Rationale for choosing SBNT and MET interventions

Two evidence-based interventions, MET⁶⁹ and SBNT,⁷⁰ were chosen to be adapted as they have been shown to be effective in decreasing substance use in a range of participants including adolescents.⁷¹

Motivational enhancement therapy is a concentrated version of motivational interviewing. This client-centred, counselling approach adds a problem feedback component to standard treatment.⁷² The problem feedback component enables the practitioner to reflect on the material elicited about the impact of drug and alcohol use on the young person’s mental health, physical health, relationships, behaviour and offending, and encourages the young person to discuss this further, for example:

We discussed the fact that your foster carer was worried about your drinking. You told me that you found that you were more irritable the day after you had drunk alcohol. Can you tell me more about this?

Within the MET approach, there is a basic assumption that the motivation and responsibility for change lie within the client, and it is the therapist’s role to create an environment to enable the client to change. A systematic review by Carney and Myers⁷³ concluded that motivational interviewing and MET have shown therapeutic promise for adolescents with problem substance use.^73–75

Social behaviour and network therapy is a counselling approach which utilises a combination of behavioural and cognitive strategies to help clients build social networks that are supportive of positive behaviour change in relation to problem substance use and goal attainment.⁷⁶ NICE recommends family interventions when working with young people presenting with complex needs, such as substance misuse and mental ill health.⁷⁷ SBNT offers an intervention with the potential to galvanise a support network for children in care that can draw support beyond the immediate family. This is important, as most forms of help focus mainly on the individual with the drug and/or alcohol problem and pay little or no attention to the social context. In addition, given the potential for family fragmentation and broken relationships, it is unlikely that the more traditional family interventions would be feasible for the population of children in care. Therefore, the challenge that Copello⁷⁶ tried to address when developing the SBNT approach was to find a way of working and helping people that takes the social context into account and uses a whole social or family system to help and support change and reduce problems, while also developing an approach that is simple enough to be used in routine practice. The principle of incorporating a support network into the intervention was believed to be suitable to use within this study, as it was hoped that the six sessions could be used as a platform to create a support network, or at least start a dialogue that could consider potentially supportive individuals, and promote cohesion that could ideally continue to be developed beyond the period when the young person is in contact with services.

Although both MET and SBNT have been shown to be effective at reducing substance use in the general population of children and young people, less is known about their effect with those who are likely to have more fragmented family relationships and are currently looked after by the local authority. For the SBNT approach, the nature of social networks and how they differ for children within the care system was paramount to understanding how to effectively engage and work with this group of young people.

Developing a theory of change model

In accordance with guidance from the Medical Research Council (MRC) on developing and evaluating complex interventions,⁷⁸ we commenced the adaptation of the interventions by building a theory of change model relating to our target population. We developed a behaviour determinants intervention (BDI) model for each intervention (RL, RM, EK and AC).⁷⁹ These models highlighted the key behaviours targeted by the interventions, the determinants for change and how the team visualised the proposed change pathways for the interventions. The models are illustrated in Figures 2 and 3. We also considered the absence of appropriate family support and supervisions, and the life experiences which led to an individual’s placement into care, as the central vulnerabilities of children in care. We predicted that an intervention seeking to decrease substance misuse by this group would need to address the behaviour determinants identified in the BDI models (see Figures 2 and 3).

FIGURE 2

Behaviour determinants intervention theory of change model: MET.

FIGURE 3

Behaviour determinants intervention theory of change model: SBNT.

When delivering the MET intervention, the therapist can elicit self-motivational statements from children in care by employing strategies to build and strengthen their motivation. By using this technique it was hoped that children in care could resolve the inherent ambivalence about their substance-misusing behaviour.⁶⁹ When developing the MET BDI model (see Figure 2), alongside the goal of strengthening motivation, we thought that it would also be helpful to provide personalised feedback to assist the young person to consider risks and tip the decisional balance.

When developing the SBNT BDI model (see Figure 3), we thought that the approach should promote the recognition of an informal network of supports that extended beyond traditional caregivers. We were aware that networks of support available to children in care would differ from those available to children and young people residing within more traditional biological families. However, evidence exists that social network support is key to helping people deal with problem behaviours, including substance misuse. Therefore, within SBNT, the therapist could usefully employ cognitive and behavioural strategies to help children in care to build social networks of positive behaviour change in relation to their goal attainment.⁷⁶

Formative qualitative research methods

In-depth one-to-one interviews, dyad interviews and focus groups were used to explore the assumptions inherent within our logic models (see Figures 2 and 3), the principles behind the MET and SBNT approaches and their relevance to children in care, and the broader therapeutic approaches, including the key behavioural and motivational domains that the interventions should address when working with the population of children in care.

All data were collected using semistructured topic guides. Guides were developed and adapted throughout the study in response to early research findings to ensure deeper understanding of emerging themes. Interviews and focus groups were audio-recorded and transcribed verbatim. Data were collected until data saturation was reached within each participant group and no new themes were emerging. Transcripts were anonymised and identifiable participant details removed. A participant key was developed and stored separately. Pseudonyms were allocated to each transcript and have been used within all reports and publications to maintain participants’ anonymity.

Interviews with children in care and carers (foster and residential) were chosen as the method of data collection. Interviews were chosen, as the semistructured nature of the topic guides meant that sensitive issues could be explored and personal experiences shared. Interviews also recognised individuals as experts in their own experiences, in respect of the topics of drug and alcohol use, any barriers and facilitators experienced when engaging with services (e.g. substance misuse services, mental health and services, such as Barnardo’s) and any potential recommendations for service improvements. The topic guides used for the interviews with children in care were developed with the SOLID PPI group, as discussed previously in Chapter 2. All study documentation relating to the formative research is shown in Report Supplementary Material 3.

Focus groups and dyad interviews were chosen as the primary method of data collection with social workers and drug and alcohol practitioners and participants were interviewed with colleagues within the same profession (i.e. social workers were interviewed together and drug and alcohol practitioners were interviewed together). The group interaction encouraged the exploration of a range of responses in a relatively short space of time. Furthermore, they proved to be an effective way to explore issues and quickly establish a range of experiences, views and knowledge. The professional focus groups and interviews enabled us to discuss the original components and principles behind the MET and SBNT interventions, alongside the proposed adaptations and whether or not they were perceived to be relevant to the context of children in care. They also enabled us to explore the broader therapeutic approaches required to work with children in care, the feasibility of delivering the interventions to this population and to consider potential barriers to delivering the interventions at scale.

Recruitment and sampling strategy

Social workers within the looked-after children and 16+ teams (teams working with children in care aged ≥ 16 years and supporting young people who are transitioning out of the care system) approached eligible children in care from their caseload. Eligible participants were defined as children in care aged 12–20 years, known by a social worker to have experience of substance use (previous or current personal use or exposure to substance use), who were able to provide informed consent and who resided in the study area. Social workers acted as gatekeepers, and shared a brief participant information leaflet with the young person. If the young person in care was willing to take part, the social worker completed a written assent or consent form with the young person and returned it to the study team. Children in care (n = 19) expressed an interest in taking part in the study. Once the form was received the study team was able to formally approach the young person in care. All 19 interested children in care were contacted by the research team to discuss the research and to arrange an appropriate time to visit and conduct an interview. Written informed assent or consent (depending on age of the child in care) was taken by the researcher before starting the interview, as described in Consent.

A purposive sample was recruited to ensure diversity with regard to age, exposure to drug and alcohol use, and placement type. The final sample was representative of the population of children in care, in so far as there was an equal mix of male and female participants and a range of placement types across the different local authority areas, as identified in Table 2.

TABLE 2

Qualitative participants

Separate one-to-one interviews (n = 13) were carried out with carers across the research sites to ensure diversity of sample in terms of age, ethnicity and carer type (i.e. foster carer/family member/residential worker). An additional focus group with four carers was also conducted as part of an already established carer support group.

Social workers (n = 8) were purposively approached to take part in a focus group and four took part. This was complemented with two further dyad interviews with four social work staff. Sampling took place to ensure diversity in terms of the local authority site in which they worked, level of experience (i.e. social worker, team manager) and sex. The social workers within the looked-after children and care leavers teams were interviewed because of their key knowledge of the context of children in care, as well as of many of the ethical issues that informed the intervention development.

A focus group also took place with specialist young people’s drug and alcohol practitioners (n = 5). Practitioners were purposively approached to ensure diversity in terms of local authority site they worked in, job title (substance misuse practitioner, service manager) and sex. The drug and alcohol practitioners had key knowledge of interventions that currently work well with young people and they used their professional knowledge and expertise to inform the adaptation of the MET and SBNT interventions.

In addition to the focus group with drug and alcohol practitioners described above, one-to-one interviews were also carried out with three drug and alcohol workers who had delivered youth social behaviour and network therapy as part of a previous trial.⁷⁰ These interviews aimed to build on previous knowledge and experience of delivering youth social behaviour and network therapy to young people within a substance misuse setting and contributed towards adapting the treatment manual for the population of children in care.

We initially proposed to carry out individual one-to-one interviews with children in care and carers, and focus groups with professional participants. For pragmatic reasons we conducted a combination of individual interviews, dyad interviews and focus groups, depending on participants’ availability and preferred method of involvement. On reflection, the research team felt that the combination of interview and focus group data collection had a positive impact on the quality of the data generated. As planned, the interviews enabled us to collect an individual’s thoughts, attitudes and personal beliefs about being involved in and interacting with the child welfare system, and the focus groups provided an opportunity to consider how the interactional data between participants resulted in similarities and differences in experiences being highlighted.

Table 2 shows the qualitative methods that participants engaged in and the demographics of the participants recruited in the first round of qualitative work.

Consent

The children in care aged < 16 years were seen with an accompanying adult (parent, carer, social worker, children’s home lead) prior to the interview taking place and they were asked to provide informed assent. If the accompanying adult did not have parental responsibility (PR), the research team contacted the adult with PR to obtain informed consent. If the parent was not contactable or, in the view of the designated social worker, it was a risk to the young person for the parent to be contacted, the social worker or local authority guardian with PR was contacted to sign the consent form. Informed young person assent and consent, dependent on age and carer consent, were obtained prior to the young person taking part in any element of the study. Information on the study was shared with parents and carers as appropriate.

For those children in care aged ≥ 16 years and for all other participants within the formative phase of the research, informed consent was taken directly from the individual concerned by the researcher. Prior to informed consent being taken, a participant information leaflet was shared with each participant, and the research team talked through the leaflet and provided an opportunity for any questions to be asked. The research team also explained that participants could withdraw at any point.

After informed consent had been given to the research team, interviews and focus groups were carried out by qualitative researchers with experience of working with young people. The data collection took place at a location convenient for the participant. For young people this was in their home or at an alternative convenient private location, which ensured the safety of both the young person and researcher. For professionals, data collection took place in a private location within their usual working environment. When interviewed the young person was given a choice of whether or not they wanted to be accompanied by a trusted adult who would act as an observer; however, only two young people requested this.

Children in care participants were remunerated for their time with a £10 ‘love2shop’ voucher.

Qualitative analysis

Transcripts were analysed thematically;⁸⁰ it was an iterative process, using the constant comparative method,⁸¹ in order to identify key themes and concepts. In practice, this entailed a line-by-line coding process and then analysis within a given transcript and across the data set as a whole. Analysis with qualitative software (NVivo) aided the organisation of thematic codes. The data were compared across the participant groups (i.e. children in care, professionals and carers), with similarities and differences being highlighted. In the first instance, data were analysed by two researchers in order to ensure intercoder consistency and agreement. The main themes and findings were presented to the wider multidisciplinary team that included expertise in a variety of backgrounds, including community child health, public health, social care, social science, drug and alcohol use, and clinical psychology. These qualitative data were used to refine the SBNT and MET approaches to ensure that they were responsive to the needs and views of substance-using children in care.

Data analysis focused on understanding internal and external drivers of behaviour and also on views about interventions promoting well-being and self-care in early life. Components of the logic models (behaviours, determinants and intervention components) were explored with participants to further refine the theory of change pathway and clarify intervention delivery issues.

Finalisation workshops: modification of interventions

A series of intervention finalisation workshops took place, during which findings from the preliminary thematic analysis of the qualitative interviews and focus groups were presented. The purpose of the workshops was to co-produce the final intervention manuals. Within the initial phase of the workshop, the research team presented the main themes that had emerged from the interview and focus group data, and participants were asked to consider the MET and SBNT interventions and discuss what the final manual should ‘look like’.

Five workshops were conducted: one with professionals (n = 14), all of whom had been interviewed earlier; and four with young people (n = 13), none of whom had been previously interviewed. All participants involved in the young people’s workshops either were currently in or historically had experience of receiving specialist drug and alcohol treatment. The workshops were inclusive of both children in care (n = 5) and non-looked-after children (n = 8). We took the decision to include young people not involved with the care system, as this element of the study needed to understand current treatment provision. Owing to social workers not systematically recording this information and gatekeeping to ‘protect’ young people, we found it difficult to identify young people in care currently on their case load with experience (current or previous) of accessing treatment. We decided to approach the drug and alcohol services involved in the study to recruit young people into the workshops. We wanted to ensure that we had maximum variation of young people, regarding age, sex and location of participants, who had experience of accessing drug treatment agencies, to discuss the developed interventions. We held a workshop in each active study site to enable participants to be involved in the study without having to travel long distances to take part. The workshops were all held within the well-established young people’s drug and alcohol services involved in the study, so participants were in a familiar environment.

The workshops provided an opportunity for the research team to present the preliminary findings to key participants. Verbatim anonymised quotes were used to identify areas of potential importance and to facilitate discussion between researchers and participants. Four researchers and participants then worked collaboratively to co-produce the final manuals. Co-production occurred through group discussions and using flip charts and paper to design worksheets and complimentary materials to be used within sessions.

The themes discussed within the workshops are shown in Table 3.

TABLE 3

Themes and subthemes relevant for manual adaptation

Areas of potential intervention adaptation were discussed within and across interviews, focus groups and workshops, and the findings from the qualitative data collection have resulted in a number of adaptations being made to the manualised interventions.

Following the workshops, the final adaptation of both manuals took place. Ongoing communication took place between the on-the-ground researchers involved in developing the manuals and the original intervention authors [Professor Alex Copello (SBNT) and Dr Gillian Tober (who has adapted both MET and SBNT for other clinical trials)], to ensure that the core components of each approach were retained throughout the adaptation process.

Theme 1: therapeutic relationships

A successful therapeutic relationship was highlighted as important by both professionals and children in care when working towards reducing substance use. The qualities of trust and genuine care were identified as the two main constructs that underpin a successful therapeutic relationship. The ability of children in care (and often inability) to trust and confide in professionals was a recurrent theme. Professionals acknowledged that children in care often experience disorganised and difficult attachment and recognised that their experiences leading up to their placement in care may have had an impact on their ability to trust other people. Therefore, although trust is one of the necessary conditions for any therapeutic relationship to be successful, it is particularly important for children in care, who may have experienced relationship breakdown and abandonment, being let down and having their essential needs unmet.

Professionals displayed a clear understanding of these complex attachment issues and discussed the need to ‘earn’ trust when engaging with children in care:

You need to put in the groundwork initially. I think with teenagers you need to gain their trust, you need to work for it. Because if they have been hurt, which they will have been, they will try to push you away. They won’t want to trust you.

Carly, social worker, focus group

Owing to the often inherent ‘lack of trust’ in professionals, practitioners recognised that they were expected to demonstrate their trustworthiness when engaging with children in care. Typically this involved practitioners being consistent and reliable. Equally, children in care described seeking qualities such as empathy, reliability and partiality, all of which are qualities that may have been missing in their early attachments. One foster carer described displaying his reliability in terms of being available ‘24/7’, stating that he is permanently ‘on call’ if a young person needs him:

. . . it is not a job because there is no job that makes you work 24 hours a day, 7 days a week and 365 days of the year, but this one does.

James, foster carer, focus group

From the perspective of children in care, the relationship between themselves and their allocated key worker within an organisation was a pivotal factor that determined whether or not they engaged with services. When establishing and facilitating a trustworthy relationship, young people explained the importance that they placed on professionals allowing them to ‘work gradually’, only sharing ‘personal information’ and making disclosures when they felt ready. Furthermore, the idea of professionals making self-disclosures was repeatedly reported: children in care felt it was important for professionals to ‘trade’ personal information, such as a hobby they enjoyed, details of a pet they owned, or an example of how they had resolved a problem successfully in their own lives. This process of sharing information was perceived to be beneficial to developing a trusting relationship, as sharing information was not completely one-sided. Children in care reported that such disclosure enhanced their sense of connection to the practitioner, as well as their own safety to disclose information:

When you work with someone you have to build a bond up first, before you can open up to them . . . It’s, well the way I’ve done [it] is just ask questions about them, and then if they tell you, then you know well if they’ve told me this then I can tell them that.

Sophie, 17, young person interview

Children in care described having multiple professionals involved in their ‘care package’ and a quality that young people desired was ‘genuine care’, inclusive of professionals going ‘above and beyond’ what is expected and providing unconditional care, although they did not always feel that they received this. Professionals, especially those within the social services teams, take on the corporate parenting role. This role dictates that safeguarding and risk management take precedence over the provision of emotional support. Therefore, much of the care a child usually receives from family members within a personal environment is provided by a professional who is employed to provide such care. To demonstrate that they care, many social workers describe being available outside their contracted working hours and going ‘above and beyond’ their role:

Myself and his YOT [youth offending team] worker had agreed between us that we would have our phones on 24/7. So that if he wanted to get in touch and check in we knew he was OK. So we did, we took turns and he did check in and he did arrange to meet up which was really good.

Steph, social worker, focus group

Children in care showed an acute awareness that social workers had a corporate parenting role to fulfil and that carers provided a role that they were ‘paid’ to do. This led to children in care emphasising the importance of practitioners who made them feel like they ‘genuinely cared about their welfare’. Interestingly, foster carers reinforced that they attempted to provide the same level of care and support to both their biological children and the children placed in their care, despite the paid position they were in:

Any child that comes to live with me, I know they are not mine, however I will work with them, I will play with them, I will live with them and I will do everything to my best ability in every area, in every arena because I want what is best for them.

Liz, foster carer, interview

Genuine care also involved professionals showing empathy to the young person and being available to provide ‘unconditional’ support. There was a belief, sometimes verbalised explicitly, at other times more implicitly, that genuine care stemmed from personal investment rather than a contractual obligation:

Like Josie talks to me, not like I’m just someone she has to work with, she talks to me like she cares.

Carla, 17, young person interview

Children in care felt that they were cared for if they were shown unconditional positive regard, regardless of their behaviour. This was a recurrent theme for professionals, who reported children in care regularly disclosing information to them regarding historical experiences that they had been witness to or subjected to. Foster carers described having to respond in a sensitive and non-judgemental way:

We had a young man who had been abused by a family member. He was feeling guilty himself about it and thought that we would feel disgusted that things like that had been done. It is letting him see that we are not disgusted. Straight away, ‘I have heard all of this before, you are not the only one. It is not your fault’.

Carol, foster carer, focus group

The ability of professionals to be non-judgemental was important to children in care, and some participants voiced concerns that practitioners would not be able to ‘cope’ if they chose to share some of the experiences that led to them being placed into care. One young person in care explained that he elected not to engage with services and open up for fear that professionals would then ‘leave him’:

. . . my family is **** up, really **** up. And if I sat there and told someone they’d probably run a mile, they probably would. So that’s why I’ve never really opened up to anyone, cause if I did they probably would run away, do you know what I mean?

Ewan, 17, young person interview

The above quotations identified the practical issues and challenges that needed to be addressed to facilitate a therapeutic relationship with children in care; therefore, it was important to acknowledge the importance of overcoming insecure attachments and incorporating methods of developing a trusting relationship.

Theme 2: engagement and challenges of working with children in care

Throughout the formative data collection, there was consensus that, if used in isolation, the more traditional one-to-one talking therapies were often unproductive for children in care. From a professional perspective, this approach was thought to be overly formal and could result in children in care disengaging with support services. Children in care also verbalised that they found it harder to engage with overly structured and formalised sessions:

It was like in a room . . . and like there’s a table there and it had like little seats round, and like, he was just on about things. Do you know, he didn’t make it very good, like, he didn’t make it very fun and enjoyable kind of thing. It was just like, boring. He was just writing things down that I was saying basically and it just upset me. He just kept on going over it and over it and over it, he was like ‘so how did that feel? Bla bla bla.’ I didn’t really feel comfortable.

Isabelle, 13, young person interview

There was a clear need for practitioners to be equipped with a number of skills and strategies to engage with young people. Practitioners needed to be responsive to the individual presenting to them and described using techniques such as ‘node-link mapping’, as used in the International Treatment Effectiveness Project,⁸² and mood cards, while staying true to the intervention they were trying to deliver:

There are not many young people who you’ll get to the point where you’re doing that one to one counselling really. It is few and far between. You’re being creative . . .

Adam, drug and alcohol worker, focus group

Many of the participants in care expressed their desire to attend sessions that enabled them to be actively involved in the work being completed, with practitioners implementing strategies that facilitated young people connecting with the work and the professional themselves, and maintaining concentration:

Writing it down or doing it like arts and crafts way because I don’t like just talking and having conversations cause I just get a bit bored and lose track, then I’ll start fiddling about.

Abbie, 18, young person interview

Alongside the necessity for an interactive approach, was the need for practitioners to be aware of the complexities of living in the care system for the young person. This awareness would help to facilitate a holistic approach to be taken to the work being conducted and could also help to identify goals that are not solely ‘substance use’ related. Children in care stated that they valued discussions that recognised the difficulties occurring in their lives. Professionals also identified the importance of taking a bespoke approach to treatment:

I think what’s coming out here is that with the kids we work with, the drug and alcohol issue is over there, if you like, and a whole raft of other issues are here. As workers we’re dealing with all of these here and that tends to sort the drug and alcohol issues out quite naturally.

Laura, drug and alcohol worker, focus group

Professionals highlighted challenges that arose owing to the transient nature of the population of children in care. It was identified that, when young people experience frequent placement changes, it can result in young people experiencing fragmented support in terms of changes to key workers, carers and professionals supporting them. It can also result in young people being eager to find friends even if these relationships are potentially destructive:

So they might, you know, have contact with their brothers or sisters, you know, it is just they get moved around, and when they are moved around they are vulnerable, they are desperate to have friends or they are desperate to have somebody to call their own . . . people get attracted to them who are, I would say, not the type of kids I would want my kids to knock around with.

Liz, foster carer, interview

Social support was identified as a potential challenge regarding the SBNT approach. Children in care and practitioners recognised that a positive support network was a central part of SBNT and accepted that social interaction is necessary in the resolution of most substance misuse problems; however, it was felt that support was not always available:

It is quite sad sometimes when they haven’t got anybody in the family, not even an uncle or a cousin or somebody who they can put down as a support really.

Steph, social worker, focus group

The challenges of finding appropriate network members was explored. In many interviews, the participant in care struggled to identify someone that they felt they could turn to, and feelings of not having support and the need to be self-sufficient were verbalised:

My boyfriend and his friends, and there’s a few of my friends. Actually they’ve got their own lives as well, they’ve got their own houses and their partners and they’re all settling down as well, so . . . there’s not really many people there. When you think about it though, how many of them can you turn to if you’ve got a problem? Cause there’s not a lot.

Abbie, 18, young person interview

As was expected within the children in care population, when individuals were able to identify individuals who provided positive support, it was often people outside the traditional family support network. This had potential for sources of support to be transient (e.g. teachers who would change with each school year). Professionals were often identified as sources of support, which could be challenging for the delivery of SBNT, as individuals may not be able to provide ongoing or out-of-hours support in the same way as more traditional family members would. Nonetheless, children in care recognised the support that was provided to them:

There’s two main people I’ve got in my life which provides me with support. One’s my boss, he’s a farm manager, I work with him most days. Another person is the manager of [name of school], he owns the company and he helps quite a lot by, when I moved out of here [residential children’s home] the first time, he’s the one that made me come back, and let me get my head back.

Philip, 17, young person interview