Akanksha A Marphatia ^1,2* Jonathan CK Wells ¹ Alice Reid ³ Aboli Bhalerao ⁴ Chittaranjan S Yajnik ⁴

• ¹ Great Ormond Street Institute of Child Health, Faculty of Population Health Sciences, University College London, London, United Kingdom
• ² Department of Geography, University of Cambridge, Cambridge, England, United Kingdom
• ³ Department of Geography, Faculty of Earth Sciences and Geography, School of Physical Sciences, University of Cambridge, Cambridge, England, United Kingdom
• ⁴ Diabetes Unit, King Edward Memorial Hospital Research Centre, Pune, Maharashtra, India

Introduction Maternal nutrition promotes maternal and child health. However, most interventions to address undernutrition are only implemented once pregnancy is known, and cannot address broader risk factors preceding conception. Poverty and socio-economic status are considered systemic risk factors, but both economic growth and cash transfers have had limited success improving undernutrition. Another generic risk factor is low human capital, referring to inadequate skills, knowledge and autonomy, and represented by traits such as low educational attainment and women's early marriage. Few studies have evaluated whether maternal human and socio-economic capital at conception are independently associated with maternal and offspring outcomes. Methods Using data on 651 mother-child dyads from the prospective Pune Maternal Nutrition Study in rural India, composite markers were generated of 'maternal human capital' using maternal marriage age and maternal and husband's education, and 'socio-economic capital' using household wealth. Linear and logistic regression models investigated associations of maternal low/mid human capital, relative to high capital, with her own nutrition and offspring size at birth, postnatal growth, education, age at marriage and reproduction, and cardiometabolic risk at 18 years. Models controlled for socio-economic capital, maternal age and parity. Results Independent of socio-economic capital, and relative to high maternal human capital, low human capital was associated with shorter maternal stature, lower adiposity and folate deficiency but higher vitamin B12 status. In offspring, low maternal human capital was reflected in shorter gestation, smaller birth head girth, being breastfed for longer, poor postnatal growth, less schooling, lower fat mass and insulin secretion at 18 years. Daughters married and had children at an early age. Discussion Separating maternal human and socio-economic capital is important for identifying the aspects which are most relevant for future interventions. Low maternal human capital, independent of socio-economic capital, was a systemic risk factor contributing to an intergenerational cycle of disadvantage, perpetuated through undernutrition, low education and daughters' early marriage and reproduction. Future interventions should target maternal and child human capital. Increasing education and delaying girls' marriage may lead to sustained intergenerational improvements across Sustainable Development Goals 1 to 5, relating to poverty, hunger, health, education and gender equality.