Coram to run UK’s first randomised controlled trial of Family Group Conferencing

  • 16 March 2020

How effective are family group conferences (FGCs) when it comes to improving outcomes for children? The evidence may be less clear-cut than many imagine. Some previous evaluations of FGCs have shown promising results. Families can feel more empowered, relationships can be strengthened, and court proceedings can be avoided. However, the systematic review of FGCs published this month by What Works for Children’s Social Care found that this evidence base is not currently strong, so it is timely that Coram is rolling out the most ambitious research project yet, to fill the evidence gap.

Every year there are perhaps thousands of family group conferences in the UK. They are entirely voluntary – no family can be forced to take part. When they come ahead of potential court proceedings, as in our study, they represent a last chance for families to plan and make decisions for children or young people who are considered at risk.

So, we are excited at Coram to have been invited to deliver the first ever randomised controlled trial (RCT) of the use of FGCs in the UK, with a view to providing the robust evidence currently lacking on the difference FGCs make. An RCT is the next step for the FGC evidence base.

We will be working with over 20 local authorities across England who are committed to producing better evidence for future commissioning, to find out what happens to families. Unlike other kinds of research, RCTs can allow us to reach conclusions about cause and effect – causality. We aim to establish through our trial whether it is FGCs that lead to better outcomes for families, excluding other factors that might explain changes.

How will it work? RCTs involve randomly assigning families to be referred or not referred for a particular service, programme or approach. This contrasts favourably with current practice. Currently, availability of and support for FGCs across the country is patchy, and they are not offered universally. There is no statutory requirement on local authorities to provide them, so families may or may not be offered a FGC according to where they live. Funding constraints also make provision uneven. This is, if anything, fairer than how FGCs are allocated currently.

In an RCT, the only difference between groups of families is whether or not they were referred – other differences are averaged out. When we compare the groups later, we can attribute any differences we see to whether or not a referral was made. This helps in the context of FGCs because families’ lives are so complex. It’s a way of holding constant all that complexity, all that noise, and hearing the signal being sent by the FGCs.

Some practitioners and researchers in children’s social care oppose randomisation on principle, but most recognise the value of this gold-standard research method, as long as appropriate safeguards for families are put in place. So we have considered the ethical trade-offs and put in place a number of design features and mitigations. We have received internal and external advice and gained the approval of Coram’s research ethics advisory board.

Families will be randomised at the point of entering pre-proceedings to either be referred or not for a FGC. The local authorities taking part in the programme volunteered last year, on the basis that they were not currently offering FGCs at pre-proceedings stage. 

All the FGCs funded through the trial will be additional FGCs that in likelihood would not otherwise have happened, so no-one will miss out on something they would otherwise receive.

Participation in an FGC is voluntary, as is participation in the study as a whole. Families who choose not to take part will receive care as usual, and will not lose out in any way. To help them choose, families will have access to free legal aid advice, thanks to the timing of the study at pre-proceedings stage.

We have also listened to FGC advocates for whom FGCs are about maximising family involvement in the process, not just improving outcomes. With that in mind, we plan to gather parents’ views. We will find out how included parents feel they have been in care planning at a point two months after entering pre-proceedings, and whether this differs between those who are referred for FGCs, and the non-referred group.

The project includes a process evaluation (interviews, surveys, document review and case studies) to discover how and why FGCs work, and to ensure we hear from family members, including young people. We will also be drawing on work by Leeds City Council, the London Borough of Camden and the Family Rights Group to assess the quality and consistency of the new model.

Referrals are due to begin in April 2020 and end in September 2021, with data collection continuing beyond this, and publication of findings in 2022. By the end of the project we hope to have data on over 3,000 families, and around 6,000 children.

We don’t provide FGCs at Coram, so have no vested interest, other than our commitment to better outcomes for children. Coram’s range of services for children and families means its research always has one foot in the real world. So we will include the voice of the child or young person in our research. This trial will provide the sector with the evidence currently missing on whether and how FGCs do make a positive difference for children and families.

We are excited to get started on this challenging but important study. More information about the RCT, including a full list of the local authorities taking part, can be found on the study website. You can also find our video on YouTube.