Session: Early-Life Influences on Health and Well-Being Across the Life Course
Tobacco smoking and vaping in pregnancy and neonatal outcomes: results from a UK cohort study (Anthony Laverty, Imperial College London)
We examined associations of smoking and vaping during pregnancy with infant outcomes using caregiver reported data for 1,896 babies in the UK Early Life Cohort Feasibility Study (9–10 months). Exposure (smoking/vaping: yes/never) formed four groups. Models adjusted for sociodemographic factors and survey weights. Most mothers (90.1%) neither smoked nor vaped; 2.6% vaped only, 5.5% smoked only, and 1.8% were dual users. Mean birthweight was 3.31kg for unexposed babies and 3.00kg for those of mothers who smoked. Adjusted analyses showed lower birthweight for babies of smokers (-0.34kg) and dual users (-0.39kg). Smoking showed dose response associations with lower birthweight and LBW; vaping showed none. Dual use was associated with longer postnatal hospital stay, while smoking or vaping alone were not. No exposure group differed in neonatal admissions.
Breastfeeding and child cognitive outcomes: Evidence from a hospital-based breastfeeding support policy (Nuno Braz, University of Reading)
This paper estimates the causal effect of breastfeeding on child cognitive development using the Millennium Cohort Study. We exploit variation in hospital-level implementation of the UNICEF Baby Friendly Initiative and instrument breastfeeding with proximity to certified hospitals. The policy increases breastfeeding initiation and duration. Instrumental variable estimates show positive and persistent effects of breastfeeding on a wide range of cognitive outcomes from early childhood into adolescence. Effects are strongest among disadvantaged children. We also find effects on maternal labour market participation, childcare use, and reading to the child for disadvantaged mothers, suggesting potential mechanisms.
Understanding health and premature mortality in men and women, role of social capital in inequality of opportunity (Alexia Bigorne, Institut de recherche santé société)
This paper considers differences in life expectancy between men and women through the lens of inequality of opportunity. Are these differences explained by differences in childhood socio-economic background (called circumstances) and considered as illegitimate because outside the individual’s choice or are they explained by differences in social capital throughout life (called effort). The National Child Development Study (NCDS) is unique in following 17 500 individuals in Great Britain from their birth in March 1958. The advantage of such a cohort is that members have experienced the same macroeconomic conditions over time. Moreover, it collects information on individuals’ socio-economic background and social capital from birth and over their lifetime. This study measures the relationship between individuals’ family socioeconomic background and social capital during adulthood, in relation to their premature death, if applicable (by the age of 62, already 21% of the cohort has died). We performed a Shapley decomposition at different stages in an individual’s life (23-33 years, 42 years, 44-46 years, 50 years, 55 years, and 62 years) to assess the contributions of circumstances, social capital, and gender to death inequalities.
The family socio-economic background is proxied by the father’s social class, the parents’ outings with the child, and whether the child shared a bedroom. Individual social capital is measured by participation in social relationships (participation in organisations and visiting friends) and by how this participation is perceived by the individual (trust in those relationships). (De Silva 2005) We added a proxy of support with the individual marital status. Preliminary results highlight a major contribution of the father’s outings with the child in premature deaths inequalities, early premature deaths being explained by the level of trust, visiting friends and being married. Finally, mothers’ outings with the child seemed to contribute more among men than among women.
To what extent do hypothetical interventions on the early childhood environment reduce inequalities in child and adolescent mental health? (Yu Wei Chua, University of Liverpool)
To inform early childhood policy, we quantified how socioeconomic inequalities in mental health would change under hypothetical interventions on the early childhood environment. We analysed parent-reported data from the UK Millennium Cohort Study (n=15513), on socioeconomic status at 9 months (maternal education), mental health at 5 and 17 years (total difficulties on the Strengths and Difficulties Questionnaire), and mediators at 9 months and 3 years in the neighbourhood, social support, home environment, adversity, and parent-child relations pathways. We estimated the change in mental health inequality (Low versus High SEC group), when mediator distributions are shifted from the Low SEC to the High SEC distribution. Mental health inequalities could reduce by one-third (5 years: -32%[-40;-23]; 17 years: -35%[-48;-26]), mostly attributed to parent-child relations and social support.
