1 Child Protection Social Work in England

In England, when children are deemed to be at risk of significant harm due to abuse or neglect, the state has a duty to intervene to protect them. This duty lies with local authority children's services who will usually hold an Initial Child Protection Conference (ICPC) when referral and subsequent enquiries under Section 47 of the Children Act (1989) find that a child is suffering, or is likely to suffer, significant harm. A child protection conference brings together professionals involved with a family (i.e., social workers, child protection chairs, and others who work within, or closely with children's services, individuals from partner agencies including police), parents and sometimes older children to agree on perceived risks to the children and produce a plan to address them. The child protection process is often difficult for parents and children who find it shaming, exclusionary and stigmatising (Gibson 2015).

There is increasing interest amongst policy-makers and practitioners in using more family-led approaches to care planning in areas of child welfare including preventative services, safeguarding work, education and court proceedings (MacAllister 2022). In these settings, a Family Group Conference (FGC)—a strength-based, family-led forum—can offer an alternative way of working which enables families to be actively engaged in decision-making and creating a plan to safeguard the child. The FGC model originated in New Zealand within the Māori community, as a response to the culturally inappropriate model of intervention from children's services (Barn and Das 2016). The model has since been introduced internationally, with many countries including the UK, USA, Canada, the Netherlands, Spain and Australia using FGC to enable families to take a lead in decision-making at different stages of involvement with social work services (Nixon et al. 2005).

Research in the UK (Taylor et al. 2024) shows that, in families who had an FGC when in pre-proceedings (legal process that can lead to a child entering care), children were less likely to enter care, and for those who did enter care, they remained there for less time. Other research indicates a higher rate of FGC use in a local authority is associated with more children who enter care living with a kinship carer (Wood et al. 2024). Quantitative evidence is mixed on outcomes such as parental satisfaction or empowerment (Nurmatov et al. 2020). However, qualitative evidence suggests FGC is experienced positively by families (Pennell and Burford 2000; Forrester et al. 2008; Holland et al. 2008) in contrast to negative experiences of families in statutory social work processes (Hall and Slembrouck 2001; Ghaffar et al. 2012; Gibson 2015; Bell 2018; Bekaert et al. 2021).

2 Safeguarding Family Group Conferencing

A Safeguarding Family Group Conference (Safeguarding FGC) is a specific type of FGC offered in some English local authorities as an alternative to holding an ICPC. The term Safeguarding FGC is used to distinguish it from FGCs delivered in other contexts. Safeguarding FGCs retain the core features of a standard FGC: An independent coordinator leads the process, and the family has private family time without professionals present to develop their own plan (see Family Rights Group 2021, 2022). However, there are important differences.

In a Safeguarding FGC, there is a sustained focus on the identified safeguarding concerns, which may or may not be the case in other FGCs. As part of the process, an interim safety plan is developed and monitored by the social worker before the FGC. The plan developed in the FGC must directly address the safety issues that led to the referral. There is also clear contingency planning. If the Safeguarding FGC does not satisfactorily address the safeguarding concerns, the local authority will proceed with an ICPC. Additionally, after the Safeguarding FGC, there is ongoing monitoring by the social worker of how the plan is implemented and whether it keeps the child safe.

This approach was developed to give children and families greater control in planning responses to safeguarding concerns and, where possible, to divert them from formal child protection processes. Originating in New Zealand, FGCs became the central decision-making forum in family welfare (Brown 2003). Their development not only sought procedural reform but also to address power imbalances between statutory agencies and families experiencing involuntary intervention (Connolly and Mackenzie 1998) and tackle the persistent overrepresentation of Māori children in state care (Love 2000). Grounded in Māori ancestral knowledge and collective notions of family—where children are embedded in wider whānau and whakapapa networks (Love 2000)—the model has also been critiqued for how indigenous knowledge is incorporated into child welfare systems (Moyle and Tauri 2016). Nonetheless, others suggest that FGCs can enable extended family networks to take responsibility for safeguarding, drawing on cultural strengths and shared authority (Ngobese 2025).

In the UK, FGCs have become widely used in child protection practice (Brown 2003). Since 2000, the Netherlands has embedded FGCs as a mechanism for families and their networks to develop plans to prevent child removal (de Jong and Schout 2018), reflecting a broader international trend in the use of FGCs as a diversionary and family-led safeguarding intervention.

Safeguarding FGC could serve the same functions as ICPCs, but in a way that is more family-led (see Figure 1). Local authorities in England who first piloted FGC as an alternative to statutory child protection meetings highlight the model's fit with the principles of restorative social work, focusing on relational, strengths-based practice, aiming to ‘work with’ children and families rather than ‘doing to’ them, whilst keeping children safe (Mason et al. 2017; Stabler et al. 2025). This aligns with the intentions of the proposed legislative change in England with the Children's Wellbeing in Schools Bill (UK Parliament 2025), which emphasises the importance of working with family networks at all key decision-making points.

Figure 1 outlines how Safeguarding FGCs differ from ICPCs. Ordinarily, at the end of child protection enquiries (a Section 47 assessment in England (Children Act 1989)), if the assessment finds child protection concerns are substantiated, an ICPC is organised. This decision is made by the social worker and their team manager (HM Government 2023). Local authorities using Safeguarding FGCs may choose to offer families the opportunity to address concerns through a Safeguarding FGC, rather than following standard procedures. An independent FGC coordinator—trained in the model and sometimes accompanied by a social worker—invites families to take part. Families who consent, including children who agree to participate, are then supported in planning for the meeting, and the coordinator takes a leading role in helping parents and children to identify the important people who could be supportive, overcoming barriers to engaging their networks, and extending invites to the wider network. A core principle of both FGC and Safeguarding FGC is that all key information from professionals, including the reasons for concern, is shared with the children, their family and their wider network in advance. Families are given the chance to clarify or challenge this information before the FGC takes place.

Like other FGCs, the Safeguarding FGC process has three stages: preparation, the meeting itself (including private family time) and a review period. The Safeguarding FGC provides an opportunity for families to lead on planning with the support of a trained independent coordinator to increase support for the child and address safeguarding concerns. The FGC Coordinator works with the family to create a conducive environment to facilitate the family taking the lead. At the Safeguarding FGC, professionals and the family network share salient information (this information is not new to the group). The family then meet in private to talk amongst themselves to formulate a plan to address professionals' concerns. Finally, the family share their plan with the wider professional network, who agree to their plan unless it is believed to place the child at risk (Holland and O'Neill 2006). Professionals then offer wraparound support including referrals to support services to bolster the plan.

Whilst ICPCs also produce a plan, the key difference is the involvement of the family network. ICPC plans are not designed by families and often fail to reflect their specific needs and strengths (Richardson Foster et al. 2021). Although plan development is the stated aim of ICPCs, much of the meeting is taken up by professional information-sharing instead (Richardson Foster et al. 2021). In Safeguarding FGCs, bringing an informed and engaged wider network together in a supportive environment enables the creation of plans that are more appropriate, sustainable and tailored to the child.

2.1 The Safeguarding FGC Study

The research aimed to evaluate Safeguarding FGC to understand enablers and barriers to implementation and identify which families are most likely to benefit and in what way (for Safeguarding FGC protocol see Stabler 2024; and our paper on barriers and facilitators to implementation (Day et al. 2025).

This article focuses on the results from a realist evaluation of a Safeguarding FGC pilot conducted from November 2023 to July 2024 in three local authorities in England. Previous attempts to use this model instead of meetings such as child protection conferences, for example, in Leeds (Mason et al. 2017), have highlighted barriers to implementation and resulted in limited effectiveness data. Our study is timely because it provides insights into how Safeguarding FGC could work well to produce desired outcomes.

The realist evaluation of the Safeguarding FGC pilot is part of a larger programme of practice and research activities. A rapid realist review (Stabler et al. 2025) drew together literature focusing on models of family-led meetings including, but not limited to, FGC. An initial programme theory was developed highlighting mechanisms that could involve families meaningfully in decision-making. This helped to develop the Safeguarding FGC model, which was piloted and evaluated by three local authorities in London. The study reported in this paper followed on from this initial pilot to explore implementation of the model in new local authorities (Day et al. 2025), and a realist evaluation of family experiences of the model. The latter part of that study is the focus of this paper.

3 Data and Methods

3.1 Methodological Approach

We conducted a realist evaluation of Safeguarding FGC. Realist evaluation is a theory-driven form of evaluation that ‘searches for and refines explanations of programme effectiveness’ (Pawson 2013, 15). A realist approach allowed us to go beyond surface-level descriptions of Safeguarding FGC to investigate the generative mechanisms explaining how the Safeguarding FGC worked, for whom, under what circumstances and why (Pawson and Tilley 1997; Pawson 2013). Central to realist evaluation is the identification of Context-Mechanism-Outcome chains (CMOs) (Pawson and Tilley 1997). CMO is an analytical tool to help realist evaluators think through how an intervention affects ‘the reasoning (mechanism) of the targeted actors to cause them to adopt an intended behaviour that in a specific context will lead to a specific outcome’ (Mukumbang et al. 2018, 344). Table 1 provides definitions of the realist terms we used.

3.2 Ethics, Data Collection and Sampling

The project was funded by the National Institute for Health and Care Research (see Stabler 2024).

We conducted extensive stakeholder engagement with professionals from local authorities who had experience of implementing and delivering Safeguarding FGC to further develop, prioritise and finalise the initial programme theory. We then tested and refined it with data from new settings. In total, three local authorities piloted the approach in our data collection period—two in London and one in south west England.

Across authorities, a total of 15 families were offered Safeguarding FGC in our data collection period. Of those 15, 12 families accepted the offer of Safeguarding FGC. Of the 12 families who accepted, four families consented to the research. Of these, three progressed to a Safeguarding FGC. Some data were collected from the family that did not have a full Safeguarding FGC, which informed the programme theory.

Qualitative data were collected between December 2023 and July 2024. Methods included semi-structured interviews, observations of meetings with professionals and families and researchers' reflective notes. Interview guides and the observation template were created based on the initial programme theory and stakeholder engagement. Families who took part in the intervention during the data collection period were offered the opportunity to take part in the research. If a family consented, professionals involved in the Safeguarding FGC with the family were also invited to participate. Families included parents, children and extended family and friends in the families' support networks. Participants were given an information sheet about the research and provided written or audio-recorded verbal consent in keeping with ethics requirements.

As a pilot, the intervention was offered to few families, limiting participant numbers. In total, we undertook 27 interviews with 15 individuals, which included family network members including young people (n = 7), FGC coordinators (n = 2), social workers and support workers (n = 4) and child protection chairs (n = 2). We aimed to interview at two timepoints, and some practitioners were involved with more than one family. We also observed and took notes at 22 meetings between professionals where they discussed Safeguarding FGC referrals and families, nine meetings between professionals and families (i.e., preparation meetings, the Safeguarding FGC and review meetings). Three researcher reflective notes also fed into our analysis.

3.3 Analysis

Meeting observations and interviews were transcribed verbatim, and notes and transcripts were pseudonymised to protect anonymity. The research team developed a four-step system to analyse data using a realist approach. This included (1) familiarisation, involving an initial read through of the data to ascertain the main themes; (2) coding and annotating the data based on CMOs from the initial programme theory using the QRS International software programme NVivo 12; (3) consolidating insights from the coding using an Excel spreadsheet; and (4) conferring with the research team and stakeholders to make decisions on refinements to the programme theory. Details about how we conducted the realist evaluation can be found in Bernheim et al. (2025).

4 Findings

The analysis resulted in a programme theory comprised of nine CMO configurations written in the form of ‘if … then’ statements (Brand et al. 2019). ‘If … then’ statements are organised into five domains (see Table 2 for programme theory overview). The programme theories connect to form a model of how Safeguarding FGC seems to work.

To build an understanding of how Safeguarding FGC compares with ‘practice as usual’, we focus on two aspects of the programme theory: information sharing and involving the wider network. Whilst the form of Safeguarding FGC maps to current practice (see Figure 1), the significant difference in function is the engagement of the wider family network, including children, in safeguarding.

4.1 Information-Sharing

A key difference between ICPCs and Safeguarding FGC is how information is shared. Information-sharing about safeguarding concerns begins when the family accept the offer of a Safeguarding FGC and continues throughout the intervention. We identified two mechanisms that show how information-sharing between families, their network and professionals could lead to a shared understanding of safeguarding concerns. First, what is needed to prepare families to hear concerns (what information is shared) and second, what it takes for families to feel comfortable to share their own concerns with professionals (how information is shared).

4.1.1 A Shared Understanding of the Safeguarding FGC Process

For information-sharing to work well between children, families and professionals, it seemed important that everyone involved developed a shared understanding of the Safeguarding FGC process and possible outcomes, including the nature of the risk that led to the Safeguarding FGC, what everyone's roles are and what the intervention is trying to achieve.

I wanted to make sure that it was really clear in terms of actually … this is running … like, this threshold is met for a child protection conference, this is happening instead of a child protection conference to try and work in a different way with your family, to be really clear … that kind of threshold had been met and that actually, to try and make sure that there was that kind of robustness and seriousness in terms of the SFCG that would be the same as if it was a traditional CP conference.

• Professional Talking About a Meeting With the Family Ahead of a Safeguarding FGC

Families should understand possible outcomes of Safeguarding FGC, potentially progressing to other child protection processes, necessitating a relational yet transparent approach communicating risk and process (Forrester et al. 2012). Ensuring the purpose and boundaries of Safeguarding FGC are clear from the start could help create the conditions for families and their networks to be receptive to hear professionals' safeguarding concerns and be transparent about their own worries.

Practitioners' strategies to foster a shared understanding of the process included discussions where the purpose and aim of Safeguarding FGC were revisited to provide opportunities to adjust expectations and clarify misunderstandings. This was crucial because we observed new risks may emerge over time, and progress on addressing root causes of safeguarding concerns (e.g., parental addiction or mental-health issues) may be slow or beyond the Safeguarding FGC's scope. Establishing realistic expectations and a shared understanding of what a successful outcome would be (i.e., a robust family-led plan) is essential.

4.1.2 Families and Their Networks Are Receptive to Practitioners' Concerns

Parents often felt fearful and unprepared for ICPCs because the process and safeguarding concerns were not clearly communicated in advance (Muench et al. 2017). Information is typically provided through lengthy written reports, and parents have little control over what is discussed, limited opportunity to challenge inaccuracies and face a strong focus on negative incidents such as police or school reports (Family Rights Group 2018). This mode of information-sharing can feel ’punitive’ and ’disciplinary’ (Jackson et al. 2020), contrasting sharply with principles of partnership and restorative practice. Unsurprisingly, many parents report feeling shamed and blamed during ICPCs and are perceived by professionals as defensive or anxious (Gallagher et al. 2011; Gibson 2020).

Safeguarding FGC aims to take a different approach to information-sharing. Once the offer of the Safeguarding FGC is accepted by the family, social workers talk through their safeguarding concerns at a time and location that is convenient for everyone, and in a way the family understands. This should include speaking with the whole family network and children, not just parents. Then, the family can be more receptive to professionals' concerns, having had time to process the information and ask questions and make their own clarifications.

The Example From Observation

The FGC Coordinator met with a parent to discuss what information would be shared with their network. The parent said they were uncomfortable with the network finding out detailed information about their personal life. The FGC Coordinator worked with them to decide what information the network was told. Information shared with the network was directly linked to the children and their needs, and no unnecessary information was shared. In this way, the parent had some control over what their network was told.

The local authorities piloting Safeguarding FGC took different approaches to information-sharing, drawing on the expertise of various professionals to communicate safeguarding concerns. In one local authority, the child protection chair visited parents with the social worker to discuss concerns ahead of the Safeguarding FGC. In another local authority, the social worker, together with the child protection chair, wrote the questions for the family to address through the family plan using language that summarised risks and was intentionally solution focused. The child protection chair was involved in these instances to help ensure risks were clearly explained and that families knew professionals' concerns were at the level of child protection. The FGC Coordinator's role was to work with families to identify their wider networks of family and friends to invite to the Safeguarding FGC and then speak to every member of the network individually to explain the process and the safeguarding concerns. FGC Coordinators help ensure appropriate information is shared with the network ahead of the Safeguarding FGC and review meetings in keeping with the principles of FGC and statutory timeframes.

When families know what will be shared and help shape it, they and their networks can process the information more effectively than hearing it for the first time in a formal meeting. Sharing the right information in advance allows everyone time to process the information and come to the meeting ready to engage and think about the future.

Throughout the research, we saw examples where information-sharing about risks fell short of these ideals.

I knew what was happening, and I knew what date and I knew what time. But the content of the meeting I didn't really have a clue about, to be honest with you … Nobody explained it to me.

• Family Network Member Reflecting on Information Sharing

The reasons for this varied Cand stemmed from, for example, a lack of clarity around the process and whose role it was to share new information ahead of the Safeguarding FGC. Time pressures in the lead-up to the Safeguarding FGC sometimes meant that practitioners had to rely on phone/video calls to share concerns with the family and their network, as there was no time to meet in person.

Information sharing is a two-way process, and it is not just about practitioners sharing information at the right time with families and their networks. The next mechanism is about what needs to be in place so families and their networks feel able to share their own concerns and perspectives.

4.1.3 Families and Their Networks Are Able to Share Their Own Perspectives

Families and their wider networks need to feel safe to share their perspectives and concerns for family-led decision-making processes like Safeguarding FGCs to work well. However, research consistently shows that families can be reluctant to be open with professionals due to fears of getting others into trouble, being judged or risking child removal (Family Rights Group 2018; Russell et al. 2025). Families' hesitance also reflects wider mistrust of children's services, particularly when child protection systems are experienced as prioritising surveillance and risk assessment over genuine support or solution-focused practice (Mason et al. 2020; Buckley et al. 2011; Gladstone et al. 2014). These fears and histories of difficult interactions with services shape whether individuals feel confident or safe enough to share their concerns and perspectives.

Within this context, private family time—a core element of the FGC model—may help to mitigate these barriers by giving families protected space to deliberate without professional scrutiny. Private family time allows networks to discuss safeguarding concerns, explore options and develop plans more openly and autonomously. Connolly and Masson (2014) show that such protected spaces can enhance family autonomy and create conditions for more honest dialogue, particularly when issues are sensitive or emotionally charged. Recent research by Lalayants and Merkel-Holguin (2023) similarly demonstrates that even adapted or abbreviated versions of private family time within child protection meetings can help families process information, reduce tension and feel more confident and empowered in shaping plans. In our study, families also described being more willing to share concerns privately, particularly when risks carried shame or stigma. In this way, private family time may help counter mistrust and fear, enabling more genuine and transparent engagement with children's services.

Outside of private family time, a more open conversation between family network members and professionals may be enabled by the coordinator's ‘relational approach’—working in a compassionate, respectful, nonjudgemental and solution-focused way with families. How this looks in practice may vary. For instance, one FGC Coordinator noticed a parent was nervous during the preparation phase of the process and responded by asking about a recent holiday—helping to ease tension and build rapport. Coordinators also supported families’ comfort at FGCs by checking room accessibility, offering food and drinks and ensuring restrooms were clearly signposted, which helped families feel welcome and at ease.

There was all lots of things [food] on the table, nice drinks, and it was just very laid back, casual. We were sat in a circle, so I think we all felt very much included, and were able to speak more just because of that circle. There was no table in between.

• Family Network Member Reflecting on the Environment of the Meeting

This stands in contrast to child protection conferences where there is no specific expectation to make families feel comfortable (Muench et al. 2017). Attention to everyone's physical needs signifies respect despite challenging circumstances. Having an independent coordinator whose role is to host meetings can help make this intentional hospitality possible. It is beneficial if local authorities have a dedicated budget to support this work. Given the context of austerity in England (Webb et al. 2022), we recognise that this is difficult particularly in those local authorities hardest hit by budget cuts (Disney and Lloyd 2020).

Other professionals also worked to create a conducive atmosphere that supported clear but respectful conversations about concerns by acknowledging the family’s strengths whilst being clear about risks. For example, we observed professionals using ‘I feel’ statements when communicating concerns and being intentional about eye contact, tone of voice and where everyone was seated in the room. One professional noted that getting the balance right between focusing on the family’s strengths and communicating risks may be more difficult for less experienced practitioners. Another way professionals enabled this atmosphere was by showing genuine interest in the points of view of the family and their network by asking questions like, ‘how has that been for you?’. These small but significant interactions may help families and their network feel comfortable to open up because they are invited to do so.

You're always under a microscope when social services are involved, and you feel like the littlest thing, you're getting like judged. Because you are getting judged initially, do you know what I mean? I feel like it… it would make people feel more at ease to work with social care if they're willing to have like meet families in the meeting, get their point of views… give their point of views.

• Family Member Reflecting on the Difference It Made to be Asked for Their Perspective in a Safeguarding FGC

Once families and their networks have a clear understanding of concerns and feel their perspectives are valued, then they can develop solutions to address risks and say what they need to succeed. These solutions are usually developed during private family time and form the basis for the plan.

4.1.4 The Capacity of the Network Is Evident

A key advantage of Safeguarding FGCs is the involvement of those most important to the child, which differs from practice as usual. When the wider network attends, families gain practical and emotional support, and professionals, parents and children can see the resources available to help keep the child safe. Bringing the network together also highlights that children are embedded within broader relationships, making protective strengths more visible. This contrasts with ICPCs, where networks are rarely engaged as sources of protective strength or trusted to contribute to decision-making.

Families said having the network attend the Safeguarding FGC showed children and their parents that people love and care for them—an important outcome in and of itself.

I just thought it's a brilliant idea that, you know, instead of having to rely on them [social services], as a family we were able to help and make plans, and for [parent] to actually see that, although the professionals are there, there are people in her life who still value her.

– Family network member reflecting on the impact of the Safeguarding FGC bringing everyone together

For struggling parents, the network showing up and offering support—which need not be tangible, their presence alone can be felt as supportive—can help them to feel like they have people on their side, help them feel less alone and demonstrate to them that they are loved. This might be particularly important where there has been a breakdown of relationships within the family network, or where a parent is feeling isolated. Where the network was not fully aware of the issues the family was facing, the Safeguarding FGC provided an opportunity for the family network to support the family by showing empathy and understanding in the face of difficulties. This can be restorative of family relationships and is an important outcome from the FGC.

Meeting together at the Safeguarding FGC helped clarify what support was available and showed the network's capacity to meet the child's needs. We observed network members make practical offers—such as providing safe places for children to stay or helping with day-to-day care—highlighting how FGCs can strengthen families by sharing caregiving responsibilities. Families developed stronger ideas when the network was encouraged to suggest solutions before professionals and when given private time to plan. This was supported by practitioners who prepared network members in advance so they arrived ready with strategies and an understanding that the aim was to create a family-led plan.

A family-led plan developed in an FGC should not be seen as a way of transferring agencies' responsibilities onto family networks. Strong plans may need a blend of formal and informal resources, drawing on the strengths of both family and professional networks as has been highlighted in other studies (i.e., Brown 2003). Practitioners acknowledged this need for professional input responding to gaps in network capacity, but we witnessed that plans did not always have sufficiently resourced support. Professionals and agencies not carrying out agreed actions can erode trust and leave families without the support they need.

“I thought it was a very good meeting, but unfortunately afterwards, some of the things they said they do they haven't done.”

– Family network member reflecting on the support from services after the Safeguarding FGC

Due to the level of concern, and the fact that some of the issues facing families might be complex and deep-seated, it is possible that the family's network may not be able to meet all the children's needs. For example, the network might be small, and individuals may have competing responsibilities, or feel burnt out from supporting the family. Collaboration between professionals and the network in supporting the child and their family is needed, not a shifting of responsibility solely onto the family network. Providing promised resources is essential for maintaining trust, and giving families the resources they need—a challenge that could be accentuated by diversion from the child protection pathway (Kirk and Duschinsky 2017).

5 Discussion

This realist evaluation developed a programme theory comprising nine CMO configurations across five domains to explain how Safeguarding FGC appears to work when used as an alternative to an ICPC. Although the procedural structure of Safeguarding FGC resembles ‘practice as usual’, its distinctive contribution lies in how it reshapes the functions of child protection decision-making—particularly in relation to preparation, information-sharing, participation of the wider network, inclusion of children’s voices and future working relationships. The cumulative programme theory shows that these elements interact to generate mechanisms of collaboration, ownership, receptivity, transparency, empathy, network mobilisation and relational repair, which together support more meaningful family-led planning in situations of high risk.

A major finding concerns the relational approach to information-sharing, which differs markedly from the model underpinning ICPCs. When information about concerns was communicated early, clearly and through relational dialogue, families appeared more receptive and better able to engage constructively, echoing criticisms of the adversarial and overwhelming nature of ICPCs (Muench et al. 2017; Gallagher et al. 2011; Jackson et al. 2020). This aligns with realist explanations that mechanisms such as openness and trust depend on contextual conditions that mitigate fear, shame and confusion (Pawson and Tilley 1997; Fitz-Symonds et al. 2024; Stabler et al. 2025). The study also shows that information-sharing becomes genuinely two-way when families feel safe to be transparent, an effect strengthened by the structured involvement of the wider network and facilitated by private family time. This supports previous FGC research demonstrating that preparatory work and emotional safety are integral to meaningful participation (e.g., Mitchell 2020; Merkel-Holguin et al. 2020) but extends it by identifying the specific causal pathways through which transparency emerges. It is important to highlight the different context in which families take part in FGC. As in the USA (Merkel-Holguin et al. 2020), information disclosed in an FGC in the UK can be used as additional information in child welfare assessments. This creates a different context for transparency compared with New Zealand where information shared in an FGC is privileged and cannot be used for other purposes. This context may be a barrier to transparency and limit the information that will be included in developing an understanding of the concerns.

Mobilisation of the wider family network proved to be one of the most significant mechanisms differentiating Safeguarding FGC from ICPCs. Participation of the network provided emotional and practical support, made available resources visible to all parties and helped families develop more realistic, sustainable plans. These findings resonate with earlier work demonstrating the value of natural supports in safety planning (Brown 2003; Munro et al. 2017), but the programme theory adds explanatory depth by showing that network mobilisation alters practitioners' and families' understandings of risk and capacity. However, the evaluation also shows that this mechanism is context-dependent: weak or overburdened networks, rushed preparation or limited professional follow-through all constrained effectiveness. This reflects wider critiques of child protection systems in England, where structural pressures and resource scarcity can undermine participatory practice (Featherstone et al. 2018; Kirk and Duschinsky 2017).

Across these domains, being together in a setting intentionally arranged to reduce hierarchy and promote comfort created opportunities for empathy and relational change between families and practitioners. Practitioners described coming to see families differently, witnessing strengths that had been obscured in traditional meetings and experiencing interactions that laid the groundwork for more constructive future relationships. These findings echo earlier research demonstrating the relational benefits of FGCs (Mitchell 2020; Taylor et al. 2024) but add theoretical clarity by identifying the mechanism of relational repair and specifying the contexts that enable it to activate.

In synthesising these findings, the programme theory developed through this evaluation contributes to the growing body of literature advocating more participatory, strength-based approaches to child protection (Featherstone et al. 2018; Dillon 2021). It complements existing FGC research by explaining not only that FGCs can work, but how, why and under what conditions they produce beneficial outcomes. The findings help reconcile variation in previous evaluations by demonstrating that effectiveness depends on specific combinations of preparation, relational practice and organisational support, which may not always be in place. They also offer timely evidence as UK policy continues to promote family- and child-led approaches to safeguarding (MacAllister 2022). Further testing and refinement of this programme theory across diverse local authorities and national contexts will be critical to understanding how Safeguarding FGC can be implemented reliably and equitably.

6 Limitations

Several limitations should be acknowledged. The participating local authorities were self-selecting and had organisational readiness, leadership support and resources to pilot Safeguarding FGC. These favourable conditions likely enabled mechanisms such as collaboration, transparency and relational repair to activate, which may be less achievable elsewhere. Anonymity requirements also limited comparison between authorities.

As a realist evaluation of a small pilot, the study cannot determine which families benefit most from Safeguarding FGC or attribute outcomes to specific programme components. We instead report practitioners' views on suitability and outline plausible causal pathways rather than generalisable effects, consistent with realist methodology but limiting certainty about differential impact.

Data collection was affected by competing local authority priorities and tight timelines, reducing opportunities for follow-up with families and children. Whilst extensive stakeholder engagement helped mitigate this, further longitudinal research is needed to understand how plans develop and how mechanisms operate over time.

Finally, although the findings show how Safeguarding FGC can support family-led planning in complex, high-risk contexts, structural challenges—such as inconsistent professional follow-through, resource pressures and existing child protection cultures—sometimes weakened key mechanisms. As others have noted (Ney et al. 2013), system constraints can inhibit family participation. Safeguarding FGC therefore requires supportive organisational and policy conditions to enable relational, participatory practice to flourish.

7 Conclusion

The aim of child protection social work is to work towards keeping children safe. Our findings suggest Safeguarding FGC may provide a viable alternative to Initial Child Protection Conferences for some families as it could result in a robust safety plan led by the family and their network and supported by professionals. Our evaluation builds on a larger programme of work but also requires further exploration. We invite researchers to test and refine our programme theory in similar and different settings as there is a need for more research that explores how family experiences of FGC lead to important outcomes and longitudinal studies that explore the impact of FGCs over time.

The research has implications for policymakers and practitioners in the UK and beyond. Policymakers in the UK should permit local authorities to work differently when there is a child protection concern and allocate resources to support family-led decision-making interventions such as Safeguarding FGC. Beyond the UK, our study indicates that family-led models introduced without the policy framework to ensure quality FGC delivery, confidentiality of families when sharing information and the infrastructure to resource family plans may not achieve intended outcomes.

Whilst the research was focused on a specific intervention, appropriate and helpful information sharing and recognising and enhancing network capacity are relevant for practitioners beyond the use of Safeguarding FGC. These mechanisms should be present in relationship-based, collaborative practice that prioritises family-led decision-making. This paper provides examples of how practitioners can work in this way in the difficult setting of child protection work.