Overview
Implementation science suggests that barriers to promoting evidence-based practice occur at multiple levels, spanning factors at the individual, organisational, intervention and systems levels.51 As the proposed feasibility trial was designed to work with children involved in the social care system, we expected the process of recruitment and engagement to be more complex than that for a trial based in the health-care environment, as more professionals tend to be involved in the former, there are complex regulatory constraints, resources are tight and this setting is less used to being involved in research.35 Recruiting children in foster care and their foster carers was expected to be significantly dependent on the support of their social workers, who act as gatekeepers in these children’s care.52 Social workers tend to view their role as to protect the vulnerable children in their care; therefore, a social worker’s beliefs about a child’s vulnerability, the potential benefits of the study and the timing (i.e. engaging in research after a recent placement transition) may affect their willingness to refer families to a study.53
The process of consent in the context of children who are looked after also poses numerous challenges, as it can be particularly difficult to determine from whom consent must be obtained and how this should be done.35 Beyond this, there may be a reluctance among social workers to consent to engage in research as many social workers do not identify their profession as a scientific discipline; therefore, they may be unaware of the methodological processes involved in a RCT.54
Dixon et al.’s53 RCT of Multi-dimensional Treatment Foster Care for Adolescents (MTFC-A) highlighted that social workers’ understandings of randomisation affected recruitment, with many social workers adopting their own clinical judgement when selecting children to enrol in the study, rather than relying on the formal eligibility criteria. Later research by Mezey et al.,55 involving a peer mentoring intervention to reduce the incidence of teenage pregnancy among children and young people who are looked after in the UK, also highlighted misconceptions among stakeholders about recruitment criteria and the purpose of randomisation as key barriers to implementation. Although there was often a theoretical acceptance of the need for randomisation, in practice this rarely occurred.56 Similarly, Dale and Watson52 noted that, if key stakeholders do not fully believe that an intervention is of tangible benefit, their engagement and acceptance of the research project can be compromised. Although these studies highlight interesting dilemmas for conducting a RCT with looked-after children, their populations largely focused on adolescents. There has been little research to date on the potential barriers to and their impact on implementing a RCT with younger children in foster care.
We expected that the climate of financial constraints and changes in organisational policy and structure within local authorities may affect our study. As many others have found in their research involving children’s services, social workers often have very large case loads, which affects their ability to engage in research tasks.57 Efforts to engage children’s social care in outcome research have encountered challenges associated with competing demands, staff turnover and financial strain.51 Other studies involving local authorities have found that the reorganisation of services and a general lack of research infrastructure and support affected services’ abilities to engage in research studies.52,55
Many of the studies that have documented challenges of implementing evidence-based programmes in a local authority context also discuss potential facilitators of the process. These have included access and support from the development team;51 face-to-face meetings and having a research ‘champion’ within a team;52 continual dialogue to address divergent perspectives; 58 social workers acknowledging the importance of strong links between research and practice;59 and contractual agreements, incentives and greater acknowledgement of ethical dilemmas.53
Studies that attempt to implement evidence-based programmes often focus on knowledge development, with little information about strategies to prepare people and organisations for this new knowledge.60 However, without this, it is likely that progress will be slow at best. The Medical Research Council has emphasised the importance of including stakeholders in complex intervention planning,61 as it allows the research team to gain insights into the processes that may contribute to positive or negative study outcomes.
Aims of the current study
This chapter presents the results of a scoping study using qualitative methodology to understand the impact, acceptability and feasibility of implementing a trial of the Video-feedback Intervention to promote Positive Parenting and Sensitive Discipline, Foster Care (VIPP-FC) in a CAMHS and local authority context. To understand the facilitators of and barriers to the implementation of the VIPP-FC, the current study adopts a qualitative approach to exploring stakeholders’ experiences of and views on the general and specific research challenges posed by a clinical trial of an attachment intervention for young children in foster care. The primary aims were to further our understanding of the intervention–setting fit, the perceived usefulness and effectiveness of the intervention for this population, the facilitators of and barriers to implementation specific to local authorities’ children services and to identify stakeholders’ main concerns and key areas for future study development.62 The qualitative data gathered from stakeholders can then be utilised to establish optimal systems and procedures for a future trial of the VIPP-FC.
This study ran in parallel with and after the case series, so that the participating stakeholders had acquired experience of the intervention (including training), the recruitment process and the challenges involved in preparing the partner sites to take part in the RCT. The majority of stakeholders involved in this initial phase (qualitative work) were selected from the research site participating in the case series project, as they had already experienced all phases of the research process, from initial engagement, planning, recruitment, obtaining consent and screening to treatment delivery. To maximise learning from this process and to gain the perspectives of other sites not involved in the case series, we also interviewed a small number of stakeholders during the early set-up phases of the feasibility RCT, whose input is also incorporated into this report.
Methods
Participants
Stakeholders involved in the early phases of the organisation and implementation of the VIPP-FC were invited to take part in the study. In total, 26 stakeholders were contacted. Of the 26 invited to interview, 15 responded and 13 agreed to be interviewed. One dropped out because of poor health, another had left their post and one had yet to begin work on the project locally as a result of delays in site set-up. Consequently, 10 stakeholders were interviewed.
Participants comprised six children’s social workers, three of whom were either service managers or deputy managers; two clinical psychologists, both of whom were service leads; one local authority commissioner; and one CAMHS practitioner. One participant worked in a CAMHS service, four worked in specialist adoption and fostering CAMHS teams, and the remaining participants were employed by local authorities. In terms of their roles in the VIPP-FC study, three participants held a strategic role (i.e. involvement in study design), five participants were involved in service delivery (i.e. offering the VIPP-FC in their service or being a trained VIPP-FC practitioner) and two participants were involved in both aspects. Two male participants and eight female participants were interviewed, with clinical experience ranging from approximately 3–20 years.
Procedure
To gather information on the facilitators and barriers that may affect the acceptability and feasibility of the intervention, stakeholders who were involved in the set-up and delivery of the VIPP-FC study in local authorities and CAMHS were invited to be interviewed. Participants who expressed initial interest were then contacted by a researcher and sent information about the study, as well as consent forms. Two of the interviews were carried out by telephone because of a lack of availability for face-to-face meetings.
Interviews
A semistructured interview schedule was developed based on Smith’s63 established guidelines. As the research involved stakeholders from various roles and professional backgrounds, four different interview schedules were drafted to suit these different positions held by the stakeholders in the study. The interview schedules addressed four general themes: (1) engagement of participants in the study, (2) supervision and training, (3) obstacles to and challenges of implementation and (4) acceptability of the VIPP-FC. The interview schedules were applied flexibly and the order in which these topics were explored was dependent on the material the participants brought. This helped to ensure that the interview flowed naturally and that spontaneous information could emerge.64 Throughout the interview, participants were encouraged to elaborate on their answers by giving specific examples. The interviews lasted 50–70 minutes and were audio-recorded. Participants were invited to add additional comments or discuss topics that the interview schedule may not have captured at the end of the interview.
Qualitative analysis
Thematic analysis was applied to the interview transcriptions to understand the key patterns arising from the data. This approach was deemed suitable for the research objectives as it allowed the researcher to capture both similarities and differences among participants’ perspectives.64 The process followed the general protocol outlined by Braun and Clarke;65 there were five phases of the analysis. NVivo 12 (QSR International, Warrington, UK) was used for data management. In the first phase, the qualitative researcher began by reading through the transcripts and familiarising themselves with their content. Once transcripts had been read through in their entirety, the researcher read the transcripts in a more active manner, taking notes of ideas, phrases or topics of interest. Phase 2 involved the initial generation of thematic codes. This involved the researcher working systematically through the transcripts, semantically identifying different features of the data.65 The researcher also took note of data that did not fit the more common narratives during this process. Following this, in phase three, the researcher analysed these codes to consider how they could be combined to form broader themes and how they might be organised hierarchically. In phase four, each of the extracts that were coded for a theme were reviewed and examined to identify an overarching, coherent thematic framework. Themes were refined and divided into subthemes or merged together to form a larger theme. Once completed, these themes were then placed in the context of the entire data set to examine whether or not they reflect the meanings that were evident in the data.65 The fifth phase involved the researcher trying to capture the essence of each theme and its application to the original narratives. This involved a final examination of each transcript to allow for theme refinement and ensuring that all nuances, contradictions and exceptions had been captured.
Credibility checks
Although thematic analysis includes some intrinsic quality-verifying procedures (e.g. checking of themes against coded data, the reviewing of themes against the entire data set), the qualitative researcher also undertook several credibility checks following good-practice guidelines developed by Barker and Pistrang,66 as well as those set forth by Elliot et al.,67 to enhance the quality and validity of their analysis and subsequent conclusions.65
In the initial stages of analysis, the researcher involved a second qualitative researcher and provided them with two transcripts to code. They then checked their coding of the two interviews against those coded by the second researcher to identify overstatements, discrepancies and potential errors.67 Following this, a consensus approach was adopted to the development of the thematic framework. The researcher also involved a third member of the research team to review a selection of the data, compare ideas and discuss the best ways to represent the data.
Results
The analysis generated three overarching themes, which comprised eight subthemes. The first overarching theme, ‘challenges of implementation’, addressed the difficulties and barriers perceived by respondents in setting up and delivering the study. The second main theme addressed ‘benefits and impact of the VIPP-FC’, which captures respondents’ feelings about the potential value for foster children, foster carers and the professional network in engaging in the study. The final overarching theme focused on ‘key initial learning’ and reflected stakeholders’ ideas about how to overcome some of the barriers that might arise during implementation, or barriers that had already arisen during the case series work or early set-up phase of the feasibility RCT.
Overarching theme 1: challenges of implementation
Box 1 presents the theme and subthemes of ‘challenges of implementation’. Throughout the interview process, participants reflected on the barriers to implementing the study, both from their direct experience of having been involved in the early stages of the project and their wider knowledge of the system. It was evident that there were three main areas that posed challenges: ‘chaos in the system’, ‘conducting a RCT in a local authority context’ and ‘elements of the VIPP-FC’. These subthemes operated on varying levels from a more macro systems, cultural level, to specific aspects of the intervention itself.
Chaos in the system
All of the participants in the study spoke about the struggles of implementing a RCT in a setting that can feel quite disordered and in which there is a lot of disruption. They mentioned difficulties within the professional network such as managing multiple priorities and the often-changing structures within local authorities as key potential barriers to successful study implementation.
Priorities
Additional work
Considerable emphasis was placed on the idea that being involved in the VIPP-FC study was something ‘additional’ to many professionals’ everyday work. It was commonly noted that social workers were often not consistently engaged in supporting the study; the study was not a mandatory activity and therefore could be easily placed on the ‘backburner’:
. . . because social workers are so busy and so stressed and have such a high demand to deal with whatever is the most crisis situation, that research, however much they support it, it’s kind of on the bottom of their list.
One of the social workers mentioned that perhaps as the study was more of a ‘mental health’ intervention, rather than a social care construct, this may have also contributed to the study being viewed as an add-on and not wholly relevant to their work. Several participants echoed this sentiment, questioning where the ‘incentive’ for social workers to engage in the research would be, as it involves extra work and they may not perceive many potential gains for them as social care professionals.
Although the ‘key learnings’ theme captures participants’ views about potential solutions to challenges in greater detail, the discussion around the ‘additional work’ theme often generated ideas about how to overcome the specific challenges within the system, such as the social workers having to juggle and manage the priorities of the local authority, on top of those presented by the VIPP-FC study. Three of the participants felt that in some way making the study an intrinsic part of one’s role (e.g. making it effectively ‘part of their job description’) would be helpful in giving them organisational permission to allocate time to and prioritise the VIPP-FC work. This underscores the need for buy-in at a relatively high level within the local authority, as well as specific discussion and organisational planning to cascade these high-level priorities down to social work teams.
Capacity
Linking to the previous subtheme, capacity and lack of resources, primarily in the local authorities, was highlighted as a challenge by all participants. Many of the participants, whether or not they were social workers by profession, were able to empathise with how ‘incredibly busy’ social workers were and recognised that research, albeit with an intervention component, would not necessarily be a priority for them. Interviewees often expressed concerns that what the study needed from the local authorities, in terms of participation, may be unrealistic given the hectic environment of many local authorities, which often do not represent ‘research-friendly environments’. Several participants expressed feelings of frustration with the system not being able to support the research processes more, while also showing awareness of why this may be the case:
. . . they have got no manpower, they’ve got no spare capacity to support the research process . . .
System structures
Structural issues
Participants spoke of the structural changes and organisational issues that local authorities often face as a significant hindrance to the smooth running of a trial. There was a distinct feeling among stakeholders that conducting research within local authorities is a uniquely ‘complex’ and demanding process. One cause of this, identified by several interviewees, was the lack of standardised organisational structures across different local authorities:
There’s so many layers, and every local authority is different, every local authority has a different structure and a different attitude . . . corporate structure and management structure about who you need to speak to, you cannot generalise information.
Not only was variability in organisational structures identified as a challenge, but many of the participants also spoke of the struggles within the network to bring people together. They commented that, in looked-after children’s services specifically, there was often a ‘fragmentation’ within the network and that perhaps if this could be overcome or avoided, then local authorities could provide a more ‘benign environment’ for research to take place. Participants also spoke of the regularity of change and the impact that this can have on study momentum. They discussed difficulties encountered in the setting up of the study because key stakeholders had moved positions, ‘lost jobs’ or became invested in something else. Alongside these issues, participants identified the ‘churn within social care’ as an important potential barrier to successful implementation of the study:
We have a huge problem with turnover and I think all of these things make a kind of perfect storm really and set up a context where it’s really easy to disengage with anything else other than the basic work that you’re meant to do.
A few of the social workers spoke directly about the ‘internal politics’ that went alongside these structural obstacles. They voiced some frustration at the system being unwilling or unable to create space for research within its organisational structures, despite, in principle, willingness. In some instances, interviewees felt that this may create unhelpful pressure on certain members of staff to try to implement the VIPP-FC without being fully supported by senior managers. An additional structural challenge that participants referred to regularly related to positioning, both in terms of who to target in the system to gain support for the study, and also more generally about the challenges of being an outside body coming into an organisation where they are not necessarily known. A few participants spoke of how difficult it can be to work in such a complex system, let alone trying to ‘infiltrate’ it as an outsider:
. . . it’s quite hard for you to know where to press on the system to get things to happen . . .
Organisational inertia
Another interesting subtheme that was primarily endorsed by the social workers who were interviewed was focused on the challenges of change and the organisational inertia that one often faces when introducing something new within local authorities. Some participants spoke of how local authorities sometimes adopt a somewhat defensive position to the prospect of change and new studies/interventions. They reasoned that this may be as a result of the system undergoing so much structural change already:
There’s been so much change in terms of policy and direction and government interference and regulatory bodies, you know, in the last 10 years it’s really ridiculous the amount of change.
Many of the participants were able to empathise with this scepticism of change, describing local authorities as being under significant pressure and constantly having things ‘coming at you from all directions’. One participant highlighted that change, in general, is difficult on an individual level, but trying to mobilise change in a system that is used to and ‘comfortable with’ doing things in a particular way is ‘almost impossible’.
Several participants spoke specifically about how this reluctance to engage in something new and innovative is perhaps reflective of a wider organisational culture. They recognised that local authorities can be quite ‘a hostile place’ to work, as employees often have to deal with difficult dynamics and real or potential criticism from external sources. They hypothesised that external researchers could be perceived as another external actor judging or interfering with their work, without understanding the struggles and pressures they face:
Of course, I mean, from the social workers’ point of view, you know, these are ivory tower people . . . you need to get down a bit.
Conducting a randomised controlled trial in a foster care context
Consent
Thinking about offering an intervention such as the VIPP-FC evoked some concerns among stakeholders; one of the key worries centred on the process of consent. Several of the participants highlighted that the foster care population poses additional challenges to the usual consenting process. Questions emerged over who to obtain consent from, legalities around parental responsibilities, and ‘best interest decisions’. Similar to other themes, a few of the participants reflected on the fact that the complex and sometimes chaotic nature of the children’s social care system meant that obtaining consent for this population was not as straightforward a process as it would be for other populations of children:
. . . there’s layers of consent, it’s just . . . they’re there to protect children, but they’re horrendous to navigate.
Respondents discussed the fact that, for this population, consent is not only multifaceted, but also distributed throughout the network, with different people holding more or less influential roles in the process. Linking with this, several of the participants spoke about the dilemma of whether or not consent would be needed from birth parents. Some social work interviewees in particular felt unsure over whether or not this might, in fact, be harmful to the child and their relationships, as they felt that the child had been removed from the family home ‘for a reason’, and, as social workers, they needed to protect the child and do what is in the child’s best interests, which might conflict with what parents would want. Other participants felt that asking the parents could also complicate the consent process, as parents may be irritated that foster carers were being offered support that they might perceive as helping the foster carer to ‘bond’ with their child, while they may not have been offered anything to help their relationship with their child. One participant role-played how they envisaged some parents to respond to this:
Well why does the foster carer get offered that intervention? Why didn’t I get offered it as part of the package to support me in my parenting, so I could become a better parent and keep my child?
Participants also provided several examples of how shared parental responsibility and whether or not court proceedings were ongoing could make consenting to research difficult and be a possible reason why recruitment rates in the case series were low. In general, the respondents reflected on the tensions between the rights of parents to be consulted and involved in decision-making, the possible conflicts of interests and complex relationships parents may have with the local authority, and the best interests of the child.
Ethical dilemmas of a randomised controlled trial
Participants highlighted a number of different causes for concern in conducting a RCT with foster children, which left them pondering whether or not it felt ethical to conduct a RCT with foster children. A significant proportion of the stakeholders felt that there was a ‘diagnostic’ element to the inclusion criteria of the RCT. They raised concerns related to the potential pathologising impact that a ‘diagnosis’ can have, especially on a young child, as well as fears as to what it may mean for a child’s prospects (e.g. for adoption) if the ‘diagnostic report’ was available to non-research staff:
Yeah you don’t want to thwart their chances of being seen as a child, an adoptable child . . .
Participants also expressed discomfort with specific elements of the exclusion criteria and with some children not receiving the intervention, a process that is expected in a RCT model. Specifically, several participants felt that the age range and focus on RAD meant that children who they felt could have benefited from being involved in the study did not meet inclusion criteria. Likewise, participants described concerns that, if a child was randomised to treatment as usual, with this population that ‘could mean [they receive] nothing’. All the participants who thought that this was an issue also felt that perhaps this could be a barrier to people’s engagement with the study, as they reasoned:
. . . it would be easier to get families to put themselves forward if they know that they would get something, whereas with this they may get something or they may not.
In this way, participants began noticing that perhaps there is a conflict between their way of looking at things as clinicians and the needs of a research study. Although the majority of the participants who noted this dilemma were able to appreciate that the study needed to have certain guidelines to ensure ‘fidelity’ and ‘study efficacy’, they sometimes found it difficult to switch from their clinical mindset and step into that of a partner in a research study.
Information-sharing
Participants provided many examples of how information-sharing with this population, particularly in a research study, can be challenging to manage. Most participants felt concerned about how and what information would be shared. They emphasised the need for researchers and those involved in the study to ensure that they took the necessary precautions to safeguard these ‘vulnerable children’. The social workers interviewed felt particularly strongly that their role is to protect the children in their care, and so they expressed the most concern about how the team as an external body would manage the data and the video-recording element of the intervention:
. . . you are, you know, a third party to these children’s lives, not directly involved in their care or offering day-to-day therapy, so, for me, there’s a little bit of that sharing information and then you step out, so I think it’s about the confidentiality . . .
However, a number of participants then also wanted information gathered in the study to be shared within their network. They felt that it would be a shame for the potential learnings and information collected about both the child’s needs and the foster carers’ skills to be lost to the wider professional network, as it could benefit future placements and planning. This, they recognised, posed a dilemma as to how much, when and with whom this information should be shared; participants feared that a report that was less positive could be misconstrued by some actors in a child’s professional network, now or in the future.
Elements of the VIPP-FC
Manualisation
Opinion among those using the model was somewhat divided on the benefits and drawbacks of the structured and manualised nature of the VIPP-FC. Many of the VIPP-FC interveners felt that having to follow a more ‘rigid’ protocol than they were used to meant that they often found themselves having to say things that did not feel fully applicable to the families with which they were working. This left some practitioners feeling ‘stuck’ and unsatisfied on occasion. Some participants shared that, although the manual was thorough, it would have been helpful if it allowed more flexibility and greater use of ‘clinical judgement’:
. . . it would just be really helpful to have the flexibility to adapt the message . . . certain messages you really, very strongly felt didn’t apply to that family, you didn’t have to read them out.
Some of the participants reflected that perhaps the VIPP-FC is a ‘different way of working’ that may not suit everyone’s style, including that of foster carers, as different foster carers may hold different beliefs about ‘how to discipline their child’.
Although many of the participants felt that the VIPP-FC could be improved by allowing some deviations from the manual, a few participants praised the manual for being mindful of these children and their contexts. Specifically, they felt that the ‘language’ was appropriate and helpful in reflecting the ‘experience of children who perhaps are in transition’.
Logistcs
Participants with experience of offering the VIPP-FC or working with children in foster care provided an abundance of examples of practical and logistical obstacles they faced in trying to deliver the intervention according to protocol. One of the most significant difficulties faced was the available time of the foster carers. They detailed struggles in getting foster carers to commit to the programme and in trying to manage a clinician’s time while fitting in birth parent visits, court dates and various appointments:
. . . so, there were four children in the placement of different ages with different contact arrangements and different needs, and a single carer.
Another key challenge centred on both the practical and ethical issues of other children being in the home. Nearly all of the participants who mentioned logistics highlighted this dilemma, detailing how sometimes it felt as though foster carers’ attention was divided by other children ‘running around’, which made the feedback process challenging:
It would be easier if the child wasn’t there during the feedback, so yourself and the parent could have a proper conversation, without the parent being distracted by trying to manage the child’s behaviour all the time . . .
Others felt that their visits were quite time-consuming; they did not want other children to feel left out, so they felt that they had to spend time interacting with them as well. Some participants questioned the timings of the VIPP-FC and spoke of concerns relating to placement timings. They explained that it can sometimes feel uncomfortable offering an intervention that may increase attachment when there was uncertainty over whether or not the child would remain with that foster carer for long.
Overarching theme 2: benefits and impact of the VIPP-FC
Box 2 presents the theme and subthemes of ‘benefits and impact of the VIPP-FC’. The likely benefits and impact of being involved in the VIPP-FC study were separated into three distinct subthemes, each relating to a different layer within the ecological system.68 The first, ‘creating an evidence base’, explores the wider system-level benefits that the study could have for developing the evidence base for support for children in foster care. The second relates to service-level benefits for engaging in the research and the impact the study could have on the practices of services. Finally, participants highlighted the potential positive benefits for the child and for the wider caregiving system around the child.
Creating an evidence base
There was a unanimous feeling among stakeholders that creating an evidence base for children in care was hugely important. The current ‘lack of evidence’ was highlighted as being ‘problematic’. Several of the participants mentioned their ‘excitement’ and eagerness to be involved in research that could potentially ameliorate this issue. Many participants elaborated on this by explaining that, even when services are following what is commonly agreed on as good-practice guidelines, they still lack high-quality, scientific evidence to support their work, and are therefore keen to work to change this. This included consideration not only of direct therapeutic interventions, but also of methods for assessing children’s needs:
I think the SDQ is poor, which is nationally what is used for foster children for outcome measures. It is not the most sophisticated tool. It doesn’t really work for some children because it doesn’t really pick up any of the nuances . . . it doesn’t capture exactly what they go through.
Stakeholders who held more strategic roles, such as managers, identified the building of an evidence base as an essential component for commissioning of services. They spoke of how research could potentially aid future decision-making, contribute to developing NICE guidelines for this population and allow services for children in care to ‘negotiate money differently’.
Benefits to services
Generally, participants expressed positive views about the VIPP-FC from a service-level perspective. Many of them referenced the benefits of being able to offer something additional that was ‘free’ to vulnerable children and families who receive ‘little support’. A few participants held the realistic standpoint that, although the VIPP-FC is not the ‘panacea that means all placements are great’, they felt hopeful that it would add to the repertoire of interventions offered, and thus give both families and clinicians more choice regarding treatment. A couple of participants also highlighted the benefits of being an organisation engaging in research, particularly when undergoing inspections or scrutiny from external bodies.
Participants whose services were actively involved in delivering the VIPP-FC spoke of how their service had developed as a result of engaging in the RCT. Service managers spoke about it being ‘nice’ to be able to offer staff new opportunities for continuing professional development that may not have been available otherwise. They also reflected that their teams had been able to share some of the learning from the VIPP-FC training, which brought new perspectives to their work.
Several participants expressed the view that, through engagement in the study, services would benefit not only from having clinicians with additional skills, but also by providing foster carers with more resources and strategies, which, in turn, may improve the quality of foster care and even potentially lead to cost savings for services:
. . . if there are less placement breakdowns and more placement stability, that’s going to save the local authority money . . .
There were mixed comments in relation to the impact on time and workload that came with study involvement. A handful of participants felt that being involved was not ‘overly onerous’, and therefore could easily fit in with the service’s usual day to day. However, some raised that the training and the intervention itself were more ‘time-consuming’ and ‘intensive’ than they had expected, which meant that there was ‘a limit to the amount of it people can do, along with their day job’.
Benefits to the families
Nearly all of the participants with experience of the VIPP-FC described it as a positive and ‘enjoyable’ intervention for both families and practitioners. The strengths-based nature of the programme was highlighted as a core feature of its positive reception. There was a sense among the stakeholders that the VIPP-FC was more ‘therapeutic’ and ‘holistic’ than many other approaches currently offered, and that it was less likely to have any aversive or ‘harmful’ effects.
Participants spoke of how the visual aspect of the intervention is empowering, as it goes beyond praise or verbal feedback and allows foster carers to see for themselves what they are doing well:
. . . things which usually stick in your mind may be the more difficult moments, whereas, when you watch it back on video, you can’t escape those positive moments, especially with the way that VIPP[-FC] really highlights them.
Several of the VIPP-FC interveners also emphasised the power that video has in building foster carers’ insight and ability to be more mindfully present. They referred to the ‘slowing down’ during the video feedback as being a key moment for foster carers, allowing the moments of sensitivity and attunement to be ‘magnified’ and further reinforced. Alongside this, participants thought that this gave them and the families an opportunity to spot things that may otherwise have been missed.
Practitioners also felt that the intervention carved out valuable time to focus on important aspects of a child’s development that are crucial for relationship-building, specifically noting the importance that the VIPP-FC attaches to ‘play’. They reflected on their experiences of seeing the children engage in games and playful interactions and how ‘special’ the experience of those minutes of interaction were for the child.
Several participants spoke of how engaging in the VIPP-FC study could have a lasting impact on families. They shared that, even if the VIPP-FC had not been completed in its entirety, families could still benefit from its core messages of sensitive responsiveness, sensitive discipline and attunement. Nearly all of the participants felt that the intervention would stand to benefit both children and families in the long term by building children’s resilience and foster carers’ understanding, regardless of whether or not they were still placed together:
. . . they’re going to be able to use those skills, not just with this one child, but with all the children who go through to stay with this particular foster carer.
All of the participants were hopeful that the VIPP-FC intervention would allow foster carers to feel better skilled and supported for working with these vulnerable children.
Overarching theme 3: key learnings
Box 3 presents the theme and subthemes of ‘key learnings’. This overarching theme consisted of two main subthemes: ‘importance of relationships’ and ‘recruitment fundamentals’. The latter comprised three subthemes, which addressed barriers to and facilitators of the recruitment process.
Importance of relationships
Nearly all of the participants mentioned the importance of adopting a relational approach to this type of study. Several of the participants spoke of the confusion they encountered at various points, both in the set-up and delivery of the VIPP-FC. Some participants felt that they were not always clear who was responsible for which aspects of recruitment:
‘Oh, I thought you were doing that’, ‘no I thought you were doing that’, ‘oh you mean we must do that?’, and so that sort of is problematic.
These participants felt that it was important to ensure that there was ‘clarity’ regarding responsibilities and expectations of the various professionals involved in the study, and widespread engagement of all of the relevant parties. It was generally felt by respondents working with local authorities that having an identified ‘key person’ in each team who takes ‘ownership’ and is the service’s first port of call for everything VIPP-FC related is enormously important. For those in sites where this was clearly in place, it allowed them to avoid falling ‘at the first hurdle’. When responsibility was more diffuse or had changed, there was a perception that this was because teams were overburdened with mandatory tasks and no one was willing to volunteer to sign up and take on this added responsibility.
Many emphasised the importance of the sharing of information and learning between the research team and local authorities. Some respondents expressed the view that it was crucial to engage the entire network and to share their findings from individual interventions with local authorities so that they could see the study as beneficial to their care-planning and place greater emphasis on supporting foster carers to take part. However, participants also noted that information-sharing in a research study, compared with routine practice, may be complicated owing to data protection considerations.
Participants noted that, even after establishing an initial agreement and plan with a local authority, there also needed to be considerable resourcing dedicated to maintaining those relationships across the lifetime of the study. One learning point that participants highlighted was that, although it is necessary to obtain support from ‘the top’, it is just as important to engage key stakeholders ‘on the ground’, as these people act as the ‘gatekeepers’ for this type of intervention, even if engaging them is very difficult and time-consuming because of social workers’ lack of availability. Some suggestions were put forward by several of the participants with regard to how to overcome some of these difficulties. A number of them discussed the idea of holding a ‘network meeting’ to share the learning from the VIPP-FC with those involved in the child’s care. However, one participant felt that:
You couldn’t include the intervener into a network meeting or something because it’s just not catered for as part of the research.
Recruitment fundamentals
Lack of understanding
One of the major areas of learning that participants identified related to the (lack of) understanding around the recruitment process. They suggested a number of reasons why uptake to the case series was low and were able to think about potential solutions to these issues.
There was a distinct view among many participants that one of the major barriers was linked to a lack of understanding of the study aims and approach at the ground level, namely children’s social workers and foster carers. Some participants questioned how well the children’s social workers understood the VIPP-FC study and the potential benefits that might be accrued if families engaged. They felt that social workers could become more engaged in the recruitment process if they could see more clearly that the screening process could provide them with valuable information about the child and, furthermore, that the intervention could help the foster carer to be better able to manage potential challenging behaviours:
Perhaps they could complete the questionnaire with the social worker and foster carer together? What might they [the social worker] glean about the foster carer that they don’t already know?
Another issue, also linked to a lack of understanding, concerned the dynamics that sometimes play out between the social worker and the foster carer. Some of the participants noted that foster carers can sometimes feel ‘judged’ by professionals, so they emphasised the importance of being clear about the potential benefits to foster carers and specifically emphasising its strengths-based approach. Communication, particularly with social workers and indirectly to foster carers, was therefore highlighted as critical.
Contextual issues
Opinions were quite consistent in relation to the contextual issues that affect recruitment. Participants felt that although they, as stakeholders, could see benefits to engaging in the study, they also understood social workers’ and foster carers’ apprehensions, as ‘foster care is not a permanent solution’. They felt that the issue of placement instability and unpredictability may block recruitment and may be a barrier to engagement:
Looked-after children, it’s just chaotic, you know, there’s no point in investing in now because who knows what is going to happen in 6 months’ time?
A few participants spoke of the fact that the framing of the intervention as ‘attachment based’ could be off-putting to foster carers, as they may be fearful of forming relationships when the child is likely to move on. They explained that it may be helpful to frame the study as a ‘training opportunity’, which may increase foster carers’ understanding and develop their resources and skills.
Presence
Linking with the earlier theme of ‘importance of relationships’, the presence and approach of the research team was stressed as being important. Several participants felt that the research team would need to be a continuous physical presence within teams and with foster carers, although they recognised the numerous barriers to that, including the lack of access to foster carers prior to obtaining consent and the heavy workloads of social workers. Nevertheless, ‘face-to-face’ communication was considered crucial:
. . . you know you can send e-mails until you’re blue in the face, but until you actually make eye contact with someone and see them, it’s a different thing.
Participants felt that it was important to be visible within the network: attending meetings and making introductions in person. It was acknowledged that this would require a lot of ‘perseverance’ and human resources, but many participants felt strongly that ‘a physical presence and a relationship’ would be crucial.
Discussion
The interviews with stakeholders highlighted the complexity of conducting a RCT in the context of foster care. Although challenges were noted in relation to the involvement of CAMHS in treatment delivery, these were not strongly emphasised in the interviews. The focus was primarily on the context of local authorities, who are a key bridge to foster carers, and on the relationship between the local authorities and the research team.
In keeping with the principles of implementation science, the themes arising from this study could be placed in a multilevel framework, with factors relevant at individual, organisational and cultural/system levels.51 A key focus, however, was on the organisational level. The qualitative interviews highlighted challenges to working in such a complex, and sometimes chaotic, set of systems. Unsurprisingly, difficulties were encountered by stakeholders attempting to secure research involvement and ongoing engagement when it may be viewed by ‘front-line staff’ as a non-statutory and additional task. Stakeholders hypothesised that this may be due to competing demands and heavy case loads, which echoes previous findings by Gilbertson and Barber.57 The implication would appear to be that it could be highly beneficial to find ways to support the local authorities in taking full ownership of the research, perhaps through co-production and joint management. Alternatively, having research members embedded within the local authorities (e.g. through honorary contracts and relevant organisational agreements) may also be a useful strategy for increasing awareness, contact and trust between the two teams.
One of the most significant challenges, as perceived by the stakeholders, was the complexity of the organisational structures within local authorities. As also noted by Mezey et al.,55 we found that the lack of an embedded research infrastructure was a significant hurdle. The stakeholder interviews also highlighted the significance of staff turnover within children’s social care teams, perhaps exacerbated by recent policy changes and budget pressures. The high staff turnover, combined with a fatigued and overstretched system, produced a challenging environment for external researchers to operate in. This echoed some of the contextual issues that Dale and Watson52 encountered in their attempts to recruit looked-after children for health research and the barriers they encountered when approaching stakeholders about their research.
Linking with this, stakeholders reflected on the readiness and preparedness of local authorities to adapt to a new initiative such as a research project. Thinking about the acceptability of the VIPP-FC in this context, participants highlighted the struggles encountered in trying to implement a study in a system that undergoes frequent change and restructuring. Although many other studies have addressed intraorganisational issues affecting the implementation of evidence-based interventions,51,69 few have discussed the defensive mechanisms that organisations may adopt when faced with new demands, which may reflect a form of organisational homeostasis at times of uncertainty, as touched on by some of the participants in this study. Stakeholders noted that holding such a position, although entirely natural, stifles innovation and, in the longer term, hinders innovation.
The study identified several key elements of a RCT that social workers considered potential roadblocks to study engagement. Randomisation was highlighted as an issue that some stakeholders found difficult. However, unlike previous studies, which suggested that this dilemma arises from social workers’ lack of understanding of the importance of randomisation in research, the social workers in this study spoke very positively about the need for bias-free research evaluation. It appeared that their difficulties were more ethically grounded, related to concerns about children not receiving treatment, which matches Dixon et al.’s53 findings, and also finding it challenging to maintain the boundaries between their role in facilitating research and having a clinical responsibility to the children.
Participants drew attention to the potential difficulties associated with using diagnostic terms and a diagnostic criterion as part of the eligibility criteria for entry to the study. Relatedly, information-sharing was also brought up by participants as a potential complicating factor. Although other researchers have encountered social workers’ wariness of engaging in RCTs, few have specifically spoken about the potential effects of sharing information within the network.55 This could be linked to the ages of the children in the current project, as, generally, younger children are more likely to be adopted than their older counterparts; thus, stakeholders may have been concerned about the impact that diagnostic information may have for the children’s adoptability.4
As recognised by Bogolub and Thomas,35 consent was viewed by stakeholders as a complex problem. It appears that there is still confusion in the children’s social care context about who must provide consent and how the process should be conducted. Indeed, this reflected our experience when we sought ethics approval for the study, whereby significant disagreement was encountered between stakeholders more closely aligned with the rights of parents versus those with responsibilities primarily to the child. Although our stakeholder interviews raised this as an issue, there was still uncertainty about how this could be overcome. This uncertainty may be quite widespread in the field and may explain apparent inconsistency in practice in different localities or ethics committees.
The interviews with stakeholders highlighted the potential benefits of the proposed research, alongside challenges they encountered or anticipated. From an organisational and systems level, it was viewed as beneficial to engage in research and contribute to developing an evidence base for children in foster care. The majority of research to date on implementing evidence-based practice has tended to focus on stakeholders as sometimes being sceptical of engaging in health and mental health research;59 however, in this study, participants were excited to be contributing to scientific research and intervention development. Considerable value was placed on developing evidence-based interventions and participants spoke of sometimes feeling frustrated when others did not understand the potential benefits of doing so.
Stakeholders were largely positive about the VIPP-FC as an intervention. It was felt that this adapted version of the Video-feedback Intervention to promote Positive Parenting and Sensitive Discipline (VIPP-SD) has great potential to be beneficial to the families with whom our stakeholders worked. They highlighted how helpful its strengths-based nature was and emphasised the importance of play.21 However, the VIPP-FC programme did elicit some questions and potential difficulties in the minds of our stakeholders. The manualised approach posed some difficulties for some interveners, relating to its perceived lack of flexibility, which could constrain the role of clinical judgement and limit scope for personalising the intervention. Greater flexibility in delivery and, especially, timing was considered worthy of further exploration.
Stakeholders generated several ideas for tackling some of the barriers to implementing a RCT. A key message was the importance of taking a relational approach, giving time and resources to building relationships with all the key players in the system, especially at the ‘front-line’. As Beidas et al.51 emphasised, implementation of evidence-based practice requires three core components: co-ordination, co-operation and communication. Stakeholders highlighted the importance of role clarity, leadership, persistence and physical presence for successful implementation. A key suggestion was to consider involving service users (foster carers) as study champions, and even to use video to disseminate people’s experiences of receiving the VIPP-FC, to build confidence and interest in the project in the wider community of foster carers and social workers.