Early-childhood linear growth faltering in low- and middle-income countries

Stunting, defined as height more than two standard deviations below international growth standards, affects an estimated 149 million children under 5 globally. This condition is associated with increased mortality, illness, impaired cognitive development, and reduced adult productivity. Current global stunting estimates rely on cross-sectional surveys, which cannot accurately capture the timing of growth faltering onset or the persistence of growth deficits. Understanding these longitudinal patterns is crucial for designing effective preventive interventions. The first 1000 days of life (including the prenatal period) are considered a critical window for intervention, but the specific timing of interventions within this period remains unclear. This study aimed to address this knowledge gap by analyzing longitudinal data to determine the typical age of stunting onset and investigate the recurrence of stunting in early life. This granular information is critical for informing the timing and type of interventions needed to prevent linear growth faltering.

Existing literature highlights the association between early-life stunting and adverse health and developmental outcomes. Cross-sectional studies using Demographic and Health Surveys (DHS) data have provided estimates of age-specific stunting prevalence but lack the longitudinal perspective needed to understand the dynamics of growth faltering. While some studies have examined age-specific incidence, few have focused on the critical period of the first two years of life. Furthermore, the extent to which catch-up growth is possible after early-life stunting is not fully understood and likely depends on the timing and severity of the growth faltering.

This study conducted a pooled analysis of 32 longitudinal cohorts from 14 LMICs in South Asia, sub-Saharan Africa, Latin America, and Eastern Europe. Data from these cohorts, collected between 1987 and 2017, were aggregated by the Bill & Melinda Gates Foundation Knowledge Integration (Ki) initiative. The cohorts included children aged 0-24 months with repeated length and weight measurements at least every 3 months. Length-for-age z-scores (LAZs) were calculated using WHO 2006 growth standards. Stunting was defined as LAZ < -2, and severe stunting as LAZ < -3. Age-specific incidence of stunting was determined by classifying children as new incident cases in three-month age periods if their LAZ fell below -2 for the first time. Stunting reversal and relapse were also analyzed. Random-effects models were used for pooled analyses, with separate analyses conducted by geographic region. The representativeness of the Ki cohorts was assessed by comparing LAZ distributions with contemporary DHS data from the same countries.

The analysis revealed that the highest incidence of stunting onset occurred between birth and 3 months of age. Stunting prevalence at birth ranged from 0.3% to 42% across cohorts, with an overall prevalence of 13%. Incident stunting onset between birth and 3 months ranged from 7% to 57%, accounting for 23% of all children stunted by 24 months. Stunting reversal was rare (less than 7% per month after 6 months of age), and relapse rates were substantially higher among children stunted at birth. South Asia showed the highest burden of early-life stunting, with 20% of children stunted at birth and another 21% becoming stunted by 3 months. Even children who never met the criteria for stunting experienced a significant decline in mean LAZ (approximately 0.5 z) from birth to 15 months, indicating that linear growth faltering is a whole-population phenomenon. Linear growth velocity was significantly lower in the first 3 months compared to later ages, reflecting the high incidence of stunting during this period. Analyses stratified by birth LAZ showed that children stunted at birth were more likely to experience stunting relapse. Improvements in LAZ following stunting reversal were neither sustained nor substantial enough to erase growth deficits.

The findings demonstrate that linear growth faltering begins very early in life, primarily during the prenatal and early postnatal periods, before most postnatal interventions are implemented. This underscores the critical importance of the first 1000 days but emphasizes the need for a stronger focus on prenatal and early postnatal interventions. The high incidence of stunting in South Asia highlights the need for tailored interventions addressing factors influencing maternal nutrition and prenatal health in this region. The limited effectiveness of postnatal interventions to improve linear growth motivates a search for alternative approaches, potentially multisectoral, targeting upstream risk factors and improving implementation of existing strategies.

This study provides strong evidence for the early onset and persistence of linear growth faltering. Targeting interventions to the first three months of life and focusing on improving maternal health and nutrition, especially in high-burden regions like South Asia, are crucial for reducing stunting. Further research is needed to identify effective interventions for the first six months of life, and to better understand the mechanisms underlying stunting relapse.

Potential limitations include measurement error in length estimates, overestimation of stunting in preterm infants due to limitations in gestational age data, and the potential for selection bias due to the characteristics of the included cohorts. The study primarily focuses on the first two years of life, limiting the analysis of long-term outcomes and catch-up growth.