RT Journal Article SR Electronic T1 OP103 Understanding social inequalities in child mortality: a population-based study in denmark and wales JF Journal of Epidemiology and Community Health JO J Epidemiol Community Health FD BMJ Publishing Group Ltd SP A50 OP A51 DO 10.1136/jech-2019-SSMabstracts.106 VO 73 IS Suppl 1 A1 ETC Lai A1 R Griffiths A1 A Akbari A1 SK Urhoj A1 AMN Andersen A1 DC Taylor-Robinson YR 2019 UL http://jech.bmj.com/content/73/Suppl_1/A50.2.abstract AB Background Children growing up in disadvantaged socioeconomic circumstances (SECs) have a higher risk of death. In this study, we aimed to examine whether perinatal factors mediate the relationship between SECs and child mortality.Methods We conducted national registry linkage studies in two countries, using data for 592,001 births in Denmark (from 2000 to 2014) and 646,303 births in Wales (from 2000 to 2016). Deaths up to age 15 years in Denmark and up to 16.5 years in Wales were identified using death registry data, whilst birth weight and gestational age were obtained using medical birth register data. SECs at child’s birth were measured using maternal education in Denmark, which was identified from the Integrated Database for Labour Market Research; whilst in Wales, SECs were assessed based on quintiles of Welsh Index of Multiple Deprivation (WIMD), which was obtained for the mother’s postcode from the Welsh Demographics Service Dataset. We built two Cox proportional hazard survival analysis models to estimate the effect of SECs on children mortality and mediation by the perinatal factors: model 1 is the baseline model adjusted for sex and year of birth, model 2 additionally adjusted for birth weight and gestational age. As a sensitivity analysis, we tested whether the effect of SECs on child mortality changes over time, i.e. we assessed the interaction term between SEC measures and birth year.Results We identified 2,664 deaths in Denmark and 2,987 deaths in Wales. In both countries, lower SECs was associated with a higher risk of child mortality [Denmark: Hazard ratio (HR) 2.78; 95% confidence interval (CI): 1.04 to 7.43 (comparing maternal education with lower secondary or lower and those with university degree); Wales: HR: 1.92; 95%CI: 1.56 to 2.36 (comparing whose mothers from most deprived quintile and those from least deprived quintile)]. After adjustment for perinatal factors, the associations were attenuated (Denmark: HR: 1.65, 95%CI: 0.62 to 4.41; Wales: HR: 1.66, 95%CI: 1.35 to 2.05). Sensitivity analysis did not show the effect of SECs on child mortality changes over time.Discussion Using data from over 1 million children across two countries, we showed that a substantial proportion of social inequality of child mortality could be explained by perinatal factors. Policies to reduce child mortality in both countries should therefore focus on improving maternal health before and during pregnancy, especially those under socioeconomic disadvantage. A limitation of this study is that SEC measures are not directly comparable across countries.