Abstract
Background: High rates of child malnutrition are a major public health concern in developing countries, particularly among vulnerable communities. Midday meals programs can be effective for combatting childhood malnutrition among older children. However, their use in early childhood is not well documented, particularly within South Asia. Anthropometric measures and other socioeconomic data were collected for children below the age of five years living in selected Sri Lankan tea plantations, to assess the effectiveness of midday meals as a nutrition intervention for improving growth among young children.Methods: The study exploits a natural experiment whereby the provision of the midday meals program is exogenously determined at the plantation level, resulting in comparable treatment and control groups. Longitudinal data is regularly collected on heights and weights of children, between 2013-2015. Standardized weight-for-age, height-for-age, BMI-for-age and weight-for-height are calculated following WHO guidelines, and binary variables for stunting, wasting and underweight are constructed. All modelling uses STATA SE 15. Random-effects regression with instrumental variables is used for modelling standardized growth variables whilst random-effects logistic regression is used for binary outcome variables. Robustness analysis involves different estimation methods and subsamples. Results: The dataset consists of longitudinal data from a total of 1279 children across three tea plantations in Sri Lanka, with 799 children in the treatment group and 480 in the control group. Results show significant positive effects of access to the midday meals program, on the growth of children. A child with access to the midday meals intervention reports an average standardized weight-for-age 0.03 (±0.01) and height-for-age 0.05 (±0.01) units higher than a similar child without access to the intervention. Importantly, access to the intervention reduces the likelihood of being underweight by 0.45 and the likelihood of wasting by 0.47. The results are robust to different model estimations and across different subsamples by gender, birthweight and birth-year cohort. Qualitative data analysis suggests a high viability of implementing similar programs within tea plantations in Sri Lanka.Conclusions: Midday meals programs targeting early childhood can be an effective intervention to address high rates of child malnutrition, particularly among vulnerable communities in developing countries like Sri Lanka.
Bayesian Estimation of Cox Models with Non-Nested Random Effects: An Application to the Ratification of ILO Conventions by Developing Countries
