An evaluation of the impact of attachment and trauma training for pupil referral unit staff

Abstract

Background:

The Gwent Attachment Service trained four pupil referral units (PRUs) staff in attachment- and trauma-informed work. Education staff received 2 days of training and then attended six skills development sessions (SDSs) on a monthly basis to embed the training concepts into their work. This model takes a “whole systems approach” to intervention, drawing on evidence suggesting that having a supportive and consistent system around a child that takes into account their attachment needs leads to better outcomes.

Methods:

Self-reported knowledge of attachment- and trauma-informed work, confidence in carrying out this work, and worries about implementing this work were collected from 64 education staff members across the four PRUs. Measures were repeated at three time points: pre-training, post-training, and post-SDS.

Results:

Knowledge and confidence increased from pre- to post-training. Knowledge did not significantly differ between post-training and post-SDS. Confidence was lower at post-SDS than post-training but remained above pre-training ratings. Worries decreased from pre- to post-training and decreased again post-SDS.

Conclusion:

Training and SDSs can improve teaching staffs’ perception of their knowledge and confidence, while reducing worries about working in an attachment- and trauma-informed way. SDSs decreased worries about working in an attachment- and trauma-informed way over and above training alone, indicating that the SDSs contribute added value. This study presents a promising starting point for improving the lives of children and young people who have experienced trauma and have attachment difficulties.

Keywords

Attachment consultation education training trauma

Introduction

From 2016 to 2018, the Gwent Attachment Service trained four pupil referral units (PRUs) in attachment- and trauma-informed work. This article reports an evaluation of this program, which involved 2 days of training as well as six skills development sessions (SDSs) on a monthly basis to embed the skills into practice. The service is a health-based multiagency commissioned service with an initiative to upskill children’s services across Gwent, including social care, education teams, youth offending, and health services. This was in response to a local case review (Kirkaldie & McDonnell, 2015) and gaps analysis (Jones & Waters, 2016), indicating an unmet need for children and young people with experience of early adversity and trauma in the local area. The training uses attachment theory (Ainsworth & Bowlby, 1991; Bowlby, 1958) and developmental trauma literature (Bellis et al., 2016) to support education staff to enhance teacher–student relationships. Attachment- and trauma-informed work relates to being able to recognize the impact that children’s early life experiences can have on their physical and emotional development as well as exploring ways to respond to and manage children’s behavior while meeting their attachment needs.

The service model

The service ethos is to support better outcomes for children and young people who have experienced early adversity and trauma. The service has been developed to offer an indirect model of intervention, by training and upskilling professionals and frontline staff across Gwent, rather than offering work directly with children and families. This model has been developed to reflect the values of a prudent health-care system (Bradley et al., 2014), aiming to cover a large geographical area with a relatively small staff resource (5.6 whole-time equivalent staffing level). This model takes a “whole systems approach” to intervention. This approach is consistent with the growing perspective that children with attachment needs have better outcomes through cumulative, consistent attachment-informed interactions with caregivers and key adults in their lives (Casswell et al., 2014; Golding, 2007; Sabol & Pianta, 2012). Furthermore, evidence suggests that direct clinical work is most effective when the system around the child is able to support it (Golding, 2010). Lastly, the model is informed by an early intervention and prevention approach, which is in line with a “spend to save” model. Money spent on early intervention in one area of the system (e.g., education) is likely to reduce spending in another (e.g., youth justice) and lead to better outcomes for young people, their families, and wider society (Allen, 2011).

Why education and PRUs?

The South East Wales Consortium for Additional Learning needs requested training for the PRUs. PRUs are an alternative educational provision for children and young people who are unable to attend a special educational needs or mainstream school (Welsh Government, 2017). This can be due to a number of reasons including exclusion from mainstream school or health and safety reasons (Welsh Government, 2017). In semi-structured interviews, parents of children who had been at risk of or had been excluded from mainstream school felt that a number of underlying issues, including attachment difficulties, had impacted their child’s ability to cope in school (Parker et al., 2016). Behavioral problems were the most common reason for exclusion; however, some parents reported that the needs behind their children’s behaviors were not understood. In a recent survey, pupils attending PRUs in Wales identified their own behavior, poor relationships with teachers, being bullied, feeling unsafe, anxiety, and mental health difficulties as some of the main reasons why they felt unable to cope in a mainstream school (Children’s Commissioner for Wales, 2014). The survey also stated that pupils currently attending a PRU in Wales are among the “most vulnerable and disadvantaged learners” (Children’s Commissioner for Wales, 2014, p. 36), often coming from chaotic and complex home environments (Welsh Government, 2017).

Current literature supports training teachers in attachment-informed principles to ensure vulnerable learners develop a sense of emotional security and build appropriate relationships (Bombèr & Hughes, 2013). Previous research has also suggested that connections and secure attachments with school staff are vital in addressing the needs of children and young people with emotional and behavioral difficulties (Midgen et al., 2019). Pupils have identified that feeling safe and having positive relationships with teachers are factors that support their concentration and engagement in class (Midgen et al., 2019). In addition, positive teacher–student relationships are associated with greater teacher well-being and retention (Spilt et al., 2011).

Training model rationale

In a thematic analysis, Berrett (2019, p. 7) found that staff at the four PRUs in question felt that the “young people who attend the PRUs have become more complex in terms of historical trauma and mental health difficulties.” Staff reported that having a number of young people who have experienced adversity, together in one setting, was a difficult dynamic to manage. Therefore, by providing training, the Gwent Attachment Service hoped to alleviate some of these concerns for PRU staff.

Research has identified several barriers that inhibit effective practice with those who have experienced trauma. These included lack of training, lack of confidence and knowledge in evidence-based skills, and worries about causing offense or distress to the clients (Read et al., 2007; Walters et al., 2016). Walters et al. (2016) found that addressing these barriers successfully improved staff ratings of their ability to assess and work with trauma.

The training delivered to the PRUs covered five themes over the course of 2 days to address some of these barriers by aiming to improve knowledge and confidence in evidence-based skills while reducing worries when working with those who have experienced trauma. The training was in line with the National Institute for Health and Care Excellence (NICE, 2015) guidelines for working with children and young people with attachment difficulties.

The training package was developed and delivered by two consultant clinical psychologists who were supported by a team that included clinical psychologists, educational psychologists, forensic psychologists, trainee clinical psychologists, assistant psychologists, and placement students.

Training day 1

What is attachment?

An overview of the history and development of attachment theory, explaining main ideas and basic principles (addressing knowledge).

Learning objectives—To understand:

Attachment theory as it applies to “normal” healthy relationships (Ainsworth & Bowlby, 1991; Bowlby, 1958)

The concept of an Internal Working Model: How early attachment relationships can shape a person’s views about the self, others, and the world (Bowlby, 1973; Thompson, 2008)

How behaviors and beliefs within children adapt when care has not been “good enough,” according to attachment theory (Bombèr, 2007; Bombèr, 2011; Geddes, 2006)

That attachment is a “relational” concept and therefore professionals must consider the attachment strategies we ourselves bring to interactions (Marmarosh et al., 2006; McFarlane et al., 2010; & Mikulincer et al., 2013)

The impact of developmental trauma and disrupted attachment (attachment and brain development)

Psycho-education and current research describing the science behind how trauma affects the developing body and brain (addressing knowledge).

Learning objectives—To understand:

The functions of the brain stem, limbic system, and the neo-cortex (Bryson & Siegel, 2012)

Critical and sensitive periods in brain development (Uylings, 2006)

That lack of physical and/or emotional safety leads to fight, flight, or freeze (Van der Kolk, 2014)

The impact of stress on the brain (McCrory et al., 2011)

That we are primed to attach and this fundamentally influences brain development and how we organize our experiences (Siegel, 2001)

That our experiences can “reorganize” neural connections for better or worse due to “brain plasticity” (Bryson, & Siegel, 2012)

Applying attachment theory to understanding children and their behavior (addressing knowledge and confidence).

Learning objectives—To understand:

That children and young people may perceive a threat in situations where objectively there is none (Bombèr, 2007; Bombèr, 2011)

How the fight, freeze, flee response may account for some behaviors in the classroom (Bombèr, 2007; Bombèr, 2011)

How young people cope with shame (Bombèr, 2007; Bombèr, 2011)

How to use the “Pyramid of Need” (Golding & Hughes, 2012, p. 213) to identify and meet children’s therapeutic needs

Training day 2

Promoting change. Attachment informed skills and approaches

Using current attachment-informed interventions and Dyadic Developmental Practice (DDP) principles to support practice when working with children and young people who have experienced disrupted attachment and developmental trauma (addressing confidence).

“DDP is a model of therapy and care based on the theories and research of attachment, intersubjectivity, interpersonal neurobiology, and trauma” (Hughes, 2017, p. 597). It is primarily used to encourage connecting and attachment security in children and young people who have experienced trauma (Hughes et al., 2015). The principles of DDP have been used to inform other settings, such as schools (Bombèr & Hughes, 2013) and social care services (Hudson, 2006, chapter 8), to ensure support and safety for vulnerable young people and families.

Attachment-informed interventions and DDP principles that were covered in the training were:

Containment—To receive a child’s emotional communication, to provide support and help them understand and regulate their feelings (Bion, 1962)

Attunement and Intersubjectivity—Being aware of and responsive to a child. To share an emotional experience by matching feeling and expression. Joining a child in their experience and experiencing it with them (Hughes & Baylin, 2012)

Connection Before Correction—To help a child regulate their emotional experience before putting in a boundary or consequence (Golding, 2015)

PACE (Playfulness, Acceptance, Curiosity, and Empathy) —A way of communicating with children to help them feel safe, regulated, and understood (DDP Network, 2020)

Remembering you are human too

Supporting a dialog about self-care with teams.

Trauma-based literature often discusses the importance of staff self-care and well-being to manage burnout and support capacity to work in an attachment informed way (Hughes & Baylin, 2012). Teachers’ well-being is entwined with the well-being of their students (Roffey, 2012). Therefore, promoting the well-being of teachers will enable schools to better meet children’s varied and complex needs. This aspect of the training felt especially relevant as the education sector has recently been highlighted by the Health and Safety Executive (2018) as having higher than average rates of stress, depression, and anxiety than all other types of industry (addressing knowledge, confidence, and worries).

Learning objectives—To:

Acknowledge the practical and emotional challenges that educational staff face

Understand what teachers need themselves to provide a secure base for a child or young person who presents with challenging behavior

Understand the barriers within educational systems and organizations that could prevent attachment informed work

Ongoing support

Research confirms the importance of post-training activities to transfer training into practice (Martin, 2010), and staff members taking part in a similar trauma-based training program stated that they would have valued more time for skills development post-training (Walters et al., 2016). Furthermore, a number of practitioners have commented on the value of consultancy for working effectively with the complex networks of young people in care (Golding, 2003, 2010; Sprince, 2000), and the value of ongoing consultation and skills development with teachers for emotional and professional support (Gill & Monsen, 1995). Change tends to be a slow process within schools and can be difficult to achieve and maintain (Monsen & Cameron, 2002). As previous studies have highlighted that teachers feel undertrained to work appropriately with children with emotional and behavior difficulties (Avramidis et al., 2000; Goodman & Burton, 2010), it was felt that ongoing continuing involvement was needed for whole school change to take place.

Skills development sessions (SDSs)

Training was followed up with six SDSs, each 2 hr long, with the aim of providing ongoing attachment informed consultation to the PRUs. These sessions were designed to embed the theory and skills learned in training into practice.

The format of the SDSs was based on three different consultation models (Caplan, 1970; Farouk, 2004; Golding, 2004). Each held aspects that were suitable for our training rationale such as increasing the skill of the consultee (Caplan, 1970) with a focus on case formulation and attachment theory (Golding, 2004). Case formulations involve forming “a hypothesis or ‘best guess’ about the origins of a person’s difficulties in the context of their relationships, social circumstances, life events, and the sense that they have made of them” (Johnstone, 2018 p. 32). Case formulations were a key part of the SDSs and enabled the facilitator to think collaboratively with each PRU team about how to make sense of children’s difficulties and how best to support them.

Like any other group of individuals or organizations, a school has its own internalized norms, behaviors, and values that make up its culture (Gruenert, 2000). This may influence group dynamics and the way teachers can perceive and label behavior and emotions (Farouk, 2004). Farouk’s Group Consultation Model was particularly relevant, not only as it had previously been used in a school setting (Hayes & Stringer, 2016), but its emphasis on reflection, group dynamics, and systemic working could address various barriers to attachment- and trauma-informed work embedded within the school culture.

The psychologists facilitating the sessions tailored the content of the SDSs to meet the needs of each PRU based on observations within sessions, formal debriefs, and feedback from the staff teams. An example of how the SDS process went for one team is as follows:

SDS 1—Recap of the training concepts and an introduction to case formulation

SDS 2—A case formulation, applying the concepts from training to the case in-depth

SDS 3—A case formulation and practicing the skills from training (i.e., PACE) using role-play

SDS 4—A case formulation and using videos to illustrate and discuss relevant training concepts (i.e., attunement)

SDS 5—Reflecting on self-care in response to a stressful period

SDS 6—Reflect on the training and SDS process. Whole team self-care exercise

Delivering skills development alongside a case formulation within the SDSs felt like a good combination as information from one activity would feed into the other. For example, when discussing a case formulation, often PACE would be highlighted as a way of making a connection with the child or young person. Staff would then be given the opportunity to practice PACE in pairs or in small groups giving a more interactive and dynamic feel to the sessions.

Attachment champions

At the end of the SDSs, two members of staff from each PRU were invited to become “Attachment Champions” whereby they would attend quarterly Attachment Champion Forums facilitated by the Gwent Attachment Service, and disseminate any information and learning to their teams. This allowed the PRUs to maintain contact with the service on a regular basis, further their knowledge and understanding of attachment- and trauma-informed work, and keep the training alive in the minds of each PRU team.

Hypotheses

It was hypothesized that implementing the training program would significantly improve education staff’s knowledge and confidence, as well as significantly reduce their worries about working in an attachment- and trauma-informed way. Furthermore, it is hypothesized that the SDSs will result in a further increase in knowledge and confidence and a further decrease in worries. Particular interest will be paid to the role of the SDSs and the added benefit they bring to improving staff knowledge, confidence, and decreasing worries as they are an additional and time-intensive resource for both the service and the PRUs.

Method

Participants and procedures

Data were collected from 64 education and teaching staff members from 4 PRUs. Questionnaires were completed in the training room on the morning of the first training session (pre-training), at the end of the second day of training (post-training), and again at the end of the penultimate SDS session (post-SDS).

Ethical considerations

The service evaluation was approved by the Aneurin Bevan University Health Board Research and Development Department; the Research Risk Review Panel determined that the study did not pose any risk. Consent was obtained from team members for the purpose of report writing and research, and all data were anonymized using identification codes at the point of data entry.

Measures

The measures comprised a questionnaire assessing participants’ knowledge of trauma and attachment-informed work, their confidence in carrying out this work, and their worries about implementing this work. Participants were required to state how much they agreed with statements such as “If a child disclosed a traumatic experience I would feel confident in how to respond and proceed” on a 5-point Likert scale (see Appendix for full questionnaire). The knowledge subscale consisted of 10 items (α = .81), the confidence subscale consisted of 4 items (α = .74), and the worries subscale consisted of 6 items (α = .71). The questionnaire was based on Walters et al.’s (2016) questionnaire assessing changes in staff knowledge, confidence, and worries after completing a trauma-based training program.

Missing data

Missing data analyses indicated that between 3% and 44% of data were missing for each item at each time point. Little’s Missing Completely at Random test indicated that the data were missing completely at random, χ²(471) = 416.44, p = .97. To maximize statistical power, missing values were estimated using fully conditional specification multiple imputation (Fichman & Cummings, 2003). All study variables and a number of auxiliary variables were used to predict missing data, as recommended by Gottschall et al. (2012). Five imputed data sets were generated. One of the five data sets was then selected for analysis using a random number generator.

Data analysis

Descriptive analyses were carried out to explore staffs’ self-reported levels of knowledge, confidence, and worries. One-way repeated measures analyses of variance (ANOVAs) were conducted to determine whether there were statistically significant differences in knowledge, confidence, and worries about implementing this work over training and SDSs. The data were normally distributed for each scale, as assessed by the Shapiro–Wilk test (p > .05). Outliers were assessed using box-plots: the knowledge scale had four outliers pre-training and one outlier at post-SDS; the confidence scale had two outliers pre-training, one outlier post-training, and two outliers at post-SDS; and the worries scale had one outlier pre-training and one outlier post-training. All outliers were replaced with a less extreme value, while retaining the subject’s rank.

Results

Descriptive statistics

Percentage mean scores and standard deviations (SD) for staffs’ self-reported knowledge, confidence, and worries at three different time points (pre-training, post-training, and post-SDS) are presented in Figure 1.

ANOVAs

Knowledge

The assumption of sphericity was violated. Epsilon (ε) was .87, as calculated according to Greenhouse and Geisser (1959), and was used to correct the one-way repeated measures ANOVA. Knowledge was significantly different at the three time points, F(1.69, 106.52) = 25.16, p < .001, partial η² = .29, with knowledge increasing from pre-training (M = 33.53, SD = 5.97) to post-training (M = 41.01, SD = 8.58), then decreasing slightly post-SDS (M = 38.50, SD = 5.18) but remaining above baseline. Post hoc analysis with a Bonferroni adjustment revealed that knowledge significantly increased from pre- to post-training (95% confidence interval (CI) [−9.793, −5.169], p < .001) and from pre-training to post-SDS (95% CI [−7.35, −2.604], p = .001). There was no significant decrease between post-training and post-SDS scores (95% CI [−0.648, 5.66], p = .165).

Confidence

The assumption of sphericity was violated. Epsilon (ε) was .86, as calculated according to Greenhouse and Geisser (1959), and was used to correct the one-way repeated measures ANOVA. Confidence was statistically significantly different at the three time points, F(1.72, 108.14) = 46.05, p < .001, partial η² = .42, with confidence increasing from pre-training (M = 12.20, SD = 3.02) to post-training (M = 16.08, SD = 2.05), then decreasing at post-SDS (M = 14.46, SD = 2.79) but remaining above baseline. Post hoc analysis with a Bonferroni adjustment revealed that confidence significantly increased from pre- to post-training (95% CI [−4.70, −3.07], p < .001) and from pre-training to post-SDS (95% CI [−3.43, −1.09], p < .001), but significantly decreased from post-training to post-SDS (95% CI [0.638, 2.61], p < .001).

Worries

The assumption of sphericity, as assessed by Mauchly’s test, was met (p = .61). Worries were statistically significantly different at the three time points, F(2, 126) = 51.11, p < .001, partial η² = .45, with worries decreasing from pre-training (M = 15.98, SD = 4.04) to post-training (M = 14.19, SD = 3.97), and decreasing again to post-SDS (M = 10.31, SD = 3.99). Post hoc analysis with a Bonferroni adjustment revealed that worries statistically significantly decreased from pre- to post-training (95% CI [0.37, 3.21], p = .009), from pre-training to post-SDS (95% CI [4.19, 7.15], p < .001) and from post-training to post-SDS (95% CI [2.55, 5.209], p < .001).

Discussion

Summary

This article reports an evaluation of attachment- and trauma-informed work training delivered by the service to four PRUs, by measuring staff self-reported knowledge, confidence, and worries at three time points (pre-training, post-training, and post-SDS).

The results demonstrated an increase in staffs’ self-reported knowledge and confidence and a decrease in self-reported worries about working with trauma from pre-training to post-SDS. In line with the hypotheses, the results reflect an initial gain in knowledge and confidence from the two-day training supporting staff members to enhance their trauma- and attachment-informed understanding. Contrary to our hypotheses, ratings of knowledge and confidence decreased from post-training to post-SDSs; however, this decrease was not significant for knowledge. It is possible that teams rated their confidence lower at post-SDS because the sessions provided a space for a deeper discussion highlighting the complexity of the theory and approach. In the field of cognitive psychology, this is referred to as the “Dunning-Kruger effect” whereby there is an inability to accurately assess one’s own capability because of a lack of insight into what is needed to be fully competent (Dunning, 2011). However, post-SDS scores remained significantly higher than the pre-training scores, indicating that the program was successful in increasing self-reported knowledge and confidence overall. These findings are consistent with previous research, showing that providing training that addresses these barriers improves staff members’ self-reported ability to work in a trauma-informed way (Cavanagh et al., 2004) with effects lasting up to 6 months post-training (Walters et al., 2016).

Consistent with the hypotheses, worries showed a significant decrease from pre- to post-training and from post-training to post-SDS. In comparison, a similar training model that did not have follow-up sessions found no significant change in worries at a 6-month follow-up (Walters et al., 2016). This suggests that the SDSs offer a unique contribution that supports people to feel less worried about implementing attachment- and trauma-informed work over and above training alone. This reduces one of the key identified barriers to working in an attachment- and trauma-informed way (Read et al., 2007; Walters et al., 2016).

The findings are particularly promising, given the wider and within-system pressures teams faced during our involvement. These pressures included high staff turnover rate, high workload, budget cuts, staff depletion, long-term sickness, burnout, and school inspections. Research suggests that services working with traumatized individuals and/or are under gross amounts of pressure can in themselves become “traumatized” and develop hostility toward external support (Hormann & Vivian, 2005). Despite this, our findings provide evidence of a positive and optimistic starting point for ways of working with school teams under complex and difficult circumstances. This raises questions about which aspects of the training model made this possible, and how they may be replicated.

Critique and recommendations

As the SDSs were based on a reflective practice model of consultation with a focus on group dynamics (Farouk, 2004), the SDSs may have provided a containing space for teachers to discuss uncertainties. Providing support for teachers is considered to be a crucial part of school interventions (Farouk, 2004; Gill & Monsen, 1995), particularly as previous literature has highlighted that teachers can often feel isolated in their work (Cole, 1997). Therefore, having space to reflect with their colleagues and being actively listened to may be particularly beneficial for education professionals and should be considered in future training programs. Prior research has suggested that reducing the pressure to attend school consultations can increase teachers’ commitment to the consultation process and investment in its outcomes (Harris & Cancelli, 1991). Ensuring the training and SDSs were accessible to teaching staff was important for maintaining relationships. Therefore, attendance in our sessions was on a voluntary basis. However, it must be highlighted that SDS attendance was varied with some sessions having to be rescheduled due to low attendance.

Schools are heavily inspected (Ballet & Kelchtermans, 2009) which posed a risk of the psychology team being viewed as the “assessor” or “expert” (Hayes & Stringer, 2016). To maintain a trusting relationship, it was important to be open and sensitive with regard to collecting feedback from school staff. Therefore, it was communicated that the role of the SDSs was to explore new concepts and difficulties drawing on the varying and equal expertise of those in the group. Not recognizing the vital role of teacher contribution could jeopardize the willingness to attend the sessions (Harris & Cancelli, 1991). It was decided that self-report measures would be used to limit the level of scrutiny experienced by school staff, as these could be completed confidentially and relatively quickly. Despite this, there are many limitations to self-report measures, and there may be an inherent risk of bias due to staff members being invested in their school and wanting to report that they had made improvements. Apprehension bias, whereby participants aim to make a positive impression on the experimenter (Miller, 2013), could have resulted in overreporting of knowledge and confidence, and underreporting of worries at each time point. Demand characteristics may have influenced staff’s reports; however, if this were the case, it may be anticipated that staff would report that their scores would have improved from pre- to post-training and again from post-training to post-SDS. This was, however, not the case for confidence, which decreased from post-training to post-SDS. Future research should explore additional, more objective ways in which competence could be sensitively measured in school settings. For example, seeking the children’s opinion could offer a more indirect and valuable method of measuring change in a school setting (Midgen et al., 2019).

This study did not have a control group due to our limited capacity as a service and the need to prioritize delivering the intervention to the PRUs in a timely manner. As a result, however, the possibility that staffs’ knowledge, confidence, and worries improved due to factors external to our training and SDSs cannot be ruled out. Although this is a possibility, there were no known external factors that could have impacted the PRUs knowledge, confidence and worries during the intervention. Future research should compare the impact of the training to an appropriate control group, such as a “treatment as usual” group (no intervention) or an “information only” group (the training information provided in a booklet) to improve the validity of the findings.

Due to the level of complexity within the PRUs, it is possible that the downward trend in knowledge and confidence between the post-training measure and the post-SDS measure could continue. The Gwent Attachment Service and the PRU staff decided that longer term involvement may be necessary to address further barriers to attachment- and trauma-informed work within the school system and its wider context. The Gwent Attachment Service is currently in the process of piloting input of a day per week between two clinical psychologists for three of the PRUs. This is a yearlong pilot intervention that allows for continuing conversations with school staff, so support can be carefully tailored to suit the varying and changing needs of each PRU and their pupils. Examples of input are ongoing training, consultations with parents and teachers, and direct work with children and young people. Because this involves a lot of time and resources from both sectors, the pilot is being evaluated and any outcomes will determine how to move forward with the input.

Next steps

As the service ethos is to support better outcomes for children and young people who have experienced early adversity and trauma, what do these findings mean for the pupils of the PRUs? It is hard to measure the impact of an intervention for this complex population because their difficulties can manifest in a range of pervasive ways that are not easily measured or captured by evidence-based procedures (Cook et al., 2017). In addition, change may take time, making it harder to determine a specific cause and effect. The services’ next steps will be to work closely with each PRU to better understand how this way of working impacts the pupils. The NICE guidelines recommend outcome measures that capture educational performance and behavioral problems (NICE, 2015). Prior studies have measured changes in school exclusions (Timpson, 2019) as well as changes in perceived well-being, school belonging and engagement in class (Libbey, 2004; Roffey et al., 2019; Ryder et al., 2017).

Conclusion

Training and SDSs can improve teaching staff’s knowledge and confidence, while reducing worries about working in an attachment- and trauma-informed way within challenging contexts. This is a positive and optimistic starting point for improving the lives of children and young people who have had a difficult start to life. The service offers an ongoing alternative approach that takes into consideration the “system” around these children and young people. This is particularly important, given the complexity of this vulnerable population and how difficult it is for mainstream services to meet their needs (Jones & Waters, 2016; Kirkaldie & McDonnell, 2015).

Footnotes

Acknowledgements

We want to thank all members of the Gwent Attachment Service, past and present, particularly those who delivered the training and SDSs to the PRUs. Without your hard work and dedication, this article would not have been possible.

Declaration of conflicting interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: All authors are employed by the Gwent Attachment Service.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Rebecca Greenhalgh

Appendix

Extremely	Very	Neutral	Not Very	Not at all
□	□	□	□	□

References

Ainsworth

M. S.

Bowlby

(1991). An ethological approach to personality development. American Psychologist, 46(4), 333–341. https://doi.org/10.1037/0003-066X.46.4.333

Allen

(2011). Early intervention: The next steps, an independent report to her Majesty’s Government. Cabinet Office.

Avramidis

Bayliss

Burden

(2000). A survey into mainstream teachers’ attitudes towards the inclusion of children with special educational needs in the ordinary school in one local education authority. Educational Psychology, 20(2), 191–211. https://doi.org/10.1080/713663717

Ballet

Kelchtermans

(2009). Struggling with workload: Primary teachers’ experience of intensification. Teaching and Teacher Education, 25(8), 1150–1157. https://doi.org/10.1016/j.tate.2009.02.012

Bellis

M. A.

Ashton

Hughes

Ford

K. J.

Bishop

Paranjothy

(2016). Adverse childhood experiences and their impact on health-harming behaviours in the Welsh adult population. Public Health Wales NHS Trust. https://doi.org/ 10.1093/eurpub/ckw167.009

Berrett. (2019). What factors affected the process of and engagement with training and SDSs? Unpublished Manuscript.

Bion

W. R.

(1962). Learning from experience. Heinemann.

Bombèr

L. M.

(2007). Inside I’m hurting: Practical strategies for supporting children with attachment difficulties in schools. Worth.

Bombèr

L. M.

(2011). What about me? Inclusive strategies to support pupils with attachment difficulties make it through the school day. Worth.

10.

Bombèr

L. M.

Hughes

D. A.

(2013). Settling to learn: Settling troubled pupils to learn: Why relationships matter in school. Worth.

11.

Bowlby

(1958). The nature of the child’s tie to his mother. International Journal of Psycho-analysis, 39, 350–373.

12.

Bowlby

(1973). Attachment and loss. Vol. 2, Separation: Anxiety and anger. Hogarth Press.

13.

Bradley

Wilson

Buss

Harrhy

Laing

Shortland

van Woerden

(2014). Achieving prudent healthcare in NHS Wales. Public Health Wales. http://www.1000livesplus.wales.nhs.uk/sitesplus/documents/1011/Achieving%20prudent%20healthcare%20in%20NHS%20Wales%20paper%20Revised%20version%20%28FINAL%29.pdf

14.

Bryson

T. P.

Siegel

(2012). The whole-brain child: 12 proven strategies to nurture your child’s developing mind. Hachette UK.

15.

Caplan

(1970). The theory and practice of mental health consultation. Basic Books.

16.

Casswell

Golding

K. S.

Grant

Hudson

Tower

(2014). Dyadic developmental practice (DDP): A framework for therapeutic intervention and parenting. The Child and Family Clinical Psychology Review, 2, 19–27. https://doi.org/ 10.1177/0308575915610943

17.

Cavanagh

M. R.

Read

New

(2004). Sexual abuse inquiry and response: A New Zealand training programme. New Zealand Journal of Psychology, 33(3), 137–144.

18.

Children’s Commissioner for Wales. (2014). The right to learn: Supporting children and young people at pupil referral units to reach their potential. Children’s Commissioner for Wales. https://dera.ioe.ac.uk//20396/

19.

Cole

A. L.

(1997). Impediments to reflective practice: Toward a new agenda for research on teaching. Teachers and Teaching, 3(1), 7–27. https://doi.org/10.1080/1354060970030102

20.

Cook

Spinazzola

Ford

Lanktree

Blaustein

Cloitre

DeRosa

Hubbard

Kagan

Liautaud

Mallah

Olafson

Van Der Kolk

(2017). Complex trauma in children and adolescents. Psychiatric Annals, 35(5), 390–398. https://doi.org/10.3928/00485713-20050501-05

21.

DDP Network. (2020, January 28). What is meant by PACE? https://ddpnetwork.org/about-ddp/meant-pace/

22.

Dunning

(2011). The Dunning–Kruger effect: On being ignorant of one’s own ignorance. Advances in Experimental Social Psychology, 44, 247–296. https://doi.org/10.1016/B978-0-12-385522-0.00005-6

23.

Farouk

(2004). Group work in schools: A process consultation approach. Educational Psychology in Practice, 20(3), 207–220. https://doi.org/10.1080/0266736042000251790

24.

Fichman

Cummings

J. N.

(2003). Multiple imputation for missing data: Making the most of what you know. Organizational Research Methods, 6(3), 282–308. https://doi.org/10.1177/1094428103255532

25.

Geddes

(2006). Attachment in the classroom: The links between children’s early experience, emotional well-being and performance in school. Worth.

26.

Gill

Monsen

(1995). The staff sharing scheme: A school-based management system for working with challenging child behaviour. Educational and Child Psychology, 12(2), 71–80.

27.

Golding

(2003). Helping foster carers, helping children: Using attachment theory to guide practice. Adoption & Fostering, 27(2), 64–73. https://doi.org/10.1177/030857590302700209

28.

Golding

(2004). Providing specialist psychological support to foster carers: A consultation model. Child and Adolescent Mental Health, 9(2), 71–76. https://doi.org/10.1111/j.1475-3588.2004.00084

29.

Golding

(2007). Developing group-based parent training for foster and adoptive parents. Adoption & Fostering, 31(3), 39–48. https://doi.org/10.1177/030857590703100306

30.

Golding

(2010). Multi-agency and specialist working to meet the mental health needs of children in care and adopted. Clinical Child Psychology and Psychiatry, 15(4), 573–587. https://doi.org/10.1177/1359104510375933

31.

Golding

(2015). Connection before correction: Supporting parents to meet the challenges of parenting children who have been traumatised within their early parenting environments. Children Australia, 40(2), 152–159. https://doi.org/10.1017/cha.2015.9

32.

Golding

K. S.

Hughes

D. A.

(2012). Creating loving attachments. Parenting with PACE to nurture confidence and security in the troubled child. Jessica Kingsley.

33.

Goodman

R. L.

Burton

D. M.

(2010). The inclusion of students with BESD in mainstream schools: Teachers’ experiences of and recommendations for creating a successful inclusive environment. Emotional and Behavioural Difficulties, 15(3), 223–237. https://doi.org/10.1080/13632752.2010.497662

34.

Gottschall

A. C.

West

S. G.

Enders

C. K.

(2012). A comparison of item-level and scale-level multiple imputation for questionnaire batteries. Multivariate Behavioral Research, 47(1), 1–25. https://doi.org/10.1080/00273171.2012.640589

35.

Greenhouse

S. W.

Geisser

(1959). On methods in the analysis of profile data. Psychometrika, 24(2), 95–112. https://doi.org/10.1007/BF02289823

36.

Gruenert

(2000). Shaping a new school culture. Contemporary Education, 71(2), 14.

37.

Harris

A. M.

Cancelli

A. A.

(1991). Teachers as volunteer consultees: Enthusiastic, willing, or resistant participants? Journal of Educational and Psychological Consultation, 2(3), 217–238. https://doi.org/10.1207/s1532768xjepc0203_2

38.

Hayes

Stringer

(2016). Introducing Farouk’s process consultation group approach in Irish primary schools. Educational Psychology in Practice, 32(2), 145–162. https://doi.org/10.1080/02667363.2015.1129939

39.

Health and Safety Executive. (2018). Work-related stress, depression or anxiety statistics in Great Britain 2018. http://www.hse.gov.uk/statistics/causdis/stress.pdf

40.

Hormann

Vivian

(2005). Toward an understanding of traumatized organizations and how to intervene in them. Traumatology, 11(3), 159–169. https://doi.org/10.1177/153476560501100302

41.

Hudson

(2006). Being adopted: Psychological services for adopting families. In Golding

K. S

Dent

H. R.

Nissim

Stott

(Eds.), Thinking psychologically about children who are looked after and adopted: Space for reflection (pp. 222–254). Wiley.

42.

Hughes

(2017). Dyadic developmental psychotherapy (DDP): An attachment-focused family treatment for developmental Trauma. Australian and New Zealand Journal of Family Therapy, 38(4), 595–605. https://doi.org/10.1002/anzf.1273

43.

Hughes

D. A.

Baylin

(2012). Brain-based parenting: The neuroscience of caregiving for healthy attachment. W.W. Norton.

44.

Hughes

Golding

K. S.

Hudson

(2015). Dyadic developmental psychotherapy (DDP): The development of the theory, practice and research base. Adoption & Fostering, 39(4), 356–365. https://doi.org/10.1177/0308575915610943

45.

Johnstone

(2018). Psychological formulation as an alternative to psychiatric diagnosis. Journal of Humanistic Psychology, 58(1), 30–46. https://doi.org/10.1177/0022167817722230

46.

Jones

Waters

(2016). An Analysis of gaps in service provision in child and adolescent mental health services in the catchment area of Aneurin Bevan University health board; strengthening the connections. Aneurin Bevan Health Board.

47.

Kirkaldie

Mcdonnell

(2015). Complex case review for children and young people’s services. Unpublished manuscript.

48.

Libbey

(2004). Measuring student relationships to school: Attachment, bonding, connectedness, and engagement. Journal of School Health, 74(7), 274–283. https://doi.org/10.1111/j.1746-1561.2004.tb08284.x

49.

Marmarosh

C. L.

Franz

V. A.

Koloi

Majors

R. C.

Rahimi

A. M.

Ronquillo

J. G.

Somberg

R. J.

Swope

J. S.

Zimmer

(2006). Therapists’ group attachments and their expectations of patients’ attitudes about group therapy. International Journal of Group Psychotherapy, 56(3), 325–338. https://doi.org/10.1521/ijgp.2006.56.3.325

50.

Martin

H. J.

(2010). Improving training impact through effective follow-up: Techniques and their application. Journal of Management Development, 29(6), 520–534. https://doi.org/10.1108/02621711011046495

51.

McCrory

De Brito

S. A.

Viding

(2011). The impact of childhood maltreatment: A review of neurobiological and genetic factors. Frontiers in Psychiatry, 2, 48.

52.

McFarlane

Burrell

Fuddy

Tandon

Derauf

D. C.

Leaf

Duggan

(2010). Association of home visitors’ and mothers’ attachment style with family engagement. Journal of Community Psychology, 38(5), 541–556. https://doi.org/10.1002/jcop.20380

53.

Midgen

Theodoratou

Newbury

Leonard

(2019). ‘School for everyone’: An exploration of children and young people’s perceptions of belonging. Educational & Child Psychology, 36(2), 2–9.

54.

Mikulincer

Shaver

P. R.

Berant

(2013). An Attachment perspective on therapeutic processes and outcomes. Journal of Personality, 81(6), 606–616. https://doi.org/10.1111/j.1467-6494.2012.00806.x

55.

Miller

S. A.

(2013). Developmental research methods (4th ed.). SAGE.

56.

Monsen

Cameron

R. S.

(2002). Supporting and developing teachers’ management of difficult and challenging pupils through coaching. In Gray

(Ed.), Working with emotions: Responding to the challenge of difficult pupil behaviour in schools (pp. 151–168). Routledge/Falmer.

57.

National Institute for Health and Care Excellence. (2015). Children’s attachment: Attachment in children and young people who are adopted from care, in care or at high risk of going into care (NICE Guideline 26). https://www.nice.org.uk/guidance/ng26/resources/childrens-attachment-attachment-in-children-and-young-people-who-are-adopted-from-care-in-care-or-at-high-risk-of-going-into-care-pdf-1837335256261

58.

Parker

Paget

Ford

Gwernan-Jones

(2016). ‘.he was excluded for the kind of behaviour that we thought he needed support with…’ A qualitative analysis of the experiences and perspectives of parents whose children have been excluded from school. Emotional and Behavioural Difficulties, 21(1), 133–151. https://doi.org/10.1080/13632752.2015.1120070

59.

Read

Hammersley

Rudegeair

(2007). Why, when and how to ask about childhood abuse. Advances in Psychiatric Treatment, 13(2), 101–110. https://doi.org/10.1192/apt.bp.106.002840

60.

Roffey

(2012). Pupil wellbeing – teacher wellbeing: Two sides of the same coin? Educational and Child Psychology, 29(4), 8–17.

61.

Roffey

Boyle

Allen

(2019). School belonging—Why are our students longing to belong to school? Educational & Child Psychology. Special Issue: School Belonging, 36(2), 6–8.

62.

Ryder

Edwards

Clements

(2017). Measuring the wellbeing of children in care: Views from the frontline and opportunities for change. National Children’s Bureau. https://www.ncb.org.uk/sites/default/files/field/attachment/Measuring%20the%20wellbeing%20of%20children%20in%20care_Final%20report.pdf

63.

Sabol

T. J.

Pianta

R. C.

(2012). Recent trends in research on teacher–child relationships. Attachment & Human Development, 14(3), 213–231. https://doi.org/10.1080/14616734.2012.672262

64.

Siegel

D. J.

(2001). Toward an interpersonal neurobiology of the developing mind: Attachment relationships, “mindsight,” and neural integration. Infant Mental Health Journal: Official Publication of the World Association for Infant Mental Health, 22(1–2), 67–94. https://doi.org/10.1002/1097-0355(200101/04)22:1<67::AID-IMHJ3>3.0.CO;2-G

65.

Spilt

J. L.

Koomen

H. M.

Thijs

J. T.

(2011). Teacher wellbeing: The Importance of teacher–student relationships. Educational Psychology Review, 23(4), 457–477. https://doi.org/10.1007/s10648-011-9170-y

66.

Sprince

(2000). Towards an integrated network. Journal of Child Psychotherapy, 26(3), 413–431. https://doi.org/10.1080/00754170010003606

67.

Thompson

R. A.

(2008). Attachment-related mental representations: Introduction to the special issue. Attachment & Human Development, 10(4), 347–358. https://doi.org/10.1080/14616730802461334

68.

Timpson

(2019). Timpson review of school exclusions. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/807862/Timpson_review.pdf

69.

Uylings

H. B.

(2006). Development of the human cortex and the concept of “critical” or “sensitive” periods. Language Learning, 56, 59–90. https://doi.org/10.1111/j.1467-9922.2006.00355.x

70.

Van der Kolk

(2014). The body keeps the score: Mind, brain and body in the transformation of trauma . Penguin.

71.

Walters

Hogg

Gillmore

(2016). Evaluation of a tailored training programme to improve the assessment and treatment of trauma in an early intervention in psychosis (EIP) service. Psychosis, 8(3), 226–237. https://doi.org/10.1080/17522439.2015.1131324

72.

Welsh Government. (2017). Education otherwise than at school (EOTAS): Framework for action. https://gov.wales/sites/default/files/publications/2018-03/education-otherwise-than-at-school-framework-for-action.pdf