Peer-reviewed article

Beyond education and contraceptive use: do caste, ethnicity, and religion influence fertility behaviour in Nepal? Evidence from six rounds of DHS.

Dhruba Raj Ghimire, Samir Kc · 13 May 2026

Abstract

BACKGROUND: Nepal has experienced a substantial decline in fertility over the past three decades; however, notable heterogeneity persists across sociocultural groups. This study examines how constitutionally defined caste, ethnicity, and religion shape fertility behaviour after accounting for education, wealth, and structural determinants. By applying a repeated cross-sectional framework, the study provides new evidence on long-term patterns and shifting trajectories across population subgroups. METHODS: The study analysed data from six rounds of the Nepal Demographic and Health Survey (1996-2022), comprising currently married women aged 15-49. Separate analyses (individual period and pooled) were done for lifetime (15-49 years) and completed (40-49) fertility. A survey-adjusted Poisson regression model was used to estimate associations with the number of children ever born (CEB). Duration since marriage was included as the exposure variable to account for variation in reproductive risk time. A pooled interaction model (caste/ethnicity x survey year) assessed temporal changes in sociocultural fertility differentials. RESULTS: Women's educational attainment was the strongest and most consistent predictor of lower fertility, with women aged 40-49 with post-secondary education having substantially lower expected CEB than those with no schooling. Among women aged 15-49, wealth, met need for contraception, and urban residence were also associated with reduced fertility. Sociocultural differentials persisted after adjustment: Muslim women had higher expected CEB than Arya women across most survey years; Madheshi women exhibited rising fertility relative to Arya women in interaction models; Janjati women showed accelerated fertility decline; and the pooled effect for Dalit women was not statistically significant. CONCLUSION: Fertility heterogeneity in Nepal reflects the combined influence of sociocultural identity and structural conditions. While education is the key driver of fertility reduction, persistent differences among Muslim and Madheshi communities indicate the need for culturally tailored reproductive health strategies. Policies should prioritise universal completion of secondary education for girls, as well as equity-focused empowerment and family planning initiatives to address remaining disparities.