3.4)

 < 0.001

2.05 (1.6, 2.5)

 < 0.001

0.000

LAZ

− 0.62 (0.95)

− 1.16 (0.92)

0.54 (0.3, 0.8)

 < 0.001

0.37 (0.2, 0.5)

 < 0.001

0.000

Weight (kg)

11.17 (1.94)

10.57 (1.47)

0.60 (0.2, 1.0)

0.001

0.62 (0.4, 0.9)

 < 0.001

0.000

WAZ

− 0.34 (1.32)

− 0.56 (1.09)

0.22 (− 0.03, 0.5)

0.083

0.44 (0.3, 0.6)

 < 0.001

0.000

Stunting n (%)

12 (6.30)

33 (18.00)

0.30 (0.2, 0.6)

0.001

0.27 (0.1, 0.6)

0.001

0.000

  1. Averages are presented as mean and standard deviation;
  2. aIn the first model co-efficient were estimated using mixed effect linear regression; unadjusted and adjusted analysis was accounted for clustering.
  3. bOdds ratio were estimated using mixed effect logistic regression; unadjusted and adjusted analysis was adjusted for clusters.
  4. cAdjusted for group, age, sex, child's baseline length, child's baseline weight, mother's height, and household asset index. Backward elimination stepwise covariate selection procedure was used; the models retained covariates with a P value of < 0.20 for the overall significance of the variables. No interaction terms with intervention were significant.
Back to article page