Global health targets now emphasize all forms of malnutrition, including overweight and obesity, alongside undernutrition. The World Health Organization (WHO) aims to prevent increases in childhood overweight by 2025, and Sustainable Development Goal (SDG) 2.2.2 targets the elimination of all forms of malnutrition in children under 5. However, progress towards these goals is insufficient. Overweight and obesity are increasing globally, particularly in upper-middle- and high-income countries, due to factors like urbanization, economic growth, and lifestyle changes. While overweight prevalence is known to increase with age after 5 years, age patterns in younger children are less understood. Undernutrition, particularly wasting, remains a significant issue, affecting millions of children worldwide and linked to food insecurity, poor diet, and disease. Wasting prevalence is classically associated with the second year of life, but recent studies indicate higher rates in younger infants. The double burden of malnutrition (DBM), encompassing both undernutrition and overweight, is prevalent in LMICs. SDG 17.18 advocates for data disaggregation, including by age, but age patterns in wasting and overweight prevalence among under-5 children in LMICs remain understudied. This study aimed to describe these age patterns using data from national surveys.

Existing literature extensively documents the increasing global burden of childhood overweight and obesity, particularly in higher-income settings. Studies have linked this rise to nutritional transitions, urbanization, and lifestyle changes. The age trajectory of overweight after age 5 is well established, showing a progressive increase. Conversely, while the high prevalence of wasting in young children is recognized, less is known about the age-specific patterns of wasting prevalence in LMICs. The concept of a double burden of malnutrition—the coexistence of undernutrition and overweight—is increasingly acknowledged, especially in LMICs. However, comprehensive analyses exploring age-specific patterns of both overweight and wasting across a large number of LMICs have been lacking, forming the basis for this research.

This study utilized data from nationally representative Demographic and Health Surveys (DHS) and Multiple Indicator Cluster Surveys (MICS) conducted in LMICs between 2010 and 2019. The study included 90 surveys representing a substantial portion of LMICs. Anthropometric data—weight and length/height—were collected for children under 5 years, using standardized questionnaires and measurement tools. Overweight was defined as a weight-for-height Z-score (WHZ) more than 2 standard deviations above the 2006 WHO Child Growth Standards median; wasting was defined as a WHZ more than 2 standard deviations below the median. Data analyses involved descriptive statistics (prevalence estimates, means, and standard deviations of WHZ stratified by age), Chi-square tests to compare prevalence across age groups, logistic regression to assess linear trends in overweight and wasting with age, and fractional polynomials to model the relationship between mean WHZ and prevalence. All analyses accounted for the complex survey design, using sampling weights and adjusting for clustering. The study also examined the variability of WHZ standard deviations by age, exploring potential measurement error issues.

The analysis of 90 national surveys from LMICs revealed several key findings. First, the overall prevalence of overweight decreased significantly with age, from 6.3% in infants (0-11 months) to 3.0% in 4-year-olds. This pattern was consistent across all income groups, although the prevalence was substantially lower in low-income countries compared to upper-middle-income countries. Second, wasting prevalence was highest among infants (0-11 months) and showed a slight decrease between the first and second years of life, with minimal variation thereafter. Lower-middle-income countries showed consistently higher wasting prevalence across all age groups. Third, mean WHZ remained relatively stable across the first five years of life. However, the standard deviation of WHZ decreased significantly with age, being considerably higher for infants than for older children. This suggests that the observed decline in both overweight and wasting prevalence with age may be partially attributable to measurement error or rapid crossing of growth channels in infants. The analysis showed that for a given WHZ value, both overweight and wasting prevalence were significantly higher in infants than in older children, further supporting the potential role of measurement error or rapid growth channel crossing.

The findings challenge the initial hypothesis that overweight prevalence would increase with age among under-five children. The observed decline in overweight and wasting prevalence with age, despite stable mean WHZ, suggests a potential confounding factor, namely, the higher variability in WHZ measurements for infants. Measurement error is plausible given the challenges in accurately measuring length/height in very young children. A small error in measurement can significantly alter the WHZ value, affecting the classification of children as overweight or wasted. The concept of growth canalization, where children tend to maintain their growth percentile after the age of 2, also offers a potential explanation. The higher standard deviations observed among infants may reflect greater variability in growth trajectories during infancy, with rapid crossing of growth channels, leading to higher prevalence of both overweight and wasting in cross-sectional analyses. These findings highlight the importance of considering the limitations of using cross-sectional data and the potential impact of measurement error in analyzing nutritional status in young children.

This study demonstrates a decline in both overweight and wasting prevalence with age in under-five children from LMICs. The consistent mean WHZ values, coupled with the decreasing standard deviations across age groups, suggests that measurement error or rapid growth channel crossing in infants likely contributes to this pattern. This study underscores the need for longitudinal studies with high-quality anthropometric data to more accurately assess the age-specific trends in overweight and wasting in young children. Further research should focus on improving measurement techniques and employing longitudinal designs to better understand the true nature of these trends.

The study's limitations include an uneven distribution of available surveys across World Bank income groups and potential outdated national estimates for countries without recent surveys. The reliance on cross-sectional data restricts the ability to make definitive statements about individual growth trajectories and the true prevalence of overweight by age. The cross-sectional nature of the data may limit the ability to accurately measure the changes in prevalence of overweight over time. Future studies using longitudinal data would be beneficial.