The British Journal of Social Work · Family group conferences
Family Group Conference Provision in UK Local Authorities and Associations with Children Looked after Rates
Born in New Zealand to give Māori families back the decisions about their children, the family group conference (FGC) is now offered by nearly eight in ten UK local authorities. The CASCADE team in Cardiff maps this provision in 2022 and, for England, sets it against the figures on children in care.
This is a reformatted republication of Family Group Conference Provision in UK Local Authorities and Associations with Children Looked after Rates, by Sophie Wood, Jonathan Scourfield, Melissa Meindl, Kar Man Au, Rhiannon Evans, Delyth Jones-Willams, Fiona Lugg-Widger, Philip Pallmann, Michael Robling, Elizabeth-Ann Schroeder, Stavros Petrou and David Wilkins, published in The British Journal of Social Work (Oxford University Press) (2024), doi: 10.1093/bjsw/bcae019, under a CC BY 4.0 licence. Prepared by Complexe Systémique in September 2026: the authors’ text is unchanged; the layout has been adapted for reading online, which constitutes a modification of the work under the terms of the licence. Two tables are presented as lists; end-of-article declarations are grouped in the notes. This edition was made neither by the authors nor by the publisher, who are not responsible for its content or for any errors. The original version prevails.
They are of value in their own right as a mechanism for families to fulfil participation rights, regardless of the outcomes in terms of children being looked after.
Sophie Wood, Jonathan Scourfield, Melissa Meindl, Kar Man Au, Rhiannon Evans, Delyth Jones-Willams, Fiona Lugg-Widger, Philip Pallmann, Michael Robling, Elizabeth-Ann Schroeder, Stavros Petrou and David Wilkins
Abstract
Family group conferences (FGCs) in child welfare share decision-making with family members by bringing the immediate and wider family together to make a plan to meet a child’s needs. This paper reports survey findings on FGC provision in the UK in 2022 and explores whether in England the presence of an FGC service and the rate of FGC provision is associated with the rate of children in care, entering care, in kinship foster care and leaving care. Seventy-nine per cent (n = 167) of local authorities in the UK provided FGCs to families, and 14 per cent (n = 29) did not. Services that were more established offered a more diverse range of FGCs. The introduction of FGCs in English local authorities was associated with a higher rate of children in care, but also higher rates of kinship foster care, a key goal of FGCs where it is not possible for children to stay with their parents. Higher rates of FGCs were associated with more children leaving care, possibly due to reunification with birth families. To understand in more detail, the circumstances of children in and leaving care in local authorities with FGCs, individual data linkage studies are needed.
Keywords: child protection, child welfare, family group conference, kinship care, out-of-home care, reunification
Introduction
A family group conference (FGC) in child welfare gathers immediate and extended family members to jointly determine the best course of action for meeting the needs of a child requiring support and/or protection. Unlike professionally led meetings like child protection case conferences, FGCs aim to involve family members in the decision-making process. (Metze et al., 2015).
The primary aims of FGCs are empowering families by increasing family participation in important decisions about children (Holland and O’Neill, 2006; Crampton, 2007), improving child safety through reduced child maltreatment, whilst also reducing state intervention and out-of-home care (Dijkstra et al., 2016; Sen and Webb, 2019) and, if children need to be removed from birth parents, keeping children within the wider family network where possible (Pennell et al., 2010). This paper describes the state of FGC provision in the four nations of the UK and seeks to understand if the provision of FGCs in England is associated with the rates of children in care, children starting to be looked after, kinship foster care and children ceasing to be looked after, at a local authority level.
The use of family group conferences in the UK
FGCs originated in New Zealand in response to the high proportion of Māori children in state care (Taylor et al., 2023). They became mandated practice there in 1989 (Connolly, 2006). In the UK, FGCs were first introduced in 1991 by the Family Rights Group (Brown, 2003), which promotes and quality assures FGCs across the UK. However, there is no legal mandate in any of the four nations that FGCs must be offered to children before coming into care. In England, Government guidance states that local authorities should consider referring a family to an FGC service before a child enters state care (Department for Education, 2014), and the policy context of FGC provision is broadly similar in the other devolved nations. Due to the lack of policy in the UK about when and how FGCs should be offered to families, the delivery and stage (e.g. early help, edge of care, reunification) at which FGCs are offered may vary between areas.
However, the Family Rights Group have developed seven quality standards (Family Rights Group, 2022). The first three standards, and the most commonly referenced, dictate that (1) the FGC coordinator is independent; (2) the family’s decision to participate in the FGC is voluntary; and (3) the FGC should be family-led and include ‘private family time’ for the family to propose a plan in response to concerns.
The current extent of FGC coverage in the UK is unknown. The most recent published study (Brown 2003), conducted ten years after FGCs were introduced in 1991, found that 38 per cent of councils in England, Wales and Northern Ireland had an FGC service and 59 per cent did not. The remainder did not respond to the survey. The author concluded that FGCs were still on the margins of practice. Anecdotal evidence suggests that momentum for FGCs appears to be increasing in the UK. The Family Rights Group (2019) estimated in 2019 that three quarters of local authorities in England and Wales run or commission an FGC service or are planning to do so.
Evidence about the effectiveness of family group conferences
Evidence about the outcomes of FGCs and, more broadly, family group decision-making have focused mainly on child maltreatment, the number and length of out-of-home placements, and involvement of social services (Dijkstra et al., 2016). However, evidence about the effectiveness of FGCs is mixed, as shown in three international systematic reviews (Dijkstra et al., 2016; McGinn et al., 2020; Nurmatov et al., 2020), and variation in quality and context (Cosner Berzin et al., 2007; Stabler et al., 2019) might explain these mixed results.
A randomised controlled trial (RCT) was recently carried out in England (Taylor et al., 2023). This was the first RCT on FGCs in the UK and one of the largest studies of FGCs internationally. It was not included in the systematic reviews cited above, as they predated it. In this trial, it was observed that children from families referred for an FGC before the initiation of care proceedings were significantly less likely to be in care twelve months after entering pre-proceedings, compared to families who were not referred for an FGC. Among children in families referred for FGCs, 36.2 per cent went into care, compared to 44.8 per cent of children in families not referred for FGCs.
Other studies have focused on outcomes such as cost-effectiveness (Dijkstra et al., 2018), parental self-efficacy (Frost et al., 2014), improved relationships (Edwards et al., 2020; Mitchell, 2020), and increased social support networks (Corwin et al., 2020).
Although most studies focus on individual outcomes of families receiving FGC services, very few studies (e.g. Sen and Webb, 2019) have explored local authority system-level outcomes after an FGC service is introduced, such as changes in the overall rates of children in care, nor how the amount of FGC provision in a local authority affects care outcomes. Studies have also tended to focus on outcomes for FGCs offered when a child is at the edge of care (e.g. Taylor et al., 2023), rather than the wide spectrum of circumstances under which FGCs are offered. This paper addresses these gaps by answering the following research questions (RQs):
RQ1: What is the extent of FGC provision in the UK?
RQ2: Do the rates of children looked after, children starting to be looked after, children in kinship foster care and children ceasing to be looked after in England change in the two years after FGCs are introduced, compared to the two years before?
RQ3: Is there a relationship between the annual rate of self-reported FGC provision in England and the rates (3-year average) of children looked after, children starting to be looked after, children in kinship foster care, and children ceasing to be looked after?
Methods
An online survey was developed in collaboration with the Family Rights Group and in consultation with FGC service managers and peer researchers with lived experience of FGCs. Publicly available local authority-level administrative data in England were sourced and matched to the survey data to investigate the association between the availability and extent of FGC provision on rates of children in care. Only English administrative data were used due to differences in administrative data collection between the four UK nations and due to England having the largest sample size.
Setting/context
Local authorities have legal responsibilities concerning child welfare. Some administer their own in-house FGC services, whilst others contract FGC services from voluntary or private sector organisations. In England, upper-tier authorities—either unitary, county councils, metropolitan districts or London boroughs—are entrusted with child welfare responsibilities. In Scotland and Wales, sole unitary authorities hold child welfare obligations. In Northern Ireland, Health and Social Care Trusts (HSCTs) are responsible instead of local authorities, although, for brevity in this paper, we generally use ‘local authority’ to denote the local governing unit across the entire UK.
Sampling
The survey was distributed in May 2022 to every local authority in England (n = 152), Scotland (n = 32) and Wales (n = 22), along with HSCTs in Northern Ireland (n = 5; Total n = 211). Additionally, we gathered data regarding local authorities that have never used FGCs and those that have withdrawn their support from such services.
Survey development
The questionnaire was made to be simple, featuring both multiple-choice questions for numerical data and open questions allowing respondents to provide detailed answers in free text boxes. It covered topics such as whether FGCs are used in the local authority, the stage of child welfare concerns when FGCs are offered, and the number of conferences run each year. See Wood et al. (2022) for a copy of the survey used.
Data collection
Survey
Two methods were employed to access FGC services. Initially, the online survey link was distributed to all heads of children’s services (or equivalents) across the UK, with instructions to disseminate the survey among relevant staff members, including FGC service managers. Additionally, the online survey link was directly forwarded to local FGC services, identified through systematic online searches. We referred to a list of services on the Family Rights Group website (https://frg.org.uk/), complemented by a Google search using the terms ‘family group conference’ (and ‘family group meeting’ in Scotland) alongside each specific local authority name.
Administrative data
Publicly accessible aggregate data were sourced from the Department for Education (2023a) for England. To access aggregate data on kinship foster care at a local authority level, a freedom of information request was submitted to the Department for Education.
Measures
Family group conference coverage
Family group conference coverage in May 2022 came from three sources: 1) the survey; 2) consultation with FGC network leads in each nation; and 3) internet searches. The knowledge of FGC network leads was prioritised over internet searches, where the two sources contradicted each other. Internet searches were only used to identify the presence of a service (rather than the absence).
Rate of family group conference provision
The survey asked respondents how many FGCs they provide on average each year. Mid-population estimates for numbers of 0- to 17-year-olds in 2022 (Office for National Statistics, 2023) in each local authority were used to calculate the average rate of FGC provision per 10,000 children.
Rates of children looked after, children starting to be looked after, children in kinship foster care and children ceasing to be looked after
Rates of children looked after, children starting to be looked after, children in kinship foster care and children ceasing to be looked after were calculated per 10,000 children by Upper Tier Local Authority District in England using data from the Department for Education (2023a) and mid-population estimates for numbers of 0–17 year-olds from the Office for National Statistics (2023).
In England, a child is considered to be looked after if they are ‘provided with accommodation for a continuous period of more than 24 hours; are subject to a care order; or are subject to a placement order’ (Department for Education, 2023b, p15). Children receiving short breaks or respite care are reported, but only those under Section 20 of the Children Act 1989. A child starts to be looked after when they start a new period of care—they may have had a break from care and are returning. New periods of respite care are not included in counts of children starting to be looked after. A child ceases to be looked after for several reasons such as returning home to live with parents, moving to a special guardianship order or ageing out of care (Department for Education, 2023b). Kinship foster care is defined in this study as being placed in foster care with a relative or friend. Kinship care can also be an informal arrangement between relatives or friends, but these arrangements are not captured in routine data returns.
Deprivation
Because the rate of children looked after is strongly correlated with area-based deprivation (Bywaters et al., 2020), analyses of the relationship between FGC provision and patterns of children looked after need to take account of deprivation. The local authority English Indices of Deprivation rank (2019) was derived from The Ministry of Housing, Communities and Local Government (2019) dataset.
Data analysis
Descriptive statistics for survey responses were produced using Stata software (RQ1). Paired t-tests were used for the pre–post analysis, and histograms with frequencies with overlaid normal density curves to visualise the distributions (RQ2). To test associations between rates of children in care and the rate of FGC provision, linear regression models were used adjusting for deprivation (RQ3).
Practitioner and service user involvement
The survey was piloted with FGC practitioners (n = 5). The study’s Families’ Research Advisory Group (n = 6) were consulted to help interpret the findings of the survey. This group is made up of adult family members with lived experience of participating in FGCs or other aspects of involvement with children’s social care services. Two peer researchers with lived experience of their own family having an FGC were part of the research team, so contributed to the design and interpretation.
Ethical approval
Ethical approval was granted from the UK Health Research Authority’s Social Care Ethics Committee. All participants provided informed consent to participate in the study.
Results
In total, the survey was completed by 160 respondents representing two-thirds (65.9 per cent; n = 139/211) of local authorities in the UK. Wales had the highest response rate of 86.4 per cent (n = 19/22), followed by Northern Ireland (80.0 per cent, n = 4/5), Scotland (68.8 per cent, n = 22/32) and England (61.8 per cent, n = 94/152).
The survey was filled in mainly by FGC managers (49.6 per cent, n = 69/139) or managers in children’s services (36.7 per cent, n = 51/139). Three participants said they were both FGC managers and children’s social services managers. Thirty percent selected an ‘other’ option (30.2 per cent, n = 42/139), describing their roles as FGC coordinators and practitioners, heads of services, strategic leads, a care proceedings case manager, commissioning leads and other roles within children’s social services.
The findings of this paper are split into three parts. The first estimates FGC coverage using survey responses supplemented with consultation with FGC service leads in each nation and internet searches. The second presents data from survey responses only across the UK. Finally, the third links survey responses from England to publicly available administrative data on the rates of children looked after, children starting to be looked after, children in kinship foster care and children ceasing to be looked after.
Part 1: FGC coverage
This study found that 79.1 per cent (n = 167/211) of local authorities in the UK provided FGCs to families and 13.7 per cent (n = 29/211) did not (Figure 1). It was unclear for the remaining 7.1 per cent if they provided FGC services. Please see supplementary material for a breakdown by nation and to see data from survey responses only.
Part 2: UK survey responses
When did FGCs start to become more common in the UK?
Respondents were asked about the initial implementation of FGCs within their local authority, which also encompassed any interruptions in the provision. Figure 2 presents when continuous coverage began in each local authority. It shows that the number of UK local authorities providing FGCs has been growing continuously for more than 25 years, but the uptake rate has seen a marked acceleration since 2016.
Nearly two-thirds (58.7 per cent, n = 71/121) of respondents said that their FGC service had expanded in size in the last three years, highlighting another sign of the growing prevalence of FGCs across the UK. One-third (33.1 per cent, n = 40/121) said their FGC service had maintained its current size, whilst 17.4 per cent (n = 21/121) reported a reduction in size. In eleven instances, multiple individuals completed the survey for the same local authorities, resulting in percentages that do not sum up to 100. Additionally, contradictory responses were provided in four cases. Some local authorities might operate multiple FGC services, potentially explaining scenarios where one service expanded whilst another reduced in size. Local authorities that had started a service more recently were more likely to say their service had expanded in size than local authorities that had been established for longer. The mean length of service for FGCs that said their service had expanded in size was 7.2 years (SD = 6.4), 10.8 years (SD = 7.7) for FGCs that said their service size had stayed the same, and 12.1 years (SD = 6.0) for FGCs that said their service had reduced in size.
Size of the FGC service
Local authorities held different numbers of FGCs annually, ranging from 5 to 800 (Table 1). The median count per year was 92.5, whilst the mean was 140.9. To understand the size of FGC provision in relation to the child population, the yearly rate of FGCs per 10,000 children within each local authority was computed. As shown in Table 1, out of the four UK nations, Wales had the highest rate of FGCs per child population, despite having the lowest raw number of FGCs annually, due to its local authorities being relatively smaller.
Table 1 — Number and rate of FGCs annually by local authority for each UK nation*.
- Raw numbers
- Mean (SD); UK (n = 104): 140.9 (157.1); England (n = 72): 163.1 (168.7); Wales (n = 18): 83.6 (71.5); Scotland (n = 10): 105.4 (182.4); Northern Ireland (n = 4): 88.3 (75.6)
- Median; UK (n = 104): 92.5; England (n = 72): 113; Wales (n = 18): 51; Scotland (n = 10): 46.5; Northern Ireland (n = 4): 60
- Range; UK (n = 104): 5–800; England (n = 72): 20–800; Wales (n = 18): 12–240; Scotland (n = 10): 5–600; Northern Ireland (n = 4): 33–200
- Rate per 10,000 children
- Mean (SD); UK (n = 104): 21.9 (19.0); England (n = 72): 20.4 (18.3); Wales (n = 18): 31.2 (21.7); Scotland (n = 10): 20.7 (18.3); Northern Ireland (n = 4): 9.2 (6.4)
- Median; UK (n = 104): 16.2; England (n = 72): 14.9; Wales (n = 18): 24.5; Scotland (n = 10): 17.1; Northern Ireland (n = 4): 7.5
- Range; UK (n = 104): 1.5–122.0; England (n = 72): 1.6–122.0; Wales (n = 18): 5.0–83.0; Scotland (n = 10): 1.5–53.8; Northern Ireland (n = 4): 3.3–18.3
* In a small number of local authorities, more than one person had completed the survey and had different responses to this question. In these instances, the higher number was taken. Source: Reproduced from Wood et al. (2022).
When FGCs are offered to families
FGCs were predominantly available at either pre-care proceedings (95.9 per cent, n = 117/122) or during the consideration of a child for a child protection plan (95.9 per cent, n = 117/122; Figure 3). Eighty-four percent (84.4 per cent, n = 103/122) of local authorities offered FGCs for reunification planning. There was also a substantial number of FGCs (71.3 per cent, n = 87/122) offering early help services, although this might stem from differing interpretations of what qualifies as ‘early’ help.
T-tests showed that the longer an FGC service had been running the more likely the local authority was to offer FGCs for early help (t(108) = –1.7, p = 0.043), reunification planning (t(108) = –1.7, p =0.043) and for children leaving care (t(108) = –1.9, p = 0.030). A relationship could also be found between the rate of FGCs per 10,000 children and types of stages of concern FGCs are offered to families. T-tests showed that the higher the rate of FGCs in an area the more likely the local authority was to offer FGCs for children in need (t(100) = –1.7, p = 0.043), during care proceedings (t(100) = –1.8, p = 0.038), reunification planning (t(100) = –2.9, p = 0.002), building relationships with birth families when a child is in care (t(100) = –2.3, p = 0.012) and for leaving care (t(100) = –1.7, p =0.044). No further statistically significant differences were found between length of service nor rate of FGC provision and the stages of concern listed in Figure 3. Mean differences in the length of FGC service and rate of FGCs for each stage of concern can be found in Table 2.
Table 2 — Mean differences in length of FGC service and rate of FGCs by stage of concern offered.
- Early help. Length of FGC service (years) M (SD)a: Offered 9.7 (6.9); Not offered 7.2 (7.1). Rate of FGCs per 10,000 children M (SD)a: Offered 22.8 (16.8); Not offered 20.3 (23.6).
- Pre-(care) proceedings. Length of FGC service (years) M (SD)a: Offered 8.9 (7.0); Not offered 11.75 (7.3). Rate of FGCs per 10,000 children M (SD)a: Offered 22.5 (19.3); Not offered 8.1 (7.0).
- Child protection plan/register. Length of FGC service (years) M (SD)a: Offered 8.9 (7.0); Not offered 11.75 (7.3). Rate of FGCs per 10,000 children M (SD)a: Offered 22.5 (19.3); Not offered 8.1 (7.0).
- During care proceedings. Length of FGC service (years) M (SD)a: Offered 9.0 (6.9); Not offered 9.4 (8.6). Rate of FGCs per 10,000 children M (SD)a: Offered 22.9 (19.6); Not offered 10.4 (7.4).
- Child in need. Length of FGC service (years) M (SD)a: Offered 9.0 (7.1); Not offered 8.4 (7.3). Rate of FGCs per 10,000 children M (SD)a: Offered 22.8 (19.5); Not offered 9.9 (9.0).
- CLA—reunification planning. Length of FGC service (years) M (SD)a: Offered 9.5 (7.1); Not offered 6.4 (6.1). Rate of FGCs per 10,000 children M (SD)a: Offered 24.3 (20.1); Not offered 10.1 (6.0).
- CLA—leaving care. Length of FGC service (years) M (SD)a: Offered 9.8 (7.3); Not offered 7.1 (6.1). Rate of FGCs per 10,000 children M (SD)a: Offered 24.3 (17.9); Not offered 17.5 (21.0).
- CLA—building relationships with birth families. Length of FGC service (years) M (SD)a: Offered 9.0 (6.9); Not offered 9.1 (7.3). Rate of FGCs per 10,000 children M (SD)a: Offered 25.2 (18.0); Not offered 16.3 (20.1).
- Other. Length of FGC service (years) M (SD)a: Offered 9.8 (7.3); Not offered 8.7 (7.0). Rate of FGCs per 10,000 children M (SD)a: Offered 23.6 (16.2); Not offered 21.3 (20.4).
CLA: child looked after. a Statistically significant differences between means identified in T-tests are in bold.
Part 3: Survey responses linked to administrative social care data in England
Comparison of rates of children in care pre- versus post-FGC introduction
In 76.5 per cent (n = 26/34) of local authorities, the introduction of FGCs was associated with an increase in children looked after rates over a two-year period. The remainder of local authorities (23.5 per cent, n = 8/34) saw a reduction in rates. See Figure 4 for the distribution of differences in rates. The paired t-test showed that there was a statistically significant difference in the rate of children looked after pre- versus post-FGC introduction. The results from the pre-test (M = 74.4, SD = 26.6) and post-test (M = 82.7, SD = 32.1) indicated that the rate of children looked after tended to increase with the introduction of FGCs, t (33) = –3.6, p < 0.001.
In 55.9 per cent (n = 19/34) of local authorities, the introduction of FGCs was associated with an increase in children starting to be looked after rates over a two-year period. The remainder of local authorities (44.1 per cent, n = 15/34) saw a reduction in rates. The paired t-test showed that overall there was not a statistically significant difference in the rate of children starting to be looked after pre- versus post-FGC introduction. The results from the pre-test (M = 32.2, SD = 15.6) and post-test (M = 32.2, SD = 13.5) indicated that there was very little difference in the rate of children starting to be looked after, after FGC introduction, t (33) = 0.03, p= 0.978.
In 58.8 per cent (n = 20/34) of local authorities, the introduction of FGCs was associated with an increase in kinship foster care rates over a two-year period. The remainder saw a decrease in rates (41.2 per cent, n = 14/34). The paired t-test showed that overall there was a statistically significant difference in the rate of kinship foster care pre- versus post-FGC introduction. The results from the pre-test (M = 17.5, SD = 9.7) and post-test (M = 23.4, SD = 16.5) indicated that there was a statistically significant increase in the rate of children in kinship foster care, after FGC introduction, t (33) = –3.5, p= 0.001.
In 50.0 per cent (n = 17/34) of local authorities, the introduction of FGC was associated with an increase of children ceasing to be looked after rates over a two-year period. The remainder saw a decrease in rates (50.0 per cent, n = 17/34). The paired t-test showed that overall, there was not a statistically significant difference in the rate of children ceasing to be looked after pre- versus post-FGC introduction. The results from the pre-test (M = 29.2, SD = 9.4) and post-test (M = 30.0, SD = 12.0) indicated that there was very little difference in the rate of children ceasing to be looked after, after FGC introduction, t (33) = –0.6, p= 0.268.
The association between annual rates of FGC provision and rates of children in care
There was not a statistically significant association between the rate of FGCs and the rate of children looked after. There was also not a statistically significant relationship between the rate of FGCs and children starting to be looked after (Table 3).
Table 3 — Linear regression output showing the association between the rate of FGC in a local authority and the rate of children looked after, children starting to be looked after, children in kinship foster care, and children ceasing to be looked after.
- Children in care. FGC rate per 10,000 childrena: B(SE) 0.39(0.20); β 0.17; (95% CI) –0.03 to 0.81. IMD rank (higher=lower deprivation): B(SE) –0.46(0.21)***; β –0.60; (95% CI) –0.56 to –0.32. R2: 0.42.
- Children starting to be looked after. FGC rate per 10,000 childrena: B(SE) 0.11(0.07); β 0.15; (95% CI) –0.03 to 0.26. IMD rank (higher=lower deprivation): B(SE) –0.14 (0.02)***; β –0.57; (95% CI) –0.19 to –0.09. R2: 0.37.
- Children in kinship foster care. FGC rate per 10,000 childrena: B(SE) 0.23 (0.11)*; β 0.24; (95% CI) 0.02 to 0.45. IMD rank (higher=lower deprivation): B(SE) –0.14 (0.06)***; β –0.42; (95% CI) –0.21 to –0.07. R2: 0.26.
- Children ceasing to be looked after. FGC rate per 10,000 childrena: B(SE) 0.15(0.07)*; β 0.21; (95% CI) 0.01 to 0.29. IMD rank (higher=lower deprivation): B(SE) –0.13 (0.02)***; β –0.54; (95% CI) –0.18 to –0.08. R2: 0.37.
B = regression coefficient; SE = standard error; β = correlation coefficient; 95% CI = 95% confidence interval. a One local authority had a very high rate of FGCs (5.5 standard deviations from the mean) and was therefore removed from the analysis as an outlier. * p < 0.05, ** p < 0.01, *** p < 0.001.
However, there was a statistically significant association between the rate of FGCs and the rate of children in kinship foster care, with higher rates of FGCs associated with higher rates of children in kinship foster care. After deprivation was accounted for, for every extra FGC per 10,000 children provided in a local authority, an extra 0.23 (95% confidence interval: 0.02–0.45) per 10,000 children were in kinship foster care. This means a 10 per cent increase in the provision of FGCs in a local authority was associated with 2.3 per cent more children in kinship foster care.
Finally, there was a statistically significant association between the rate of FGCs and children ceasing to be looked after, with higher rates of FGCs associated with higher rates of children ceasing to be looked after. After deprivation was accounted for, for every one extra FGC per 10,000 children provided in a local authority, an extra 0.15 (95 per cent confidence interval: 0.01–0.29) per 10,000 children cease to be in care. In other words, a 10 per cent increase in FGCs provision was associated with 1.5 per cent more children ceasing to be looked after.
Discussion
This study found that 79 per cent of local authorities in the UK provided an FGC service in 2022; however, the size and characteristics of the services varied. There has been a marked acceleration of the provision of FGCs since 2016 and nearly two-thirds of services claimed they had expanded in the last three years. Compared to Brown’s (2003) survey of FGC services in the UK, since 2001 the percentage of local authorities providing FGCs has more than doubled. Brown found that the majority of local authorities introduced FGCs as a tool for implementing the principles of partnership, participation and empowerment. Some laws and policies certainly support such principles and greater family engagement in children’s services (Mitchell, 2020), facilitating the wider implementation of FGCs. For example, the Children Act (1989) emphasises partnership with parents and involving looked after children in decision-making (Edwards et al., 2020), and a key principle of the Social Services and Wellbeing Act Wales (2014) is family voice and control over desired outcomes. However, the Care Crisis Review (Thomas, 2018) noted a move away from the Children Act partnership principle in more recent child welfare policies and practices. Quite possibly the more relevant context to explain the apparent increase in FGC provision is the concern about rising rates of children looked after (e.g. Trowler, 2019).
Results from the survey showed that more than 90 per cent of local authorities that offered FGCs did so at pre-proceedings stage or when the child was being considered for a child protection plan. However, a substantial proportion also offered services for reunification planning (84 per cent) and early help (71 per cent). This study did not explore how many FGCs are held at these stages in these local authorities, but they were, in theory, on offer. Local authorities with more established services were more likely to offer FGCs for reunification planning, early help and children leaving care, than local authorities with less established services. Similarly, local authorities with higher annual rates of FGCs were more likely to offer reunification planning and leaving care services, among a few other categories of FGC. Whilst it may not be surprising that the longer a service has been running or the larger a service is, the more diverse its offerings may be, these distinctions add important contextual insights to the findings that follow.
These next findings about the association between FGC provision and rates of children in care should be interpreted with caution. This is due to the cross-sectional and self-reported nature of the survey and the possibility of multiple intervening variables or confounders which were not accounted for.
Roughly, three quarters of local authorities saw a significant increase in the rates of children in care in the two years after introducing FGCs compared to the two years before. This supports findings from some of the studies from the mixed international evidence base, such as the RCT conducted in the Netherlands by Dijkstra et al. (2019) that found that FGCs resulted in more out-of-home placements than care as usual. However, the amount of FGC provision in a local authority was not statistically significantly associated with rates of children in care.
There were no statistically significant differences in the rates of children starting to be looked after in the two years after an FGC service was introduced. However, there could be other unknown changes over time in these LAs. The recent RCT in England (Taylor et al., 2023) found that children whose families were referred to an FGC were less likely to have care proceedings issued (59 per cent) compared to those not referred (72 per cent) after eighteen months, and an RCT is a more robust design for evaluation of this outcome, allowing for comparison between families who have had FGCs and families in similar circumstances who have not. It may be that more time is needed for an FGC service to achieve an effect of de-escalating intervention at a local authority level if it has been introduced in the context of rising rates of children looked after. The rate of FGCs was also not statistically significantly associated with the rate of children starting to be looked after.
However, our study found that the introduction of FGCs was significantly associated with an increase in kinship foster care rates over a two-year period in nearly 60 per cent of local authorities. McGinn et al.’s (2020) meta-analysis of the effectiveness of FGCs found an overall positive association between FGCs and kinship care, but this was not statistically significant and evidence from studies was heterogeneous. In our study, FGC rates were also associated with rates of kinship foster care, after controlling for deprivation, a 10 per cent increase in the provision of FGCs in a local authority was associated with 2.3 per cent more children in kinship foster care.
Finally, our study found that the introduction of FGCs was not significantly associated with changes in rates of children ceasing to be looked after over a two-year period. However, higher rates of FGCs were significantly associated with higher rates of children leaving care. After deprivation was accounted for, a 10 per cent increase in FGC provision was associated with 1.5 per cent more children ceasing to be looked after. This points to the earlier finding that local authorities with higher rates of FGCs were more likely to offer FGCs for reunification planning and therefore, perhaps more children leaving care to live with their birth families. The increase in exits from care also supports findings from some studies—for example Sheets et al., (2009) and Pennell et al., (2010) both found that exits from care were faster for families that had an FGC compared to standard care.
Study implications
The expansion of FGC services across the UK, as highlighted in this study, offers significant implications for service users, families and the broader public. Notably, the increase in FGC services since 2016, reflecting principles of partnership and empowerment, could potentially signify a shift towards more inclusive family involvement in child welfare decisions. The study's findings, however, present a complex scenario, with no direct correlation between the rate of FGC provision and the rates of children in care, suggesting that the impact of FGCs might be more nuanced and dependent on local contexts and individual circumstances. The rise in kinship foster care rates following the introduction of FGCs points towards a positive move in child welfare towards whole family approaches.
However, it is also worth considering the implications of these findings in context of indigenous populations where FGCs originated (Nygård and Saus, 2019). For example, indigenous FGCs are deeply rooted in community traditions and values, emphasising extended family involvement, collective responsibility and cultural continuity (Ban, 2005), which contrasts with the more generalised application of FGC principles observed in the UK. It is possible that FGCs may not be as effective in the UK, as they may not fully capture the depth of community integration seen in the Māori communities where FGCs originated (Ban, 2005). This underlines the significance of culturally responsive and adaptable FGC models that can effectively incorporate whole family and community approaches and cultural nuances, offering valuable insights for enhancing FGC implementation in various sociocultural settings (Barn and Das, 2016).
Strengths and limitations
Our study provides an updated picture of FGC provision across all the UK nations. It is also the first study to look at the impact of the amount of FGC provision in an area (i.e. FGC rate) on children’s social care outcomes. It also looks at the influence of FGCs at all stages of concern (e.g. early help, pre-proceedings, children in care), whereas the review by Nurmatov et al. (2020) showed that many studies focus more narrowly on child protection and the prevention of out-of-home care. Our study looks at the influence of FGCs on system-level outcomes at the local authority level and is therefore not biased by individual sample selection.
The pre–post analysis was limited by the relevance and availability of historic data before 2013 and therefore only thirty-four local authorities were included in this analysis. The increase in rates of care after the introduction of FGCs could be a reflection of a longer-term trend across the UK of rising rates of children in care.
The study was also reliant on self-report data for information about FGC characteristics, such as the rate of FGCs and so may not be completely accurate. This affects the reliability of RQ3, and these results should therefore be interpreted with caution.
Finally, although deprivation was accounted for in some analyses, detected differences or changes over time may of course be confounded by variables other than FGC provision. The study did not include measures of other facets of anti-oppressive practice, for instance. It is vital to acknowledge the role of race, ethnicity, gender, sexual orientation, disability, and age in the context of FGC services. These intersecting factors, alongside socio-economic status, could significantly influence both access to and outcomes of FGCs, potentially leading to disparities. For example, systemic biases related to race and ethnicity might affect access to and use of support services (Valenti, 2017), whilst gender dynamics could impact family participation and decision-making within FGCs (Holland et al., 2005; Sen et al., 2018). Disabilities and age-specific needs also require attention to ensure equitable participation (Franklin and Sloper, 2006; Edwards et al., 2020). A comprehensive understanding of these diverse factors is crucial, as it may reveal variations in the effectiveness and accessibility of FGC services across different groups.
Conclusion
The recent English RCT (Taylor et al., 2023) provides more robust causal evidence of FGC effectiveness in preventing care proceedings, because of the counterfactual set up by the randomised design. However, our study provides some additional insights into patterns of child welfare system outcomes associated with FGC provision.
Our findings show a mixed picture for England. The introduction of FGCs in a local authority was associated with an increased rate of children in care, but that increase may be due to more children in kinship foster care, a key goal of FGCs where it is not possible for children to stay with their parents. Furthermore, higher rates of FGCs were associated with more children leaving care, perhaps due to reunification with birth families. With proper reunification planning through FGCs, are more children who leave care under these circumstances than remaining with their families? However, as highlighted above, there are limitations to our study, and therefore, to understand in more detail the circumstances of children in and leaving care in local authorities with FGCs, individual data linkage studies need to be undertaken (see our protocol paper for planned work in this area Wood et al.).
It is also important in closing to note the view expressed by some who are strong advocates of FGCs (e.g. Morris and Connolly, 2012) that they are of value in their own right as a mechanism for families to fulfil participation rights, regardless of the outcomes in terms of children being looked after. Family members participating in FGCs will have a wide range of personal outcomes they want to achieve (Mitchell, 2020) and these may or may not include reducing state intervention.
Complexe Systémique: key points
For a systemic reading, the value of this study lies in shifting the focus from the family to the system that receives it: it does not measure what an FGC changes within a family, but what the existence of an FGC service changes in the flows of children in an area. The result is deliberately uncomfortable. After FGCs are introduced, care rates tend to rise, but part of that rise goes through relatives, and the authorities that hold the most conferences see more children return to their families. In other words, the FGC seems less to prevent care than to redistribute responsibility across the wider family network, which is precisely its wager. The limitations are serious and the team owns them: self-reported data, a pre–post analysis of only 34 authorities, a national upward trend in care, and no measure of families' experience. The conclusion is worth keeping in practice: the value of an FGC cannot be reduced to a care figure; it also lies in families' right to take part in decisions. Read alongside the study on what family group conferencing teaches about child welfare outcomes, and the article on family dynamics in kinship care.
Notes from the original
Acknowledgements. We would like to thank the Family Rights Group for their support in promoting the survey and for their valuable input regarding questionnaire content and interpretation of findings. Additionally, we appreciate the guidance provided by FGC managers in shaping the questionnaire content. We are especially thankful to all the respondents who generously dedicated their time to participate in the survey.
Funding. This study is funded by the NIHR Health and Social Care Delivery Research (HSDR) Programme (Grant number: NIHR131866). The views expressed are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care. The CASCADE partnership receives infrastructure funding from Health and Care Research Wales. Jonathan Scourfield receives funding from Health and Care Research Wales as a senior research leader.
Conflict of interest statement. None declared.
References
Ban P. (2005) ‘Aboriginal child placement principle and family group conferences’, Australian Social Work, 58(4), pp. 384–94.
Barn R., Das C. (2016) ‘Family group conferences and cultural competence in social work’, British Journal of Social Work, 46(4), pp. 942–59. https://doi.org/10.1093/bjsw/bcu105
Brown L. (2003) ‘Mainstream or margin? The current use of family group conferences in child welfare practice in the UK’, Child & Family Social Work, 8(4), pp. 331–40. https://doi.org/10.1046/j.1365-2206.2003.00293.x
Bywaters P., Scourfield J., Jones C., Sparks T., Elliott M., Hooper J., McCartan C., Shapira M., Bunting L., Daniel B. (2020) ‘Child welfare inequalities in the four nations of the UK’, Journal of Social Work, 20(2), pp. 193–215. https://doi.org/10.1177/1468017318793479
Connolly M. (2006) ‘Fifteen years of family group conferencing: Coordinators talk about their experiences in Aotearoa New Zealand’, British Journal of Social Work, 36(4), pp. 523–40. https://doi.org/10.1093/bjsw/bch273
Corwin T. W., Maher E. J., Merkel-Holguin L., Allan H., Hollinshead D. M., Fluke J. D. (2020) ‘Increasing Social Support for Child Welfare-Involved Families Through Family Group Conferencing’, The British Journal of Social Work, 50(1), pp. 137–56.
Cosner Berzin S., Thomas K. L., Cohen E. (2007) ‘Assessing Model Fidelity in Two Family Group Decision-Making Programs’, Journal of Social Service Research, 34(2), pp. 55–71. https://doi.org/10.1300/J079v34n02_05
Crampton D. (2007) ‘Research review: Family Group decision making: A promising practice in need of more programme theory and research’, Child & Family Social Work, 12(2), pp. 202–9.
Department for Education (2014) Court orders and pre-proceedings for local authorities, available online at: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/306282/Statutory_guidance_on_court_orders_and_pre-proceedings.pdf (accessed July 31, 2023).
Department for Education (2023a) Children looked after in England including adoptions, available online at: https://explore-education-statistics.service.gov.uk/find-statistics/children-looked-after-in-england-including-adoptions/2022 (accessed July 31, 2023).
Department for Education (2023b) Children looked-after by local authorities in England Guide to the SSDA903 collection 1 April 2022 to 31 March 2023—Version 1.2, available online at: https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/1148938/CLA_SSDA903_2022-23_Guide_Version_1.2.pdf (accessed July 31, 2023).
Dijkstra S., Creemers H. E., Asscher J. J., Deković M., Stams G. J. J. M. (2016) ‘The effectiveness of family group conferencing in youth care: A meta-analysis’, Child Abuse & Neglect, 62, 100–10. https://doi.org/10.1016/j.chiabu.2016.10.017
Dijkstra S., Creemers H. E., van Steensel F. J. A., Deković M., Stams G. J. J. M., Asscher J. J. (2018) ‘Cost-effectiveness of Family Group Conferencing in child welfare: a controlled study’, BMC Public Health, 18(1), pp. 848. https://doi.org/10.1186/s12889-018-5770-5
Dijkstra S., Asscher J. J., Deković M., Stams G. J. J. M., Creemers H. E. (2019) ‘A randomized controlled trial on the effectiveness of family group conferencing in child welfare: Effectiveness, moderators, and level of FGC completion’, Child Maltreatment, 24(2), pp. 137–51. https://doi.org/10.1177/1077559518808221
Edwards D., Parkinson K., Fisher T., Owen J. (2020) ‘Looked after children as decision makers: Family group conferences in practice’, Child Care in Practice, 26(3), pp. 318–30. https://doi.org/10.1080/13575279.2019.1575186
*Family Rights Group (2019) Family Group Conferences and Lifelong Links, available online at: https://web.archive.org/web/20190101000000/https://www.frg.org.uk/involving-families/family-group-conferences (accessed July 31, 2023).
Family Rights Group (2022) Family Group Conference: Quality Standards, available online at: https://frg.org.uk/wp-content/uploads/2023/01/FGC-Quality-Standards.pdf (accessed July 31, 2023).
Franklin A., Sloper P. (2006) ‘Participation of disabled children and young people in decision making within social services departments: A survey of current and recent activities in england’, British Journal of Social Work, 36(5), pp. 723–41. https://doi.org/10.1093/bjsw/bch306
Frost N., Abram F., Burgess H. (2014) ‘Family group conferences: context, process and ways forward’, Child & Family Social Work, 19(4), pp. 480–90.
Holland S., O’Neill S. (2006) ‘We had to be there to make sure it was what we wanted’: Enabling children participation in family decision-making through the Family Group Conference’, Childhood, 13(1), pp. 91–111. https://doi.org/10.1177/0907568206059975
Holland S., Scourfield J., O'Neill S., Pithouse A. (2005) ‘Democratising the family and the state? the case of family group conferences in child welfare’, Journal of Social Policy, 34(1), pp. 59–77. https://doi.org/10.1017/S0047279404008268
McGinn T., Best P., Wilson J., Chereni A., Kamndaya M., Shlonsky A. (2020) ‘Family group decision-making for children at risk of abuse or neglect: A systematic review’, Campbell Systematic Reviews, 16(3), pp. e1088. https://doi.org/10.1002/cl2.v16.3
Metze R. N., Abma T. A., Kwekkeboom R. H. (2015) ‘Family group conferencing: a theoretical underpinning’, Health Care Analysis: Hca: Journal of Health Philosophy and Policy, 23(2), pp. 165–80. https://doi.org/10.1007/s10728-013-0263-2
Mitchell M. (2020) ‘Reimagining child welfare outcomes: Learning from family group conferencing’, Child & Family Social Work, 25(2), pp. 211–20. https://doi.org/10.1111/cfs.v25.2
Morris K., Connolly M. (2012) ‘Family decision making in child welfare: Challenges in developing a knowledge base for practice’, Child Abuse Review, 21(1), pp. 41–52. https://doi.org/10.1002/car.v21.1
Nurmatov U., Foster C., Bezeczky Z., Owen J., El-Banna A., Mann M., Petrou S., Kemp A., Scourfield J., Forrester D., Turley R. (2020) Impact of shared decision-making family meetings on children's out-of-home care, family empowerment and satisfaction: a systematic review, London: What Works Centre for Children's Social Care, available online at: https://whatworks-csc.org.uk/wp-content/uploads/WWCSC_Family_Group_Conferencing_Report.pdf (accessed July 31, 2023).
Nygård R. H., Saus M. (2019) ‘Is family group conferencing a culturally adequate method outside the origin of New Zealand? A systematic review’, Social Work and Social Sciences Review, 20(1), pp. 78–108. https://doi.org/10.1921/swssr.v20i1.1164
Office for National Statistics (2023) Population estimates—local authority based by single year of age, available online at: https://www.nomisweb.co.uk/datasets/pestsyoala (accessed July 31, 2023).
Pennell J., Edwards M., Burford G. (2010) ‘Expedited family group engagement and child permanency’, Children and Youth Services Review, 32(7), pp. 1012–9. https://doi.org/10.1016/j.childyouth.2010.03.029
Sen R., Morris K., Burford G., Featherstone B., Webb C. (2018) ‘When you're sitting in the room with two people one of whom… has bashed the hell out of the other’: Possibilities and challenges in the use of FGCs and restorative approaches following domestic violence’, Children and Youth Services Review, 88, 441–9. https://doi.org/10.1016/j.childyouth.2018.03.027
Sen R., Webb C. (2019) ‘Exploring the declining rates of state social work intervention in an English local authority using Family Group Conferences’, Children and Youth Services Review, 106, 104458. https://doi.org/10.1016/j.childyouth.2019.104458
Sheets J., Wittenstrom K., Fong R., James J., Tecci M., Baumann D. J., Rodriguez C. (2009) ‘Evidence-based practice in family group decision-making for Anglo, African American and Hispanic families’, Children and Youth Services Review, 31(11), pp. 1187–91. https://doi.org/10.1016/j.childyouth.2009.08.003
Stabler L., O'Donnell C., Forrester D., Diaz C., Willis S., Brand S. (2019) Shared decision-making: What is good practice in delivering meetings? Involving families meaningfully in decision-making to keep children safely at home: A rapid realist review, London: What Works Centre for Children's Social Care, available online at: https://whatworks-csc.org.uk/wp-content/uploads/WWCSC_Shared_Decision_Making_Rapid_Realist_Review_full_report.pdf (accessed July 31, 2023).
Taylor S., Blackshaw E., Lawrence H., Stern D., Gilbert L., Raghoo N. (2023) Randomised Controlled Trial of Family Group Conferencing at Pre-Proceedings Stage, Foundations, What Work’s Centre for Children & Families.
The Ministry of Housing, Communities and Local Government (2019) English indices of deprivation 2019, available online at: https://www.gov.uk/government/statistics/english-indices-of-deprivation-2019 (accessed July 31, 2023).
Thomas C. (2018) Care Crisis Review: Factors Contributing to National Increases in Numbers of Looked after Children and Applications for Care Orders, London, Family Rights Group.
Trowler I. (2019) Care Proceedings in England: The Case For Clear Blue Water, Sheffield, University of Sheffield.
Valenti K. (2017) ‘Family Group Conferencing with BME Families in Scotland’, Practice, 29(2), pp. 121–36. https://doi.org/10.1080/09503153.2016.1173667
Wood S., Scourfield J., Au K., Evans R., Jones-Willams D., Lugg-Widger F., McCormack E., Meindl M., Pallmann P., Robling M., Schroeder E., Petrou S., Wilkins D. (2022) A UK-wide Survey of Family Group Conference Provision, Children’s Social Care Research and Development Centre.
Reformatted republication of Family Group Conference Provision in UK Local Authorities and Associations with Children Looked after Rates, by Sophie Wood, Jonathan Scourfield, Melissa Meindl, Kar Man Au, Rhiannon Evans, Delyth Jones-Willams, Fiona Lugg-Widger, Philip Pallmann, Michael Robling, Elizabeth-Ann Schroeder, Stavros Petrou and David Wilkins, The British Journal of Social Work, vol. 54, no 5 (2024), doi: 10.1093/bjsw/bcae019, under a CC BY 4.0 licence. Edition and layout: Complexe Systémique, September 2026 — the work has been modified under the terms of the licence (two tables as lists). Neither the authors nor the publisher are responsible for this edition; the original version prevails.
This is the original article “Family Group Conference Provision in UK Local Authorities and Associations with Children Looked after Rates”, published in The British Journal of Social Work (2024) under a CC BY 4.0 licence. Republished by Complexe Systémique: the author’s text is unchanged; only the presentation has been adapted for reading online, as set out at the head of this page.
Read the original articleHow to cite this article
Wood, S., Scourfield, J., Meindl, M., Au, K. M., Evans, R., Jones-Willams, D., Lugg-Widger, F., Pallmann, P., Robling, M., Schroeder, E.-A., Petrou, S., et Wilkins, D. (2024). Family Group Conference Provision in UK Local Authorities and Associations with Children Looked after Rates. Complexe Systémique. https://app.complexe-systemique.com/en_GB/articles/family-group-conference-provision-in-uk-local-authorities-and-associations-with-children (Original work published in 2024 in The British Journal of Social Work, 54(5), 2045-2066 (2024); republished in 2024 by The British Journal of Social Work, https://academic.oup.com/bjsw/article/54/5/2045/7613004)
To go further
- Repenser les *résultats* en protection de l’enfance : ce qu’enseigne la conférence familialeWhat FGCs change in outcomes.
- Les *dynamiques familiales* dans l’accueil chez des prochesLiving in kinship care.
- Des réunions de décision *menées par la famille* à la place des conférences initiales de protection de l’enfance : résultats d’une évaluation réaliste de l’expérimentation des conférences familiales de protection en AngleterreFGCs in child protection.
Related articles
Article Family group conference provision across the lifespan of child welfare intervention: A mixed-method realist-informed evaluation comparing three stages of welfare concern
Early help, child protection or children in care: among 155 UK families, experiences and outcomes of family group conferences vary very little by stage.
Article How can we BeST? meet the needs of adoptive families? Relevant findings from a major Randomised Controlled Trial of infants in foster care in the UK
The BeST? trial (488 infants in care in the UK) reread for adoption: holistic, ongoing assessment of children and a move to dual registration.
Article Implementing Family Group Decision-Making within statutory child protection social work in Scotland: Challenges and opportunities
Family group decision-making offered automatically in Scottish child protection: families find a voice, social workers doubt the plans, and everyone wants it offered earlier.
Article Parenting after critical illness: a biopsychosocial review of post-intensive care syndrome, family systems, and child outcomes
After post-intensive care syndrome, parenting falters, the co-parent is worn down and children help in silence. A biopsychosocial review proposes how to spot them.