Peer-reviewed article

Changing patterns of family formation among internally displaced populations in Yemen: evidence from cross-sectional surveys.

Elnakib S, Zimmerman LA, Attal B, Alkebsi T, Aldaram A et al. · 3 Dec 2024

Abstract

Introduction Yemen has one of the highest rates of child marriage in the Middle East and North Africa region and is home to one of the world's worst humanitarian crises. How the conflict and resulting displacement have impacted family formation patterns is not entirely clear. In this study, we investigate the impact of displacement on child marriage and early childbearing by comparing time-to-first-marriage and time-to-first-birth among displaced and non-displaced girls. Methods We used data collected through cross-sectional multistage stratified cluster household surveys in three governorates in Yemen with high concentrations of internally displaced persons (IDPs). Employing an analytic sample 1,861 girls aged 15-24, we compared entry into first marriage and time to first birth between IDPs and non-displaced persons using Kaplan Meier curves and Cox regressions. Results We found high rates of child marriage in this population, with 37.8% and 34.5% of ever-displaced and never-displaced girls aged 20-24 married before age 18. Overall, those who were displaced had 30% higher hazard of overall marriage compared to their host counterparts (95% CI 1.16-1.47), and 23% increased hazard of child marriage (95%CI 1.01-1.51). Stratification by governorate revealed heterogeneity across governorates, with displaced populations at higher hazard of marriage in Aden and Maarib but not in Hadramout. We found that child brides and displaced participants were more likely to initiate earlier childbearing compared to non-child brides (HR = 3.44; 95% CI 2.94 4.0) and host counterparts (HR 1.35; 95% CI 1.15-1.58). Despite generally having lower age at first birth, child brides experienced a 19% lower hazard of first birth after marriage compared to those married over 18, indicating that child brides tended to wait longer before first birth (95% CI 0.72-0.92). We did not find a significant association between displacement status and marriage to birth interval after adjusting for child marriage status. Conclusion Our results highlight the impact of forced displacement on time-to-first-marriage and suggest that those who are displaced are more vulnerable to child marriage. We did not find compelling evidence for child marriage driving an increase in early childbearing. Nonetheless, our findings highlight the need for interventions that focus on prevention and mitigation of child marriage in this setting, particularly among IDPs who are at increased vulnerability.