Dance Like a Butterfly and Sting Like a Bee: A Preliminary Evaluation of a Six-Session Group for Foster-Carers and Adopters of Children with Attachment Difficulties

Due to early trauma, looked after children and young people often demonstrate a high level of emotional and behavioural difficulties which can be challenging for foster-carers to manage long term. In order to reduce placement breakdown, which will have a further significant negative impact on the young person’s psychological wellbeing, carers need to be supported to provide care that is attuned, reflective and responsive. The primary aim of this study was to examine qualitatively and quantitatively the effects of a six-week attachment-focused group for foster-carers and adopters on placement security, family and child distress and carer confidence in managing difficult behaviour. Thirty five foster-carers and adopters completed a series of visual analogue scales before and after attending the groups. Qualitative feedback was also collected. Qualitative and quantitative analyses were employed. Following the group, there was a significant increase in carers’ confidence and a significant decrease in distress caused to the family by the problem behaviours. There was no significant change in carer-rated child distress or placement security. Specific themes extracted from the qualitative analysis included: Learning and acquiring practical skills and developing the ‘attitude’ for reflective capacity about self and actions, reflecting about the child, challenges, learning environment and increases in confidence. Brief attachment-focused groups may increase foster-carer’s and adopter’s confidence in caring for children with attachment difficulties and may reduce familial distress. Future research should investigate how to accurately measure changes in reflective capacity of carers and whether such changes translate into child behaviour change.


Introduction
Due to prior experiences of trauma and loss, children in care of the local authority (commonly referred to as 'looked after children') and adopted children often display behaviour which is a challenge for carers to understand and manage [1]. A number of epidemiological studies have documented the high rates of emotional and behavioural problems in this population [2,3] and such problems reliably predict placement breakdown [4]. The disruption that is caused by placement breakdown has been shown to exacerbate existing behavioural problems [5] and such negative outcomes contribute to a self-perpetuating cycle in which subsequent placement break down becomes more likely.
In order to reduce the likelihood of placement disruption, fostercarers and adopters need to be supported and equipped to provide sensitive and attuned parenting to the children in their care. Group work is generally thought of as a good way to support parents; as well as being an efficient use of limited resources; it provides the opportunity for individuals to learn from others' experiences. For children who live with their biological families, there are a number of well validated group parent training programmes, such as The Incredible Year Parenting Programme [6]. National guidelines recommend such parent training programmes for families where there are children with conduct or attention-deficit problems [7]. These programmes tend to draw on principles from Social Learning Theory (SLT). In brief, SLT proposes that behaviour is initiated and reinforced by the social environment; therefore if a parent changes their behaviour, then the child's behaviour will adjust accordingly [8]. Whilst principles from SLT may also be applicable to looked after and adopted children, there is also a need to account for the significant trauma and attachment disruption that children who have been removed from their biological families are likely to have experienced. Attachment Theory [9,10] provides a useful framework for understanding the behavioural and emotional manifestations of early trauma and thus is applicable to looked after and adopted children as well as their carer's. Central to Attachment Theory is the importance of early relationships and experiences in shaping a child's development. When the bond with the carer enables a child to feel safe and protected then a child will develop a secure attachment and will seek proximity to their care giver in the face of real or perceived threat. When early relationships and experiences do not provide safety or comfort, as is often the case for children who become looked after, children are more likely to demonstrate insecure attachment patterns characterised by ambivalent or avoidant behaviour towards caregivers. As insecurely attached children have not had the experience of caregivers providing security in the face of threat, they develop strategies to survive their early environment [11]. Whilst having such strategies is adaptive within the environment in which they are developed, problems arise when the children are removed and placed into a safe environment as they are likely to perceive many common family situations as if they posed a theat. Examples include carers telling a child to do something or imposing some control on their behaviours. Children who have not learnt that adult control is usually safe and reliable may show avoidant or challenging responses. Behaviours such as lying, violence and sexualised behaviour are often hard to understand and a challenge to manage by carers and professionals but can be understood using Attachment Theory.

Journal of Psychology & Psychotherapy
Providing reflective, attuned and responsive care that is informed by the child's history and aimed at helping the child develop a more secure sense of themselves and relationships with others has been shown to be the most effective long term intervention for looked after children [12]. A number of group-based interventions, which draw on principles from both SLT and Attachment Theory, have been developed to achieve this [13][14][15][16][17][18]. For example, The Fostering Attachments Group [13] is an 18 session group which consists of three modules: Attachment Theory, developing a secure base and building relationships and managing behaviour. Preliminary evaluations of the group have shown high levels of carer satisfaction, although actual change in the child's behaviour has been harder to demonstrate [12]. More recent evaluations have indicated some sustained changes in carer-rated improvement in specific behaviours exhibited by the child [14]. Whilst results are promising, the length of existing programmes can make them hard to offer as part of a routine clinical service. Therefore, research is needed to establish whether programmes shorter in length, and therefore more cost-effective for both carers and professionals, can have equally good outcomes. In addition, existing studies have tended to evaluate the programmes offered to either foster carers or adoptive parents. Services who work with both foster carers and adoptive parents may find it hard to extrapolate from such studies. This paper therefore reports on a preliminary evaluation of a group shorter in duration and for a mixed group of foster carers, kinship carers and adoptive parents.

Service context
Thinking Allowed (TA) is a specialist part of the child and adolescent mental health service (CAMHS) in Bristol, UK, that provides a clinical and consultation service to the Local Authority's looked after and adopted children, their families and carers. Bristol looks after over 700 children at any one time with approximately 200 entering and leaving the care system each year. The local authority makes about 40 adoptive placements per year and there are two independent adoption agencies in the region which provide placements to an additional 20 children per year. TA's clinical responsibility is defined by the population and has no geographical limits.
Each year approximately 200 referrals are processed in TA using a Choice and Partnership Approach system. The aim of the initial 'choice appointment' is to formulate a shared understanding of the looked after or adopted child's relevant difficulties based on case information as well as an extended clinical interview with the carers, social worker and other professionals. Following the initial choice, 'partnership consultations' are usually offered to the carers. If at the choice appointment it is decided that the young person would benefit from individual therapy, this is sought from community CAMHS or provided by the TA team if the young person's placement circumstances are too unsettled. The parenting model proposed by Hughes draws on both SLT and Attachment Theory and emphasises the importance of playfulness, acceptance, curiosity and empathy (PACE) in caring looked after children is particularly influential in the team's clinical work [19,20].
TA also offers attachment-focused training for carers called 'Why and How' to carers. This introduces carers to Attachment Theory, uses this to explain the emotional and behavioural manifestations of attachment difficulties, considers how the theory influences caring for looked after children and introduces PACE (most effectively described by [19]. A smaller workshop group called 'Dance like a Butterfly and Sting like a Bee' (henceforth Butterflies and Bees) has been developed by TA to help carers apply PACE in the home environment. The overall aims of Butterflies and Bees are to: • Increase carers' capacity to understand the children in their care and the reasons behind their difficulties so that they can provide reflective, attuned and responsive parenting.
• Increase the skills of the carers in delivering therapeutic reparenting through the practical application of PACE in their every-day parenting.
• Help carers feel more confident in caring for children with attachment difficulties.
The aim of this preliminary study is to describe and evaluate the effect of Butterflies and Bees on child and family distress, carer confidence in managing difficult behaviour and placement security.

Materials and Method Participants
Seven groups of five foster-carers and adopters completed Butterflies and Bees between 2012 and 2014. The inclusion criteria were as follows: being a foster or kinship carer or adopter, completion of Why and How and currently having a looked after or adopted child or young person placed with them. The majority of carers who completed the training were female (N=26). The group series were specifically aimed at children up to 11years old or those over 11 years. The majority of participants were foster-carers (N=27). One carer identified themselves as a kinship carer and four as adopters [1]. Twenty-two carers completed the over 11 years old series and 12 completed the series for younger children [1].

Overview of butterflies and bees
The group took place over six, weekly, two-hour sessions. Each session had an overriding theme, however, carers were encouraged to bring to the sessions material from their week and the facilitators worked with the material presented by the carers. Broadly, the first session revised Attachment Theory and what children learn in their early experiences about themselves, adults and relationships. This was linked to the behaviours that the children exhibit now. Carers were encouraged to draw parallels with their own emotional responses to events and situations (their 'red buttons'). The therapeutic stance of PACE was re-presented. After the first session, homework was set for the carers; they were invited to look at a small sample of behaviours in their children, consider where they originated in the context of what they knew about the children's history. The next session introduced the idea of self-care including managing their 'red buttons' . From Session 2, role plays were used to give carers the opportunity to practice responding to difficult situations using the PACE approach. Other topics which were covered included: understanding and working with children who do not like physical touch, providing discipline with empathy and understanding, interactive repair after a relationship rupture [21], noticing and challenging carers' negative automatic thoughts, recognising and responding to shame [20]. In the final two sessions, carers were supported to complete a formulation of the child or children in their care by drawing on the different principles which had been discussed in previous weeks.
The manner in which sessions were delivered was integral to the approach that was being promoted. In line with PACE, facilitators modelled using empathy, acceptance and curiosity. Admitting mistakes or difficulties was also important and this was often done using humour and playfulness. Whilst the sessions had a loose structure and carers were provided with a work book for the sessions, it was important for the facilitators to be flexible and use the material that the carers brought each week; the facilitators therefore modelled and promoted a reflective state of mind [15].

Outcome measures
Visual analogue scales: Four visual analogue scales were developed by one of the named authors to measure the effect of the group. Carers were each invited to identify up to three difficult behaviours. For each of the behaviours, carers rated the following: 'How confident am I dealing with this behaviour?' (from unconfident to confident); 'How much distress does it cause my family' (from great distress to no distress); 'How much distress does it cause my child' (from great distress to no distress). The fourth question asked carers to rate: 'How secure is this placement' (from secure to insecure).
Qualitative feedback: Carers were invited to provide qualitative feedback each session. The evaluation sheet contained four open-ended questions which asked about the most and least helpful aspects of the group, use of the skills gained in the group and suggested changes to the group programme. There was also a space for any additional comments.

Data analysis
Quantitative analysis: Ratings on the visual analogue scales were measured. As the majority of carers identified more than one difficult behaviour, a mean rating was calculated per carer. Examples of behaviours identified by carers included: lying, stealing, anger and promiscuity. Paired samples t-tests were used to compare pre-group and post-group scores of the items on the rating scales.
Qualitative analysis: Written feedback from the carers was transcribed by a named author (TWJ). Data were hand coded and analysed thematically using a six-stage process: familiarisation with the data; generation of initial codes; searching for themes; reviewing of themes; defining and naming of themes; and finally writing a report. Analysis in this study was done by two of the authors (TWJ and PN). Validation of themes was sought through inter-rater reliability with two other authors of the paper (AL and AF). Following credibility checking, the interpretation of the data and the conclusions drawn were made in a Thinking Allowed team meeting.

Quantitative results
Carers reports indicated that they felt more confident in dealing with the child's behaviour having attended the group to a significant degree (before group, M=51.3, SD=19.7; after group, M=70.1, SD=16.7; t (34) = -5.74, p<0.001). They reported a significant decrease in the distress caused to the family by the behaviours after attending the group (before group, M=42, SD=22.9; after group, M=57.2, SD=20.9; t (34) = -3.70, p ≤ 0.001). At the group level, there were no significant differences in carers' ratings of the distress caused to the child by the behaviour or the security of the placement [1] although when the rating scales were examined at an individual level, some carers did note an increase in child distress after the group.

Qualitative results
The thematic analysis resulted in one transcending theme and six sub themes. The theme that transcended all the data was 'Thinking and reflecting within the group sessions' . Within this transcending theme, there were embedded six sub themes: 'Learning and acquiring practical skills and developing the attitude' , 'Reflecting about self and actions' , 'Reflecting about the child' , 'Challenges' , 'Learning environment' , and 'Increases in confidence' . Representative quotations have been used throughout the results section to illustrate the themes.
Thinking and reflecting: What was evident throughout the majority of the feedback forms was that carers felt that the group had been an opportunity for them to think about and reflect on their role as caregivers for looked after and adopted children and young people. Reflecting about the child: Comments illustrated that the process of the group, and knowledge gained, helped carers to reflect on the child's behaviours and emotions in their context of their individual histories. A number of carers reflected that the group enabled them to think more deeply about their understanding of the children in their care.
'Very helpful looking at child's history, looking behind the behaviours, strategies very useful. ' (1:4) 'Greater understanding of how to deal with certain situations. It keeps fresh in mind the feelings of the child. ' (2:2) 'Understanding the feelings of being complimented for these youngsters, how it could make them feel and how different that is to what we would take for granted. ' (2:5) 'Reminder of attachment helpful when understanding behaviour' (4:1) 'I will think more about the reason for [Child's name] behaviour' (4:4) 'I hope to understand more about how the shame / guilt mechanism affects my daughter's behaviour over the next weeks' (4:5) ' Accepting the child's past' (4:6) 'Thinking about the child's red buttons. Thinking about how they might be feeling' (6:3) 'Think about how [Child's name] past can affect her behaviour' (6:4) 'Recognising behaviours as possibly resulting from shame' (6:5) 'Help me understand my children' (7) Helpful learning environment: Carers commented on the nature of being in a group having the opportunity to share experiences. A number of carers seemed to find this both supportive and helpful for their learning.
'I felt the entire group were able to take a part and share information. ' (1:1) 'Feel comfortable to express my views and interested in those of the other carers, as I find this the best way to learn from their experiences. ' (2:2) 'Listening to others and maybe using some of their experiences in my placement if appropriate' (6:1) 'Learning from each other, different views' (7:1) Increase in confidence: Carers indicated that thinking and reflecting in the group increased their confidence in providing good enough care. As well as learning practical skills and strategies, carers seemed more confident with not always knowing the answer or being able to fix the situation.

'I have been given so many different ideas enabling me to help the children with emotional problems and development I would reflect back on my work and have learnt different approaches' (1:6)
'Overall the course enabled me to look reflect and feel that I am doing a good job and that I am not on my own. Also to remember to stay with the feeling. ' (1:6) 'I/we are not alone -others have similar problems, gave me confidence. ' (5:1) ' Able to deal with behaviours much more confidently' (6:6) Challenging: Whilst the majority of the feedback regarding the group was positive, some carers commented on how challenging they had found thinking about and reflecting on the children in the group.
'Very hard, got you thinking about other ways to deal with situations' (1:2) 'The practical exercises were very hard this week but an eye opener' (1:4) 'It was emotionally draining' (1:4)

Discussion
The aim of this study was to conduct a preliminary evaluation of Butterflies and Bees, a six session attachment focused group for fostercarers and adopters. Whilst there was no significant change in placement security which remained high throughout, following the group, carers felt significantly more confident in managing difficult behaviours and reported a reduction in familial distress caused by the children's behaviours. The qualitative analysis indicated that carers felt they were more able to reflect on the reasons behind both their own and their child's behaviour but also those they had learnt and developed practical skills. Feedback from the carers also suggested that the group format provided a useful learning environment, although more challenging aspects of attending Butterflies and Bees were highlighted by some.
A qualitative study undertaken by Samrai, Beinhart and Harper [22] explored foster-carers' perceptions of placement support and success. Receiving support from services, including training on specialist topics such as the impact of early separation and loss, was perceived as useful by carers and they felt it contributed to placement success [22]. Previous studies have hypothesised that increasing carers' confidence is one mechanism through which training contributes to placement success [23,24]. In the current study, both the qualitative and quantitative analysis demonstrated that carers felt more confident following the group in managing difficult behaviours. Despite this, there was no significant change in placement security. This is in line with other group evaluations. For example, Herbert and Wookey [23] reported a significant increase in carer confidence after a cognitive behavioural group programme but no significant change in placement security. It may therefore be that increasing confidence of the carers does not necessarily equate to changes in placement security. However, carers who attended Butterflies and Bees rated the stability of the placement as high before and after the group and therefore the lack of change in placement stability may be attributable to ceiling effects. Carers who deem a placement as stable to start with may be more willing to commit to a group programme such as Butterflies and Bees.
The process of developing and maintaining reflective capacity has been attributed to the placement stability of children in foster care and the ability of their carer's and children to gain insight and understanding to their internal experiences which in turn effects their ability to identify, monitor and regular their emotional and behavioural experiences [25]. Current theory and empirical research has identified that developing carers' reflective capacity for thinking about their child's state of mind and the reasons behind their behaviour promotes resilience and psychosocial development in the child [15,26,27]. In addition, qualitative research suggests that carers' also need to have the capacity to reflect on their own behaviours as traumatised children can often arouse feelings of inadequacy and insecurity in caregivers which in turn may activate the carers own attachment behaviours [28,29]. One possible conclusion from the current findings is therefore that carers' increased confidence is related to their enhanced capacity to reflect. Reflecting on the learning and emotions underlying the child's behaviours as well as their own emotions and how their responses might be perceived were key themes that emerged from the qualitative analysis.
In recognition of the importance of carers' reflective capacity, a number of group interventions for foster and adoptive carers cite increasing reflective capacity of carers as the main aim [15]. However, evaluations of such groups have not yet been able to show reliably that this as the mechanism of change. One study attempted to measure mind-mindedness, a construct similar to reflective capacity, before and after the Fostering Attachments Group and found no significant change in the mentality attributed to the child by the caregiver [14]. Bammens and colleagues evaluated the effect of a brief mentalisation-based psychoeducational programme on carers reflective functioning [18]. Compared to a control group, carers who completed the mentalisation group demonstrated a greater capacity to think reflectively about themselves and their children in a speech task. The authors concluded that increasing the reflective capacity of the carers through the group intervention should in turn should enhance and strengthen the understanding and the relationships between the carers and the child and reduce negative outcomes [18]. Unfortunately, this was not examined. Future research should consider how to accurately capture abstract concepts such as reflective capacity which may contribute to carer confidence, carer-child relationships and placement stability. It will also be important to consider how best to examine whether changes in reflective capacity of the carer translates into child behaviour change.
Foster family stress has been associated with placement instability.
For example, in a recent study which surveyed over 600 foster-carers about factors which may influence the decision to continue fostering, foster-carers with higher levels of tension in the family reported more intention to leave fostering [30]. In the current study, it was demonstrated that a relatively brief group training led to a significant decrease in carer rated familial distress. Brief attachment-focused training, such as Butterflies and Bees, may therefore be one way to improve carer retention. Reducing stress in the family may also allow foster-carers to provide more reflective and attuned care and therefore feel more confident in their ability to parent the children in their home.
Qualitative feedback from the carers in the current study indicated that they benefited from the non-specific aspects of the group such as learning from others' experiences. Cohesion and mutual support within a group setting have been shown to be of critical importance in enabling personal exploration [15]. However, some carers commented on how challenging they had found thinking about and reflecting on themselves and the young people they care for in the group. Having the time to think and reflect can mean that people may engage with uncertain and uncomfortable experiences which are typically shut off from conscious awareness [31]. Paradoxically, experiencing this uncertainty and discomfort can leave people more resilient and available to those around [31]. The children with whom foster-carers and adoptive parents work with will be best served by carers who are well but grounded and reflective within themselves, rather than burntout and defended against their own, and others' , pain and distress. A group such as Butterflies and Bees may therefore offer a space in which to be curious about themselves and their work.
The results of this study have important implications for services working with looked after children and their families. The group format is advantageous as a large numbers of carers can be provided with specialist attachment-focused training relatively efficiently. In addition, the group was short in duration. Butterflies and Bees is therefore a cost-effective option for resource limited services. In addition, the group programme can be tailored for foster-carers, kinship carers and adopters. In terms of service provision, this is important as it had been reported that kinship carers and adoptive carers do not receive the same level of ongoing training and support as foster-carers despite them caring for children with a similar level of need [32,33]. As demonstrated in this study, providing support in a group format may be one way that services are able to offer support to a diverse group of carers looking after children at different stages of development. The manual for Butterflies and Bees will continue to be refined so that it can be disseminated more widely.

Limitations
There were limitations to this study. Without a control group it is not possible to determine whether the effects were attributable to anything more than non-specific group effects. The sample of carers was relatively small and diverse which limits the generalizability and applicability of the results. A larger number of carers would enable the effects of a range of potentially important variables to be examined, such as the child's age at removal, number of previous placements and years of carers' experience. Carers were not followed up and therefore it is not known whether the observed changes in confidence and distress were maintained over time. This is something we plan to address in subsequent group evaluations. The main outcomes of interest were measured using a non-validated rating scale. Lastly, the impact on the child's behaviour was not directly measured. Further research is therefore planned to establish whether the children benefit from the carers attending the group as well as the best way to measure carer's perceived changes following the group.

Conclusion
Butterflies and Bees are six session attachment-focused group training aimed at carers of looked after and adopted children. Whilst caution has to be taken due to the limitations of the preliminary evaluation, this study illustrates that a brief training can increase carers' confidence, reflective capacity and provides them with practical skills to aid the therapeutic re-parenting of children who have experienced significant early trauma. Future research will aim to establish the process of change during the group and whether such changes influence the child's behaviour.