Global Health
Seasonality of underweight among children 1-11 months old in Niger: a population-based analysis of repeated cross-sectional data from a cluster-randomized trial Brittany Peterson* Brittany Peterson Ahmed M. Arzika, MPH Abdou Amza, MD Ramatou Maliki, MSc Bawa Aichatou, BSc Ismael Mamane Bello, MSc Diallo Beidi, BSc Nasser Galo, MD Elodie Lebas, RN Catherine E. Oldenburg, ScD, MPH Travis C. Porco, PhD Benjamin F. Arnold, PhD, MPH Thomas M. Lietman, MD Kieran S. O’Brien, PhD, MPH
Malnutrition in children is a risk factor for mortality, and around 45% of deaths of children under 5 globally are linked to malnutrition. Niger has strong seasonal patterns in rainfall, which influences infectious disease transmission and food security. Studies on the seasonal effects of weight show that nutritional status may fluctuate by season related to factors such as rainfall, food insecurity, and time of harvests. This analysis used data from the cluster-randomized AVENIR trial which compared the effect of biannual distribution of azithromycin versus placebo on mortality in children 1-59 months old. A subset of 133,781 children aged 1-11 months were included in the analysis, and weight-for-age z-score (WAZ) was assessed every 6 months over 2 years in a population-based census. We pooled time of measurement data to four seasons: dry, pre-rainy, rainy, and post-rainy, and used season and months since harvest as the exposures. Linear regression was used to estimate the cluster-level mean difference in WAZ by season using the dry season as a reference with mean difference, P-values, and 95% confidence interval reported. Linear regression was also used to determine the cluster-level mean change in WAZ by months since the last harvest. Compared to the dry season, the post-rainy season had the largest change in WAZ with a -0.11 reduction (95% CI -0.15 to -0.07, P-value <0.001). WAZ was also lower during the pre-rainy season, -0.07 (95% CI -0.11 to -0.03, P-value <0.001), and during the rainy season, -0.07 (95% CI -0.12 to -0.02, P-value 0.003). WAZ increased substantially 3 months after the harvest by +0.39 (95% CI 0.26 to 0.53, P-value <0.001). Given the mortality risk associated with low weight, the seasonality of WAZ young children in this setting may have implications for the timing of child survival and malnutrition programs and results of this study can be used to guide the timing of future programs.
