Children in Family Justice Data Share

1. Children in Family Justice Data Share MoJ, DfE and Cafcass Bridgette Miles, MoJ

2. A brief history of data linking…. What’s the problem? The data share is a comprehensive and detailed database which provides opportunities for analysis that have not been possible before. Some of the key strategic evidence gaps the data share has the potential to address include: Building a picture of the demographic profile of children with experience of the family justice system including their age, gender, ethnicity and special educational needs (SEN). Providing longitudinal data on the pathways of children from their first entrance to the family justice system through to their case and educational outcomes. Linking detailed case-level data such as the type, pattern and timing of applications and orders made with educational attainment at Key Stages 1 to 5 will build our understanding of how different experiences and decisions made within the family court can impact on children’s educational outcomes. Comparing the demographic profile of children, both in public and private law, with children in the general population, as well as to observe how the profile of children in the family justice system has changed over time. What’s in it for me? Understanding the potential of the data share • The Family Justice Review (FJR), published in November 2011, was critical of the management of information and summary data held by the government in relation to cases in the family justice system.  • The review found that disparate stand-alone IT systems and the lack of communication and co-ordination across government departments and agencies led to inefficiencies and delays within the system. • It also prevented the development of a coherent picture of a child’s journey through the family court system and its potential impact on their life chances and outcomes. • Significant gaps in management information and lack of robust data impacted on the development of evidence-based policy. • Final report recommended that an integrated IT system should be developed for the family justice system and wider agencies. • As an interim, the government conduct an urgent review of how better use could be made of existing systems.

3. Some facts and figures What’s in the Children in Family Justice data share? How many children are included? *The total child count is higher than the number of individual children because a child will be counted twice if they have been subject to both a public and a private law case.

4. Data linking process and timeline A simplified data linking process How long did it take? Unique identifiers and key personal details required for matching transferred Children were identified and matched by variables that were recorded in each individual dataset and combined to create a dataset where each child is represented once. The matching rates across the datasets are currently between 70% and 75%.

5. The first output – Public Law Applications to Orders (PLATO) tool https://public.tableau.com/profile/moj.analysis#!/vizhome/ChildreninFamilyJusticePublicLawApplicationstoOrdersTool_0/FrontPage

6. What challenges did we face along the way?

7. What have we learnt? Importance of senior buy-in and sponsorship in all organisations Engaging with experts Be pragmatic and realistic re scope of project Formal project structure Dedicated resource needs to be made available Making best use of different talents – great example what cross-profession working can achieve

8. If you want to find out more: • Contact: • Data • Bridgette Miles • Bridgette.miles@justice.gov.uk • Research and analysis • Amy Summerfield • amy.summerfield@justice.gov.uk • Advanced analytics & data science • Olivia Lewis • olivia.lewis@justice.gov.uk

9. Out of Hospital Cardiac Arrest A Scottish Government sponsored data linkage project 18 July 2018 GSS methodology symposium Dr Claire Wainwright Data Statistics and Outcomes Division Digital Directorate Scottish Government

10. Case Study: OHCA Two high level aims: Increase survival rates (save 1,000 additional lives by 2020) Equip an additional 500,000 people in Scotland with CPR skills by 2020

11. How does the linkage lead to improved resource allocation, policy delivery and patient outcomes? • Linking ambulance data with other clinical datasets in Scotland allowed patient outcomes to be tracked alongside monitoring of system performance and the impact of service changes. • How many people in Scotland who had an OHCA were resuscitated and were discharged from hospital alive? • How many patients who had an OHCA received bystander CPR? • Are there geographical variations in survival rates following an OHCA? If so, of what magnitude? • Are there differences in care delivery systems around Scotland which impact on survival rates following an OHCA? If so, of what magnitude? • The project considered several other factors as part of the analysis including, patient case-mix, geography, deprivation, healthcare delivery system (from initial 999-call through paramedic resuscitation, hospital care to return to the community).

12. Case Study: Out of Hospital Cardiac Arrest Who are our partners? Scottish Ambulance Service University of Edinburgh NHS National Services Scotland Scottish Government

14. Challenges in delivering the linkage project • Timescales – Scottish Government reporting • Changes to IT systems impacting on data provisioning • Complexity and volumes of data required • Permissions processes - SG Analytical Leadership Group permission secured in August 2015, data provisioned a year later • 75.7% of OHCA linked to CHI • Data Quality – completeness and quality of bystander CPR data is poor and work is underway to improve it

15. What outcomes have we seen so far? • This project started in 2016, with two rounds of analysis already completed (next wave summer ’18) • The results have already fed into: • SG Strategy Report (which is published every 2 years) • Analytical Report (published annually) • SG funding for an analyst sitting in Scottish Ambulance Service – will work on data management and data quality. Working towards new publications and enhanced data. • Future work: engagement with the Fire Service and Police Scotland – building on successes to date • Further work to explore placement of defibrillator devices

16. Lessons Learned • Explore data quality and availability at the outset and schedule time to identify, prepare, clean and QA the data prior to linkage • Ensure regular formal communications channels established at the outset – project teams, roles and responsibilities • Adequate time to navigate permissions processes – schedule time to obtain approvals and anticipate questions • Strong Project Management needed - to ensure actions, owners and timescales agreed, detailed specification of data provisioning to ensure correct datasets ready for linkage. • Preserve the learning and linkage outputs where justified – to get the best from all the effort put into bringing the data together. An OHCA “epistry” (epidemiological registry) is being developed and we are now updating the linked dataset.

17. Questions? You can also contact us at: DataLinkage@scot.gov