The association between paternal labour migration and the growth of the left-behind children—evidence from a birth cohort in Dhanusha district, Nepal

TL;DR

This study finds that children, especially girls, of fathers who migrated for work in Nepal had slower growth, especially in early life.

Contribution

The study identifies gender-specific effects of paternal migration on child growth and highlights critical early-life windows.

Findings

Daughters of migrant fathers had lower height-for-age z-scores compared to non-migrant fathers.

Negative growth effects were strongest in children under 6 months old.

No significant association was found between remittances and child growth outcomes.

Abstract

We aimed to determine the association between paternal labour migration and the growth of the left-behind children in Dhanusha district, Nepal, where child stunting and international labour migration are highly prevalent.

We used growth data at birth, 6 months, 1 year and 2 years from a birth cohort study conducted 2012–2014, and growth data at age 6 years collected in 2018. We collected household migration history data to determine the children’s exposure to paternal migration. The primary outcome was child length/height-for-age z-score (HAZ). Children’s body circumferences, skinfold thicknesses, body composition, tibia length and grip strength were secondary outcomes measured at 6 years. We tested (i) the overall association between paternal international migration and the growth of the left-behind child; the roles of (ii) the duration of migration (≤12 mvs >12 m) and (iii) child age (≤6 mvs 12–72 m) as moderating factors; (iv) the association between receipt of remittances from the migrant father and child growth outcomes; and (v) stratified the main analyses by child gender. We fitted mixed-effects linear regression models for longitudinal data and linear regression models for cross-sectional data, adjusted for potential confounders.

Analysing across all time points, daughters of labour migrants had lower HAZ than daughters of non-migrants (−0.13, 95% CI −0.24 to –0.03), but no overall association was found in boys. The negative associations were largest at <6 m (girls: −0.23, 95% CI −0.41 to –0.05), but in boys only if the father had recently (≤12 m) migrated (−0.26, 95% CI −0.51 to 0.00). Children of migrants showed a tendency towards smaller body sizes compared with children of non-migrants. We found no association between remittances and any measure of child growth.

Interventions should target support for pregnant women and mothers with young infants to provide gender-equitable childcare, especially if their husband just left for work overseas.

Figures