Longitudinal assessments of child growth: A six-year follow-up of a cluster-randomized maternal education trial

Abstract

Background & aims: Child growth impairments are rampant in sub-Saharan Africa. To combat this important health problem, long-term follow-up studies are needed to examine possible benefits and sustainability of various interventions designed to correct inadequate child growth. Our aim was to perform a follow-up study of children aged 60-72 months whose mothers participated in a two-armed cluster-randomized education intervention trial lasting 6 months in rural Uganda when their children were 6-8 months old with data collection at 20-24 and at 36 months. The education focused on nutrition, hygiene, and child stimulation.

Methods: We measured growth using anthropometry converted to z-scores according to WHO guidelines. We also included assessments of body composition using bioimpedance. We used multilevel mixed effect linear regression models with maximum likelihood method, unstructured variance-covariance structure, and the cluster as a random effect component to compare data from the intervention (receiving the education and routine health care) with the control group (receiving only routine health care).

Results: Of the 511 children included in the original trial, data from 166/263 (63%) and 141/248 (57%) of the children in the intervention and control group, respectively, were available for the current follow-up study. We found no significant differences in any anthropometrical z-score between the two study groups at child age of 60-72 months, except that children in the intervention group had lower (P = 0.006) weight-for-height z-score than the controls. There were no significant differences in the trajectories of z-scores or height growth velocity (cm/year) from baseline (start of original trial) to child age of 60-72 months. Neither did we detect any significant difference between the intervention and control group regarding body composition (fat mass, fat free mass, and total body water) at child age 60-72 months. Separate gender analyses had no significant impact on any of the growth or body composition findings.

Conclusion: In this long-term study of children participating in a randomized maternal education trial, we found no significant impact of the intervention on anthropometrical z-scores, height growth velocity or body composition.

Trial registration: Clinical Trials (clinical trials.gov) ClinicalTrials.gov ID NCT02098031.

Keywords: Body composition; Growth impairment; Growth velocity; Maternal education; Stunting; Sub-Saharan Africa.

Conflict of interest statement

Conflicts of interest The authors declare no conflict of interest.

Copyright © 2021 The Author(s). Published by Elsevier Ltd.. All rights reserved.

Figures

Fig. 1. Flow chart showing the enrollment of study participants into the original randomized trial and those attending the current follow-up study.

Fig. 2. Changes in growth z-scores from baseline (6−8 months) to 60−72 months.

(A) Height-for-age z-score (HAZ), (B) Weight-for-age z-score (WAZ), (C) Weight-for-height z-score (WHZ), (D) Mid-upper arm circumference-for-age z-score (MUACZ). The horizontal middle (yellow) bar represents mean change; green and red lines are upper and lower 95% confidence interval of the mean change, respectively. The line Y = 0 (black dashed line) indicates no change between the two time points. Each dot represents the value from one child.