Childhood interactions with social services and risk of poor health and social outcomes in adulthood: a population wide data linkage study

In 2015 around 5.5% of young people (aged 0-18 years) in Northern Ireland were known to social services as a child in need (n=23,834), with 2,875 in the care of the Health and Social Care Trusts (a "looked after" child). The "looked after" population in Northern Ireland is increasing, with a 31% increase between 2002 and 2014, despite a total population decrease of young people of 2.9%. In addition, Northern Ireland is experiencing a mental health crisis with the highest ever recorded prevalence of poor mental health in the population and the highest suicide rate in the UK. Mental health is a major concern, and half of children "looked after" by social services report some form of mental health difficulty compared to just 10% not in care. But very little is known as to what happens these children after they leave the care of social services. This is important as poor mental health in childhood can not only lead to poor mental health in adulthood but to physical health problems, employment problems, economic problems and relationship breakdown. It is vital to understand who is most at risk, so policy makers can target interventions to reduce childhood mental health difficulties and thus reduce poor adult health and social outcomes.

However, research involving children known to social services (which includes "looked after" children, children in need and children on the child protection register) is difficult due to their high mobility and difficulties surrounding gaining parental or informed consent. To study adult outcomes would involve long-term follow up which takes a lot of time and money. Some existing longitudinal studies have tried to study adult outcomes of children known to social services but only have very small samples within their study with limited information on why the individual entered care. Other survey based studies have asked adults to recall their experiences of social services as children but these are subject to a range of possible errors. Large scale longitudinal studies with detailed information on the circumstances surrounding entry to care, placement type and duration, and health and social outcomes after leaving care are critical to understanding which individuals are at an elevated risk of poor mental health outcomes and which individuals show resilience and prosper in difficult circumstances.

Northern Ireland is unique within the UK in that it has an integrated health and social care system, and therefore holds data centrally on all individuals known to social services since 1985, providing an optimal environment for the study of long-term outcomes. This project will utilise 30 years of de-identified social services data linked to Census returns, medication data and death records to create the UKs first historical, population-wide cohort studying the long-term social and mental health outcomes of children known to social services. The research will be in partnership with the Administrative Data Research Centre - Northern Ireland, one of four such centres across the UK, who will make the data anonymous and hold it in their secure data environment. This project represents one of the most ambitious record linkage studies undertaken in the UK and therefore I will be collaborating with a world renowned Scandinavian team from the University of Helsinki who have extensive experience in large scale register based studies and can provide help and advice on data complexity and analytical methods. The project will also provide a significant training opportunity, developing my skills in large scale data analysis.

Basically, this study will use already existing administrative data and link it to better understand the long-term mental health and social trajectories of children known to social services. The research will indicate potential predictors of both poor and good adult mental health and social outcomes which could be used to target resources and interventions toward those most in need.

Aims: To develop the UK's first historical, population-wide cohort of children in contact with social services by linking 30 years' of social services data to three waves of Census returns, prescription medication data and death records to investigate the effect of contact with social services in childhood on adult mental health and social outcomes.

Objectives: To first complete advanced training, building upon my existing skills set, to become an expert in the manipulation, management and analysis of large, longitudinal administrative datasets with a focus on life course analytical techniques and Multilevel Modelling. Then secondly, to investigate the effect of contact with social services in childhood on adult mental health and social outcomes via four specific research questions focusing on (1) changes over time, (2) determinants of poor outcomes and variation within families, by reason for entry to care or placement type, (3) critical periods, and (4) generational effects.

Methodology: The is an administrative data linkage study utilising 30 years of social services data enhanced via unique linkage to all contemporaneous Census returns, data on medications and death records. Life course epidemiological analysis and Multilevel Modelling techniques will be utilised to tease apart observed associations.

Scientific Opportunities: This Fellowship will allow me to build capacity in the manipulation and analysis of large administrative data and to generate new knowledge surrounding the effect of a history in the social care system in childhood on adult social and mental health outcomes. It will also result in the development of a unique data resource for use by other UK researchers in the field.

The proposed study will provide a greater understanding of the mental health and social outcomes of adults who were in contact with social services as children and will have an immediate impact on policy makers, third sector organisations, the Administrative Data Research Network (ADRN), public health professionals and on me, as a research professional.

Policy Makers: The Department of Health (DoH) who are responsible for the delivery of social care in Northern Ireland, including social work strategy and policies, will benefit from this study as it relates to children's social care; in particular, analysing the long-term adult mental health and social outcomes of children in need, children on the child protection register and "looked after" children. A representative from the DoH will sit on the project Steering Group and will be consulted with regard to the analysis plan, pertinent questions and dissemination of results. The Steering Group will meet at the beginning, middle and end of the project to allow input from all stakeholders at all stages of the project and foster collaborative working between academics and policy makers. Improving outcomes for children from adverse backgrounds, including those in the care of social services, is a key priority of the DoH and is core to their "Early Intervention Transformation Programme" (EITP) and with a paucity of research in this area, this Fellowship will provide unique population wide information on the effect of being known to social services in childhood on adult mental health and social outcomes, providing insights into resilience, critical periods and inter-generational effects. The research will indicate potential predictors of both poor and good adult mental health and social outcomes which could be used to target resources and interventions toward those most in need.

Third Sector: The results of this study will also have a direct impact on organisations who deal on a daily basis with children known to social services or adults with a history of social services contact, such as the charitable organisation Voices of Young People in Care (VOYPIC), Include Youth and the Northern Ireland Association for Mental Health (NIAMH). The findings will provide insight into those children most as risk of poor adult outcomes and potentially shed light on understanding why some adults known to social services as children have poor mental health and social outcomes.
ADRN: Methodologically, the DoH, academia and the ADRN will benefit from the experience of establishing a data sharing gateway and successfully linking these administrative data sources. This knowledge may be carried forward to future linkage projects in the coming years utilising similar data and with similar data custodians.

Public Health Practitioners: Indirectly, individuals such as employers and primary care health professionals may be able to utilise this research in their day to day activities, with a raised awareness of the factors that are associated with risk and resilience for poor mental health and social outcomes amongst adults known to social services as children.

Research Fellow: Personally, I will also benefit immensely from this Fellowship, developing my skills in both the manipulation and analysis of large scale administrative data, with a particular focus on advancing my skills in life course epidemiological methods and Multilevel Modelling. This Fellowship will allow me to develop independence a researcher, to gain expertise in the field, to publish high quality research and move toward my goal of obtaining an academic position in the next 5 years.