OP22 Perinatal and Early Life Risk Factors for Becoming a Looked-After Child. A Population-Based Cohort Study Using Routinely Collected Data in Wales

Background The number of Looked-After Children (LAC) in local authority care in the UK has risen dramatically over recent years, with stark inequalities. These children are at higher risk to experience poor health and educational outcomes. Few studies, none in the UK, have used registry-based data at an individual level to assess how socioeconomic disadvantage and perinatal exposures compound in the LAC population. We therefore assessed how socio-demographic factors, maternal and child health are associated with being looked after in Wales. Methods We conducted a national linkage study in Wales using the Secure Anonymised Information Linkage (SAIL) Databank, which contains anonymised individual-level population-scale data sources from electronic health records and administrative data, including novel data on LAC. We built an electronic birth cohort of 276,505 children born in Wales between 2000–2018. Our outcome was being a looked-after child between 2016- 2018 . Exposures of interest were demographic factors at birth, including the Welsh Index of Multiple Deprivation (WIMD), child health at birth, and perinatal indicators of maternal health, including mental health and drug and alcohol use captured in the period 2 years prior to birth. We used logistic regression to assess the odds of being looked-after for the characteristics of interest. As a sensitivity analysis, we used the Townsend Deprivation Index in replacement of WIMD, validating our results. Results Between 2016–2018, 3,074 children in our population were LAC, with a steep social gradient: 52% were born to mothers from the most deprived WIMD quintile, compared to 4% from the least deprived (OR 4.5; 95%CI 3.7–5.5). Odds of being looked after were increased for children of mothers who have: been treated for common mental health problems (OR 1.7; 95%CI 1.5–1.8) and/or depression (OR 1.25; 95%CI 1.09–1.42); a diagnosed learning disability (OR 4.22; 95%CI 3.44–5.17); reported problematic drinking (OR 2.43; 95%CI 1.77–3.34), smoking (OR 1.67; 95%CI 1.54–1.80) and substance misuse (OR 5.79; 95%CI 4.89–6.85). Odds of being looked after were increased for children with developmental delay (OR 2.99, 95%CI 2.73–3.28) and congenital anomalies (OR 1.74, 95%CI 1.24–2.45). Conclusion Half of LAC’s mothers in Wales lived in the most deprived quintile of WIMD, with serious risks associated with maternal health. We provide insights into the mechanisms of intergenerational transfer of disadvantage in one of the most vulnerable sections of the child population. Limitations included: LAC status currently only available for 2016–2018 and for children over age 5, and information bias in routinely collected data.