Fostering attachments in looked after children: further insight into the group-based programme for foster carers.

Laybourne, Gemma ; Andersen, Jill ; Sands, John 等

Introduction

Many children who are placed in local authority care have been the victims of abuse and neglect (Quinton et al, 1998). Research indicates that they have significantly higher rates of mental health difficulties than children in the general population and an especially high incidence of disruptive behaviours (McCann et al, 1996; Dimigen et al, 1999). Many of these vulnerable children are cared for in family placements with foster carers. Sinclair et al (2000) report that the two most common causes of placement breakdown are the child's difficult behaviour and the foster carer's lack of confidence to manage the child. As a result, carer training programmes have been developed and evaluated with varying results.

Initially, parent training programmes based heavily on social learning theory principles, developed for birth parents of children displaying challenging behaviours (such as Webster-Stratton, 1997), were identified as a potential source of training for foster carers. While these programmes have been methodically and comprehensively evaluated to yield positive effects when delivered to birth parents, there is limited evidence to suggest their benefits for carers trying to parent looked after children who have typically suffered trauma and abuse (Allen and Vostanis, 2005).

Hill-Tout et al (2003) subsequently found that providing support to foster carers in terms of improving their knowledge base, skills and confidence to manage the complex and challenging issues often displayed by the children in their care is acknowledged as fundamental to placement stability and positive care outcomes. In an attempt to overcome the identified shortcomings of previous training programmes, researchers have focused on the need to adapt existing social learning theory-driven training programmes. Specifically, this has involved acknowledging the importance of attachment difficulties, which are understood to be at the root of many of the emotional and behavioural difficulties experienced by children within the looked after system (Fahlberg, 1996; Hughes, 1997; Howe and Fearnley, 1999). Indeed, Golding and Picken (2004) have reported outcomes of a training programme based on attachment theory and related parenting techniques. Preliminary evaluation of this training detailed carers reporting they had a better understanding of the child's difficulties and improved parent--child interaction. In addition, they found that the children were displaying fewer conduct problems.

South Tyneside is a small local authority with a well-established format for multi-agency training and strong links between South Tyneside Fostering Service and the Clinical Psychology for Looked After Children Service (CPLAC). Jointly, these services had continued to think about how best to meet the needs of, and secure positive outcomes for, the group of children who present especially challenging behaviour rooted in attachment difficulties; also how to train and support the foster carers charged with the task of caring for these children on a daily basis.

Following on from Golding and Picken's (2004) preliminary programme evaluation came the release of the training programme, available for purchase in CD format under the title Fostering Attachments in Children who are Looked After and Adopted (Golding, 2006). This release coincided with an investigation by South Tyneside local authority into resources available to enhance the training and support offered to carers. South Tyneside clinical psychology and fostering service professionals were reviewing training developments and were keen to look at whether the initial positive evaluation findings identified by Golding and colleagues could be replicated within our local authority.

The present article describes how the authors co-ordinated their own implementation of the Golding (2006) attachment-focused training programme and conducted a comprehensive evaluation of it. Our evaluation is specifically focused on the attachment training programme originally produced and evaluated (Golding and Picken, 2004; Golding, 2006) and was undertaken through a replication of the quantitative evaluation measures implemented by Golding. In addition, a qualitative methodology was employed to explore the views and perceptions of foster carers who had been involved in the training.

The training programme

Content and structure

This training programme used the manual developed by Golding (2006) which, she reports, was written using ideas from a number of different authors (Fahlberg, 1996; Hughes, 1997; Delaney, 1998; Howe et al, 1999).

Module 1 focused upon attachment theory and the development of different patterns of attachment. Carers were encouraged to think about these patterns as adaptive to the child's early environment and how certain behaviours developed around this time continue to be maintained as they provide a level of safety and security for the child living in an adverse environment. The Dynamic-Maturational Model suggests that children who are not safe learn to behave in ways that will increase their feeling of safety, and that this may take the form of eliciting adult protection or reducing the possibility of adult threat (Crittenden et al, 2001). Small and large group discussion and specific exercises helped the carers to link this theory to their own foster children, enhancing their understanding of the particular behaviours and emotions displayed.

Modules 2 and 3 moved on to consider the application of attachment theory to practice. Special consideration was given to the parenting model employed by Dan Hughes (1997). In this model, carers are encouraged to provide a positive family atmosphere where they are in a position to be able to control the emotional rhythm of the house. The model also considers ways to avoid being drawn into confrontation that resembles the child's early experience. Through this, the carers are encouraged to think and learn about how to create a secure base within which the child may potentially develop the ability to regulate their own emotions. Of particular focus is the importance of emotional attunement and how discipline is recognised as important but should be accompanied by empathy. While much of the training focused upon the child and parenting him or her, part of it also addressed the needs of the carers, identifying their own personal stressors, when to seek out support and the importance of looking after themselves.

Training methods

The training, which comprised 16 sessions, took place on a weekly basis over approximately six months (allowing for school holidays), with carers attending for morning sessions comprising three hours with regular breaks. Training ended each week with a group lunch. This seemed to suit carers and acted as a useful debrief and opportunity for them to reflect on the training together.

Wherever possible, the facilitators tried to use the methods recommended for each session outlined in the manual; these involved direct teaching components of the course, role plays, case study scenarios, group exercises, use of video clips and book excerpts. One of the main methods employed was group discussion. Specifically, this involved carers thinking about the material from the manual delivered by the facilitators and then providing details regarding incidents and examples from their own direct experiences with foster children. Often, the discussions that took place around these 'real-life' examples appeared to be a useful way of helping the carers to understand some of the concepts covered in the training, arguably creating appropriate space to think and voice their opinions.

Programme facilitators

Facilitators of the programme were seconded from local psychology and fostering services. They comprised a clinical psychologist, an assistant psychologist involved with looked after children and a fostering support officer, each of whom had extensive previous experience of group facilitation within their relevant disciplines. Facilitators from both services worked hard to combine knowledge and skills to provide a broad range of expertise to the training group.

Participant selection

Fostering and clinical psychology services identified eligible foster carers potentially as those experiencing significant placement difficulties and/or those with children identified as having significant attachment difficulties. Twelve carers were selected in this way, each of whom received a home visit from their fostering support officer and one of the programme facilitators. This involved a brief discussion of what the programme entailed, obtaining a brief history of the child from the carer's perspective (to aid facilitators' understanding of each child and the carer's views of the child's difficulties), and what they hoped to gain from the training course. Each carer was asked to consider whether they could commit to the first module (six sessions). They were advised there would be a review after module 1 with the potential for them to continue through modules 2 and 3 should we all (carer and group facilitators) agree this to be appropriate. Following from this home visit, two carers felt that they could not commit to the course at that time (due to other training commitments) and the group facilitators agreed to enlist the remaining ten.

Participants

All ten carers (two male, eight female) successfully completed module 1. The same group agreed to continue on with modules 2 and 3 but due to unforeseen circumstances two of them were unable to finish the course. So, in total eight carers completed modules 1, 2 and 3, an 18-session programme (six sessions per module) delivered over six months. Each carer missed no more than three of the 18 sessions with the majority missing two or less. Evaluation material pre and post programme is available for seven of the eight carers.

Programme evaluation: methods

Verbal feedback

As this was the first time South Tyne-side had used the programme manual, it was agreed by the facilitators that the primary focus of the evaluation would be on ensuring that the delivery and content of the programme were appropriate and comprehensible to the carers. With this in mind, facilitators regularly requested verbal feedback from the carers regarding how they were finding the programme and their understanding of the material. This feedback was obtained through the use of a regular timeslot allocated at the end of each session for carers to discuss their views and opinions on content and structuring of the course. This was then recorded by one of the facilitators and used as a weekly discussion point in the planning meetings, particularly in terms of how best to deliver the next section of the training manual.

Quantitative evaluation measures

A second focus was on evaluating the programme objectively through administering a series of questionnaires (recommended in the manual and previously used to evaluate the programme by Golding and Picken, 2004). These were completed anonymously by carers in the first session of module 1 and then again at the final session of module 3. The following measures were employed (both pre- and post-group):

Parenting Stress Index/Short Form (Abidin, 1995)

This is a direct derivative of the Parenting Stress Index full-length test. A brief questionnaire that focuses on three key areas within the parent-child system: the parent, the child and their interactions, and the self-reported stress the parent experiences in each of the three domains.

Strengths and Difficulties Questionnaire

This is a short 25-item behavioural screening questionnaire for common child mental health problems. It was completed at the beginning and the end of the group. It assesses the level of behavioural difficulty presented by the child in four areas (emotional, conduct, hyperactivity and peer relationships), as well as the level of pro-social behaviour displayed. This assessment is now widely used within Child and Adolescent Mental Health Services (CAMHS) nationally and has proven reliability and validity (Goodman, 1997). It has also been used in similar populations to that discussed in this article (Minnis and Devine, 2001; Pallett et al, 2002)

Relationship Problems Questionnaire (Minnis et al, 1999)

This is an 18-item parent-report questionnaire, used to assess Reactive Attachment Disorder (RAD) behaviours, a psychological disorder sometimes found in young children who have failed to form attachments with one or more primary caregivers or who have had this attachment severely disrupted (Hanson and Spratt, 2000).

Intervention Questionnaire (Golding and Picken, 2004)

This self-report questionnaire was devised by the project team who produced the manual. Respondents are required to rate aspects of their care of and relationship with the child at the beginning and end of the group. The questionnaire generates a combined score (range 10-60) based upon ten self-report ratings of how well the carer understands the child, how well they respond to the child, how rewarding they find the child and how confident they feel in caring for him or her.

Qualitative evaluation--carer responses

The third focus of the programme evaluation utilised qualitative methodology. After completing the programme, six carers were visited at their home by a psychology professional independent of the programme. This entailed a semi-structured interview that lasted around 30 minutes whereby carer responses were recorded verbatim. The aim was to allow carers to discuss what they had learned from the training, any particularly helpful/useful aspects and what they had found difficult. It also explored whether they believed the training had made any difference to their parenting of their foster children and if they felt they still had skills requiring further development. Data were then analysed using thematic analysis whereby the recorded information was read through numerous times and relevant concepts coded. From further analysis of the coding, a number of themes emerged.

Programme evaluation: findings

Verbal feedback

The verbal feedback component of the evaluation proved most helpful when it came to considering areas that carers were struggling with and that may have gone unnoticed had we not incorporated this evaluative process into the training. In particular, carers highlighted difficulty understanding some issues in module 1 related to the terminology of certain attachment concepts, such as different attachment patterns and the meaning of the internal working model. Contemplating this feedback, facilitators were able to integrate this into following sessions, where necessary providing a re-cap of the previous week's training or spending part of the session revisiting some of the issues and concepts that carers found difficult. In terms of modules 2 and 3, verbal feedback identified areas where carers had struggled when attempting some of the interventions that had been considered in the previous week's training and where they wanted further guidance and support on how they had managed in these situations. This information allowed facilitators to 'fine-tune' the next session to consider some of the practical issues raised in terms of managing the children's behaviour. For instance, one issue raised many times by carers was the appropriateness of empathy in certain situations. Identification of this issue enabled facilitators to provide more detailed information on when and how empathic responses may be used and the times when carers ought to remain focused on managing behaviour at the expense of contemplating empathy.

Further to this, the feedback worked well as a forum to reflect on wider issues that would not have been identified using only quantitative measures. For instance, it became apparent within module 2 that some carers were increasingly reporting a degree of difficulty in their spouse/partner's ability to understand the carer's parenting of the child as it shifted away from traditional parenting methods. Several carers reported this affecting their relationships and facilitators were pleased that they had been alerted to this issue and were able to start contemplating ways to tackle it.

Quantitative evaluation measures

For the purpose of the objective component of the evaluation, statistical analysis of pre-and post-training evaluation measures took place. Specifically, t-tests were applied to all scales incorporated into each of the questionnaires for the seven carers who completed the pre-and post-programme evaluation pack. A summary of these analyses are presented. Further, in relation to the qualitative interview, carer responses were analysed and any emerging themes highlighted.

Parenting Stress Index/Short Form (PSI/SF)

A comparison between the pre-and post-programme scores of the PSI/SF showed positive results. Statistically significant decreases in post-programme scores, when compared with pre-programme scores, were observed for the total stress scale and the parental distress, both revealing post-training improvements and a large effect size. No statistically significant difference was observed for the difficult child subscale, although a marked decrease in post scores when compared to pre scores was noted (see Table 1 for a comparison of pre-and post-mean scores).

The total stress score is designed to provide an indication of the overall level of stress an individual parent is experiencing. The parental distress subscale determines the distress a parent is experiencing in his or her role as a parent, as a function of personal factors that are directly related to parenting. The fact that these scales/subscales have significantly reduced after completing the programme, we believe, is extremely important. It is possible to speculate from these findings that the programme helped carers understand better the children and the issues they present, which is helpful in reducing stress. As the course progressed, facilitators also identified a reduction in carers blaming themselves directly for the difficulties presented by the children in their care. It is possible that this reduction in self-blame reduced their self-reported stress. The increased understanding and empathy for the difficulties that carers experience, combined with the shared experiences of their own personal difficulties, were potentially very supportive, ultimately alleviating possibly damaging stress.

Such findings are particularly encouraging as there is research to suggest that placements frequently break down as a result of stress in carers and feelings that they are unable to cope and manage children experiencing distress (Wilson et al, 2000; Farmer et al, 2005).

Strengths and Difficulties Questionnaire (SDQ)

When comparing pre-and post-SDQ scores, there were no statistically significant differences for any of the components assessed as part of the measure. However, decreases were noted for the subscales of conduct issues, hyperactivity issues and peer problems, coupled with a marked decrease in the total difficulties score (see Table 1 for a comparison of pre- and post-mean scores).

Relationship Problems Questionnaire (RPQ) and Intervention Care Questionnaire (ICQ)

No statistically significant difference between pre- and post-programme scores was observed for either the RPQ or the ICQ. However, a marked decrease in post scores was observed for both questionnaires, when compared to pre-programme scores (see Table 1 for a comparison of pre-and post-mean scores).

Despite the SDQ, RPQ and ICQ findings not reaching statistical significance, it is worth noting that all post-programme results, when compared to pre-programme results, had moved in a positive direction.

Qualitative evaluation: carer responses

As no qualitative evaluation took place before the programme, it is not possible to make pre/post-training comparisons; nonetheless, the responses obtained from the carers post training are still a rich source of information. The three main themes that emerged from the thematic analysis are 'carers developed an understanding of the issues for their foster children', 'personal growth and development, and self-reported reduced stress in foster carers', and 'course delivery and future attachment training'. These themes are discussed below, along with a description of each one and quotations from foster carers.

Understanding the issues for their foster children

Carers talked of the importance of the early years of life and understanding how attachments formed at an early age can affect the way a child views him or herself and others:

I hadn't really thought about attachment much before. But now I know how important all interactions are from the point the child is born and the effect that not experiencing safety, security and comfort can have on a child.

I'd never heard of the 'internal working model' before. So, to hear how this template that is formed is so fundamental to how the child views themselves and others was fascinating. It made so much sense and I was able to think about the internal working model of the child I care for, and consider how this links to how he struggles to approach certain situations or manages himself in situations, say with peers at school and with members of my family.

A number of carers reported that by having this increased understanding of attachment, they felt they now responded differently (and more appropriately) towards the child:

Understanding the importance of the early lessons the children have learned about themselves and primary care givers and how these patterns of interaction are the only way they know how to relate to people.

Learning about attachment was extremely helpful to me in terms of not taking things personally, such as how she reacts to me on occasion, and this means I am much more able to be calm and predictable in the way I respond to her.

Carers reported having developed a working knowledge of how the children struggle to understand and manage their emotions and how empathy can be a powerful way of communicating to the child that somebody understands them:

I now feel that I am putting into his head that I understand how he is feeling, subtly letting him know that it is not his fault, he is not to blame for what happened to him ... and I really believe he is sharing things with me much more than before, and talking to me more about how he is feeling and about how he felt about things that happened in his past.

It is important to note that, initially, empathy was introduced to the carers very much as another 'tool' they could use when they felt it was appropriate and there was a lot of discussion about when this should happen. What is extremely pleasing is that carers were reporting taking this a step further, to the point where they felt that being empathic was becoming a natural response for them:

I really think learning about empathy has been one of the most valuable things I learned about. I think I was doing it to a certain degree before, but thinking more closely about what it is and the impact empathy can have for these children has been enormously beneficial. I now feel after the training that being empathic has become a very natural way of responding for me.

They also reported that they looked at certain emotional issues differently now. In particular, a new-found consideration of the importance of 'shame' for these children seemed to have developed, and carers had a more insightful understanding of the importance of their own role in relation to how they respond to the child's behaviour so as not to induce and/ or exacerbate their feelings of shame:

I used to think about the kids experiencing anger and sadness a lot, but not so much shame. So when we started to talk about shame in the group, it really wasn't something I'd considered a great deal. Also, I felt that my idea of shame was different to how the facilitators were using it. But we discussed this in a lot more detail and I began to understand it better.

I now think about how the children have come to experience so much shame from their earlier experiences, and also understand how the things you say and how you respond to the child can affect whether the child experiences shame. For example, sometimes I used to reach the point where I'd shout--this is making the shame worse. I'm so much calmer in my responses now.

It was acknowledged that some of the material was at times difficult to grasp. But carers felt comfortable enough to speak up and ask for extra help with some of the attachment issues they were struggling to understand, such as some of the theory presented in module 1:

First couple of sessions were intense. I didn't really understand some of the words and it felt a bit daunting being in the group. But when we did the re-caps every week, I thought it was put across really well and this seemed to be helpful, as revisiting what we had done seemed to put it clearer in my mind; I understood what it meant. As we went along the course, everything fell into place, particularly when we got to modules 2 and 3, as that was when we really started putting things into practice.

Personal growth and development, and self-reported reduced stress in foster carers

Carers reported that they understood that attempts to parent these children with 'normal' parenting skills are often not enough. They seemed to grasp that parenting looked after children requires adaptation from usual parenting strategies:

Before I was trying to parent with normal parenting, but this just wasn't working. Now I think less about 'bad behaviour' and more about 'choices' and how much the child can manage at any given time. I now take responsibility by eliminating choices if I feel this will be unmanageable for the child.

Many carers acknowledged the group setting as excellent. This provided a supportive network and helped carers understand there are some similarities in the difficulties these children have:

To hear the difficult experiences of other carers and know that you are not the only one helped me a lot.

Good to listen to stories from other carers who had their own problems and be able to hear their experiences.

Knowledge and understanding of the issues helped carers realise that behaviour is not personal against them. Many behaviours and interactions are ways the children have learned to cope in difficult times/traumatic situations. This understanding and support was acknowledged as being supportive in maintaining some of the placements:

I no longer think, 'What am I doing wrong?' all the time.

Understanding things from the child's point of view helps me keep a lot calmer now. I no longer get angry very quickly; instead I am able to keep calm and think about how to respond before I react.

Significantly, there was a strong feeling for some carers that placements would possibly have broken down had they not had this attachment training. In fact, several carers identified that previous placement breakdowns they had been involved in could have been prevented had they been aware of some of the issues, such as how much the children attempt to control situations:

If I hadn't had the training, I feel that I would have failed the little boy I have now because the placement would have broke down. His behaviour was putting such a strain on our family and I just couldn't see a way out apart from him leaving ... The training has changed my outlook. By understanding more about his controlling behaviour and different ways I can handle this, and by my showing empathy much more often, this has actually begun to change how the little boy expresses himself and behaves ... I really feel I am starting to build an attachment with him and he is managing to fit in much more in the family home.

The carers all reported that the understanding they now had of attachment issues and the support emanating from knowing that other carers had children displaying similar emotional and behavioural patterns seemed to reduce the stress they experienced, in particular in relationships between them and the children, but also in the personal stress of dealing with challenging children:

Hearing that other carers had experienced similar difficulties to me made me ease off myself a little. My stress levels seemed to come down in the sessions, knowing that people there understood what I was talking about and could offer ways to help.

Course delivery and future attachment training

Carers reported that the way in which the course was delivered made them feel comfortable to participate:

I can't sing the praises enough of this training. Without it, I think I would have handled problems with the boys a lot differently.

Was nice not to be lectured. Enjoyed the interaction of the course--the feeling of not being lectured but listened to.

I enjoyed the humour that was used at points throughout the course. Felt this was needed at times because we were dealing with such a serious subject.

Carers expressed strong views regarding the future of this attachment training. They felt it was crucial and should be offered to all foster carers. At the time, South Tyneside local authority was only in a position to offer the training to foster carers but carers strongly voiced their beliefs that adoptive parents should also have this training if children are to stand the best chance of remaining with an adoptive family:

I think this training should be offered to all people who foster or adopt children because it has done wonders for us. Ultimately the insight and understanding I have gained from this training, I think, has helped give the children I care for a better chance.

Discussion

When comparisons are made with the original Fostering Attachments training (Golding, 2006), it is important to acknowledge from the outset that there were unavoidable differences in the delivery of the present training. The original programme comprising 18 sessions was delivered over 18 months in two-hour sessions. The present training comprising 18 sessions was delivered over six months in three-hour sessions. Comparisons must therefore be tentative and must take into account possible variables, including differences in the trainers, carers, training environment, group dynamics and the mental health of the young people being looked after, all of which could not be evaluated as part of this 'small-scale' research.

As previously alluded to, continuous verbal feedback from carers was encouraged as an important source of information for facilitators, guiding and steering the shape of preceding training sessions and alerting facilitators to any issues arising as the training programme progressed. One of the findings from this aspect of the evaluative process was that some carers found aspects of the content of the course hard to grasp. In such circumstances, the process of this continuous evaluation via verbal feedback was invaluable, enabling us to 'pause' and 'rewind' at times throughout the training, ensuring that key concepts were grasped before moving on. Some variability in carers' learning is inevitable. However, it can be said with confidence that the majority regularly reported benefiting from the course, suggesting that Golding's framework for 'fostering attachments' is robust and well worthy of consideration by fostering training services. Carers regularly talked of the benefits of group-work support, learning and sharing the struggle with difficult concepts.

Impact of training on relationship with spouse/partner

A further and most striking finding from this 'verbal feedback' was the reported effect on the relationship between some of the foster carers and their spouse/partner, an issue not included on the training. We were alerted to the difficulties for the non-training spouse/partner in understanding the change in approach to the young person. To our knowledge, the earlier evaluation (Golding and Picken, 2004) did not identify this issue nor has other research which has since evaluated training with an attachment focus (Allen and Vostanis, 2005), and therefore this outcome was somewhat unexpected. It was beyond the scope of the present appraisal to evaluate this issue (although it is most certainly advocated that it be explored) and we can only speculate as to why this might have been the case. Perhaps this issue came about because of the frequency and intensity of training sessions, ie the present training was completed on a weekly basis over six months, whereas Golding and Picken (2004) conducted their training over a longer time period. The potentially much faster pace of change in the foster carer's approach to parenting in the present training may have been more difficult for spouse/partners to adjust to. Whatever the potential reasons, it was felt that the tension that appeared to have been created between several couples needed to be addressed. Subsequently, an attenuated six-session training package was run for spouses and partners only, to help them with the transition. No formal measurements were taken but informal feedback suggested that this additional training was helpful to the couples concerned. This is clearly another area of potential future research.

Further themes and initiatives

Golding and Picken (2004) found a statistically significant difference both in reduced stress in carers and in the improved behaviour of the young people. The present research also identified a statistically significant reduction in the stress reported by foster carers. We believe the carers were much less likely to blame themselves for the poor behaviour of the young person in their care. This finding lends weight to the assertion that aspects of the present training had a positive impact on the stress experienced by foster carers. Despite finding no statistically significant improvement in the behaviour of the young people, most measures were pointing in the right direction. The difference in the length of the course may be an important factor here. For instance, it may be that the behaviour of children with attachment issues takes longer to change than that of carers' stress and confidence. Or perhaps the carer's increased understanding of attachment issues creates a more informed appraisal of the issues presented by the child, thus influencing carer responses on the objective measures. We intend to consider this idea further by repeating the measurements at a later date to allow an 18-month comparison with the Golding training (2006).

Golding and Picken (2004) identified numerous themes from their qualitative analysis related to carers' views of the training. They report that carers found the group supportive and the content of the training informative, although the authors acknowledge that as the facilitators of the training also conducted evaluation interviews, there is a potential for bias. In a bid to overcome this issue, the present evaluation used a trainee clinical psychologist, not connected to the training in any way, to undertake the interviews. With this control in place, it is encouraging that the present training's qualitative evaluation findings identified themes similar to that of Golding and Picken (2004)--'carers developed an understanding of the issues for their foster children', 'personal growth and development, and self-reported reduced stress in foster carers' and 'course delivery and future attachment training'.

It was concluded that in addition to objective measures used, this qualitative aspect of the evaluation process was extremely informative, providing rich and detailed accounts of carers' views, knowledge and experiences of the attachment training. It is recommended that future attachment training continues to consider such research methods as a way of gaining important feedback from participants.

A potential shortcoming of the present training was the absence of fostering support officers (with the exception of one officer who co-facilitated the training). There was no scope to explore whether carers' increased knowledge and adapted parenting approach made the fostering support officers' regular sessions with foster carers more challenging. We acknowledge that offering no attachment training to fostering support officers, who are a potential source of additional support for the carers involved and who are ideally placed to reinforce some of the important concepts, was an opportunity lost. South Tyneside fostering services and CPLAC are currently repeating this training with a second cohort of foster carers but are involving the fostering support officers to a greater degree.

Our services are also in the process of considering an initiative called ATTACH, the aim of which is to broaden the understanding and support available to the foster carers looking after young people with significant attachment difficulties. The ATTACH initiative includes recruiting 'champions' from all the corporate parenting agencies. These 'champions' will receive additional training and support via regular group meetings. Their goal will be to 'champion' the understanding of the importance of attachment within their own corporate agency or staff group. It is hoped that such an initiative will provide an attachment-focused 'wrap-around' service to the young people and their foster carers.

The need for new research and development

As well as the service development issues alluded to, it is important to consider the expansion of research in this area. This is especially significant as previous parenting and behaviour modification training programmes, underpinned by social learning theory and primarily developed for use with birth parents of children displaying challenging behaviour (Webster-Stratton, 1997), have not been shown to be particularly beneficial for foster carers and looked after children (Allen and Vostanis, 2005) and attachment difficulties are now understood to be at the root of many of the difficulties presented by looked after children (Hughes, 1997; Howe and Fearnley, 1999). As a result, there is a lack of substantive research that identifies exactly what training with carers may be required to have optimum benefits for the attachment needs of such children. The present evaluation, although small, is one contribution to fill this void. Research with larger group sizes, or combining evaluation data from multiple groups against other foster carer training programmes and against controls (ie carers who have not received the training), is proposed as a plausible way forward to identify the potential for such attachment training programmes and what they can achieve in comparison to others, such as Fostering Changes, currently available.

Research should also attempt to develop the evaluation methods used and explore further ways of considering the impact of attachment training. For instance, there are limited objective measures currently available that specifically focus on attachment issues in looked after children (Golding, 2006). It may be that behavioural changes within the children do take place due to the training of carers, but might be missed owing to the lack of available evaluation material. Or, given that it takes a long time for children with attachment difficulties to recover, perhaps we need to be considering to a much greater degree longer-term follow-up evaluation.

Development of such materials and longer evaluation periods would go some way to help identify whether carer stress reduction alone is fundamental to attachment training and is what we should be targeting, or whether we should perhaps consider developing training that attempts to create a more substantial change in child behaviour. Perhaps attachment training tailored for carers and their foster children together might be beneficial, although the age of children is clearly an issue here. Or perhaps, in order to promote a greater change in the behaviour of children, it is important that interventions in addition to carer training take place which directly target the carer--child relationship. Also, do we need to be looking at the neuro-developmental difficulty which is increasingly acknowledged as developing from early abuse and neglect (Gerhardt, 2004) and thinking about such issues in relation to what behavioural change may feasibly take place in such children?

We believe the present evaluation is small, yet important in its offerings within the field, providing further insights into the original fostering attachments training (Golding, 2006) and creating ideas that may be developed. According to carers, the attachment training has undoubtedly been highly beneficial to them and the care they offer to looked after children. However, the challenges of developing such training and conducting research which aids us in our quest to explore what more we can offer with regards to attachment training is clearly something that merits much greater attention within the field.

Conclusion

Whilst strict comparisons between the preliminary findings of previous training and evaluation (Golding and Picken, 2004) and the present article's training and evaluation are difficult, South Tyneside fostering services and CPLAC are convinced, from the outcomes obtained, that the attachment training devised by Golding (2006) is of enormous benefit to foster carers. With this in mind, our services are taking steps to ensure that it becomes an established and integral part of that offered to foster carers in South Tyneside. We would recommend that other services consider doing the same.

Acknowledgements

With thanks to Dr Stephanie Richardson, clinical psychologist, for co-facilitating the group programme and offering support and advice throughout, and to Brian Morgan, Fostering Services Manager, South Tyneside Local Authority for his unfailing support and commitment. Further thanks to Dr Nicola Richardson, clinical psychologist, for conducting post-evaluation interviews with foster carers. We are indebted to the foster carers who willingly gave their time to participate not only in the 18-session programme, but also in the evaluation.

References

Abidin RR, Parenting Stress Index: Professional Manual (third edition), Odessa, FL: Psychological Assessment Resources, 1995

Allen J and Vostanis P, 'The impact of abuse and trauma on the developing child: an evaluation of a training programme for foster carers and supervising social workers', Adoption & Fostering 29:3, pp 68-81, 2005

Crittenden PM, Landini A and Claussen AH, 'A dynamic-maturational approach to treatment of maltreated children', in Hughes JN, La Greca AM and Conoley JC (eds), Handbook of Psychological Services for Children and Adolescents, Oxford: Oxford University Press, 2001

Delaney RJ, Fostering Changes: Treating attachment-disordered foster children, Oklahoma City: Wood & Barnes Publishing, 1998

Dimigen G, Del Priore C, Butler S, Evans S, Ferguson L and Swan M, 'Psychiatric disorder among children at time of entering local authority care: questionnaire survey', British Medical Journal 319, p 675, 1999

Fahlberg V, A Child's Journey Through Placement, London: BAAF, 1996

Farmer E, Lipscombe J and Moyers S, 'Foster carer strain and its impact on parenting and placement outcomes for adolescents', British Journal of Social Work 32:2, pp 237-53, 2005

Gerhardt S, Why Love Matters: How affection shapes a baby's brain, Hove: Routledge, 2004

Golding K, Fostering Attachment in Children who are Looked After and Adopted, Unpublished, 2006

Golding K and Picken J, 'Group work for foster carers caring for children with complex problems', Adoption & Fostering 28:1, pp 25-37, 2004

Goodman R, 'The Strengths & Difficulties Questionnaire: a research note', Journal of Child Psychology & Psychiatry 38:5, pp 581-86, 1997

Hanson RF and Spratt EG, 'Reactive Attachment Disorder: what we know about the disorder and the implications for treatment', Child Maltreatment 5:2, pp 137-45, 2000

Hill-Tout J, Pithouse A and Lowe K, 'Training foster carers in a preventative approach to children who challenge: mixed messages from research', Adoption & Fostering 27:1, pp 47-56, 2003

Howe D, Brandon M, Hinings D and Schofield G, Attachment Theory, Child Maltreatment and Family Support, Basingstoke: Macmillan, 1999

Howe D and Fearnley S, 'Disorders of attachment and attachment therapy', Adoption & Hostering 23:2, pp 19-30, 1999

Hughes DA, Facilitating Developmental Attachment: The road to emotional recovery and behavioural change in foster and adopted children, London: Aronson, 1997

McCann J, James A, Wilson S and Dunn G, 'Prevalence of psychiatric disorders in young people in the care system', British Medical Journal 313, pp 1529-30, 1996

Minnis H and Devine C, 'The effect of foster carer training on the emotional and behavioural functioning of looked after children', Adoption & Fostering 25:1, pp 44-54, 2001

Minnis H, Devine C and Pelosi T, 'Foster carers speak about training', Adoption & Fostering 23:2, pp 42-47, 1999

Pallett C, Scott S, Blackeby K, Yule W and Weissman R, 'Fostering changes: a cognitive-behavioural approach to help foster carers manage children', Adoption & Fostering 26:1, pp 39-48, 2002

Quinton D, Rushton A, Dance C and Mayes D, Joining New Families: A study of adoption and fostering in middle childhood, Chichester: John Wiley & Sons, 1998

Sinclair I, Gibbs I and Wilson K, Supporting Foster Placements: Reports 1 and 2, York: Social Work Research and Development Unit, 2000

Webster-Stratton, C, 'From parent training to community building: families in society', The Journal of Contemporary Human Services 78, pp 156-71, 1997

Wilson K, Sinclair I and Gibbs I, 'The trouble with foster care: the impact of stressful events on foster carers', British Journal of Social Work 30, pp 193-209, 2000

[c] Gemma Laybourne, Jill Andersen and John Sands 2008

Gemma Laybourne is Assistant Clinical Psychologist, Clinical Psychology for Looked After Children, North-umberland, Tyne & Wear NHS Trust

Jill Andersen is Fostering Support Officer, South Tyneside Local Authority

John Sands is Consultant Clinical Psychologist, Clinical Psychology for Looked After Children, North-umberland, Tyne & Wear NHS Trust

Table 1
Comparison of mean pre- and post-programme carer ratings

 Significance
N = 7 Pre Post of change Effect size
 Mean (Std. Mean (Std. (p value) (Eta squared)
 Deviation) Deviation) two-tailed
PSI/SF
Total stress 111.6 (13) 87.9 (19.9) 0.02 0.37

Parental 33.9 (7.8) 23.9 (6) 0.02 0.37
distress
SDQ

Total 22.1 (7.1) 18.4 (5) 0.28 (NS) 0.1
difficulties
RPQ

Total score 21.1 (7.3) 17.6 (10.8) 0.48 (NS) 0.04
ICQ

Total score 48.4 (5.4) 45.9 (6.4) 0.43 (NS) 0.05