• By Dartington SRU
  • Posted on Friday 11th March, 2011

Why keeping it in the family doesn’t work

If it often takes a long time for policy makers to accept good practice based on firm evidence, it is also sometimes the case – but less well remarked upon – that there is resistance to ditching what has proven not to work. Family group conferences (FGC) are an interesting case in point. There are variations on the method, but the essential ingredient is bringing family members together to solve their own problems. FGCs emerged from an inquiry in New Zealand in the 1980s into why Maori children were over-represented in public systems.They feel right. Once family members are given responsibility for resolving their own difficulties they will be more motivated than professionals to making the resolution work – or so the belief goes. And there is no loss of motivation consequent on professionals taking charge of matters. The idea is that once professionals are out of the room, family members will talk about things they might otherwise have kept secret. Wider family, often excluded or frightened away by child protection processes, are integral to the FGC approach, thus increasing resources brought to bear on the problem.Because FGCs worked with Maori families, the assumption was that they would translate successfully when used with the other ethnic groups. There are many process studies into FGCs, but few that ask whether they work. Of the few that do, the answers are revealing because the results do not give room for optimism. Knut Sundell and Bo Vinnerljung followed up 239 children for three years, 97 of were subject to an FGC and 142 who were dealt with under routine child protection procedures in Sweden. Before the intervention, the FGC children were experiencing more risk than those in the control group, a difference taken into account by the Swedish investigators.Three years later the FGC children had experienced higher levels of re-referral to child protection services, were more often referred by family members because of concerns about maltreatment and they spent longer in placements away from their family home than children in the control group. Sundell and Vinnerljung did find, though, that there was less intrusive involvement by child protection professionals with the FGC children. This is consistent with the approach.Stephanie Cosner Berzin followed up two groups of children randomly allocated to a form of family group conferencing in two California counties. In Fresno, the intervention was used with children from birth to 18 years of age who were at moderate to high risk of maltreatment as assessed using a state-wide decision-making tool. In Riverside County, FGCs were used with children aged two to 12 years of age, placed in foster families or with relatives and who were at risk of moving placements. Berzin found the FGCs had no effect on child safety (by which she means substantiated evidence of maltreatment) or placement stability.Nobody would claim that these evaluations are optimal. Studying processes like FGCs is a relatively new science and there are many complications when an intervention is bound up with complicated legal arrangements demanded by child protection. Nor do these studies say that FGCs are dangerous.But both studies contain comparison groups, have good follow-up procedures and make reasonable attempts to control for bias. At very best the conclusion is the verdict of Sundell and Vinnerljung: that the impact on the approach is scant.There may be reasons why policy makers want to persist with FGCs, but on the present evidence this cannot be because they are effective in comparison with traditional child protection systems. They are not.These two studies are, respectively, seven and five years old. The Swedish study was published in 2004 with the Californian investigation following two years later. But advocacy for FGCs remains strong. In England, despite significant financial cut backs, the Department for Education (DfE) continues to offer support for agencies seeking to promote the approach. In its recent submission to a major review of government policy on family justice, the Association of Directors of Children’s Services, representing 150 chief officers in local authorities in England, called for “the strengthening of family group conference as a means to promote family owned solutions”.The DfE is populated by smart people and directors of children’s services are no fools. So why is there a commitment to something that probably does not work?ReferencesKnut Sundell and Bo Vinnerljung, ‘Outcome of family group conferencing in Sweden: A 3 year follow-up’, Child Abuse and Neglect, 2004, 267-287 Stephanie Cosner Berzin, ‘Using sibling data to understand the impact of family group decision-making on child welfare outcomes’, Child and Youth Services Review, 28, 2006, 1449-1458Links:See: Michael Little's blog on Prevention Action on "Allen Review and Family Group Conferencing..."

Back to Archives