A landmark study comparing the characteristics, psychological needs and stability of children/young people in independent fostering agencies (IFA) versus in-house local authority (IHLA) foster placements using data from the BERRI assessment. It reveals that children in IFAs are typically older and experience greater psychological needs than children in IHLA fostering. Despite these differences, children in both placement types were found to be similarly susceptible to risk, and demonstrated comparable stability and trends in mental health outcomes.

• There often seems to be the assumption that all children in foster care are one homogenous group, and that the two models of delivery (in-house and IFA) can be compared like for like. However, this research shows that children in IFA foster care are older and have more complex needs. This needs to be recognised in wider discussions when comparing costs and delivery models.
• The findings suggest that IFAs perform an alternative function to IHLAs, where they appear capable of providing support that better accommodates children with more complex needs. Where children’s needs are higher, additional support and access to clinical input is appropriate, and IFAs are one model of providing that greater wraparound support.

The findings of this study support the link between events that make the child feel less stable and their behaviour and mental health.

It is already well known that multiple placement changes have a negative impact on children’s behaviour and wellbeing. This study used BERRI data to expand the way we looked at placement instability to include changes in key relationships or the child knowing about a potential change of placement, and shows that these can also function as destabilising factors. This aligns with attachment theory, which shows how critical it is for children to have lasting, attuned relationships, and proposes that repeated disruptions in caregiving relationships are associated with difficulties in emotional regulation and limited identity formation. We know that a lack of enduring secure relationships alters developmental trajectories, with enduring impacts into adulthood. Findings in this new research suggest that Placement Instability and Behavioural and Emotional Wellbeing issues may be interrelated. Therefore interventions addressing both may be most effective, particularly for children in residential settings. However, our literature review showed no existing interventions for children in care that explicitly target the effects of Placement Instability, and almost all existing interventions to support relationships with caregivers focused solely on children in foster care.

• Creating and sustaining lasting relationships for a child can have a large impact on their behaviour and mental health, and in turn impacts their outcomes into adulthood. We need to understand the importance of these stability elements in how care is delivered and commissioned.
• It is critical to measure these destabilising changes in a child’s care journey and to support them to have lasting relationships with empathic caregivers, as well as the most reliable possible relationships with other people who are important to them, such as within their birth family, school, education or professional networks.

This study aimed to investigate the patterns of vulnerability to harm from external antisocial pressures (such as drugs, alcohol, criminal exploitation, sexual exploitation or negative peer influences) alongside risk to self (of self-harm and/or suicide) among children living in staffed settings like residential care and supported accommodation. It found that children in semi-independent accommodation had higher levels of harm from external risk than those in children's homes. It also found that children who experience a greater number of recent life events have higher levels of risk vulnerability in residential care, both in terms of vulnerability to external risk and harm to self. Older children appear vulnerable to greater harm from external risk and girls appear more vulnerable to greater risk to self. Supported accommodation for young people aged 16 to 17 is designed to provide increased independence and lighter touch support as they transition towards independent living. Ofsted rules say that it is not to be used where children have higher levels of need and require care. However, at the moment, BERRI data shows that there are many young people in supported accommodation with higher care needs and higher levels of risk. This research suggests policy makers need to adequately address the harm from external risks that children face in this setting and the appropriateness of supported accommodation as a placement option for this age group.

• Children in care are subject to adultification in many ways. They are typically expected to be autonomous and self-sufficient at much younger ages than children living in families. This is particularly true where children are in residential care, where the “cliff edge” of turning 18 and no longer having a placement funded by children’s services has been long-recognised as a major issue and a time of great risk (with 1/3 of care leavers becoming homeless within the first 3 years of leaving care).
• There have been changes to expand provision for older adolescents since the grooming gangs scandal became headline news. This is a good thing, as before then too many teenagers were being treated as pseudo-adults; left without safe places to live and in exploitation without support. However, the new range of supported accommodation has only just fallen under Ofsted regulations, and the boundary line between children who need residential care and those where accommodation and support would be sufficient, has still not been sufficiently defined.
• This paper highlights the increased risks to this group and the value of having full-time carers who can help to manage risk, compared to just providing accommodation and lighter touch support. However, we recognise that supported accommodation is a category that contains many models of delivering support. Further research is needed to explore how supported lodgings and staying put foster placements compare to semi-independent living services, and how we identify the type of support needed for each young person.

This study looked at the BERRI questionnaire to see whether it was reliably measuring the domains that it claims to cover, and whether the five categories within BERRI (Behaviour, Emotional wellbeing, Risk to self and others, Relationships, and Indicators of psychiatric or neurodevelopmental conditions) are statistically supported. Tests were done to check the structural validity and internal consistency of BERRI. The results showed that the tool had high internal consistency, was statistically strong, and the factor structure was clinically meaningful and related to the existing five categories. 

However it also suggested that

• The boundary between “Behaviour” and “Emotion” should be defined more by whether the items related to externalising (behaviour and dysregulation, caused by exposure to trauma) or internalising (low mood and a negative sense of self) rather than by whether they were actions or thoughts/feelings. 
• “Risk” should be subdivided to recognise the differences between risk to self (of self-harm and suicide) and risks from external antisocial pressures (such as drugs, alcohol, criminal exploitation, sexual exploitation and negative peer influences). 
• “Relationships” should cover variables including social skills, attachment/trust, and care blockers.
  Issues such as sleep/routines and privacy/hygiene might need to be considered separately from the existing headings.

• The analysis showed that BERRI is statistically strong but that a few items should be moved to different sections, and the interpretations we offer in reports could benefit from bringing through the clinical themes that showed in the factor analysis.
• Changes were made to feed this learning into the BERRI assessment. This included enhancing our risk analysis to reflect the different domains and that behaviour is best thought of alongside dysregulation because neurological changes happening in response to trauma mean some people struggle to manage their emotions and tend to go into “fight or flight” mode under stress.

Using BERRI data, this review was the first to establish the type and prevalence of mental health disorders and symptoms among children in residential care. Almost half of children living in residential care had behaviours indicative of a clinical level of mental health concern and trauma-related symptoms were common among the children. It suggests that children and adolescents with higher levels of mental health symptoms and disorders might well be managed in residential care settings, given the higher levels of support.

• More needs to be done to identify, track and support the mental health needs of children in care; further fueling BERRI’s mission!