INTRODUCTION

Multisystemic therapy (MST) is a community-based, family intervention for young people on the edge of care or custody with significant systemic behavioural difficulties (Henggeler & Borduin, 1990). Originally developed in the United States to target serious antisocial behaviour (Henggeler et al., 1998), it is based on a social–ecological model (Bronfenbrenner, 1979) following the assumption that antisocial behaviour is multi-determined and is related to the risk factors of the young person and the wider ecology around them (Henggeler et al., 2009). Intervention aims to empower the caregiver to support their child with their behaviours across different systems. It is typically delivered intensively over 3–5 months. MST is probably the most extensively evaluated intervention worldwide for antisocial behaviour in young people, with research demonstrating reductions in rates of offending, substance misuse and out-of-home placements (MST Services, 2024).

Gangs and child exploitation

Meta-analysis has shown that negative peer associations are the most powerful predictor of antisocial behaviour amongst young people (Lipsey & Derzon, 1998). Negative peer association has been found to mediate outcomes for young people engaged in multidimensional treatment foster care (e.g. van Ryzin & Leve, 2012). Youth gang involvement can be considered an extreme manifestation of negative peer associations (Carson et al., 2017) which likely presents a real challenge to positive outcomes. Child exploitation is highly connected with gang involvement (Children Commissioner, 2019). The terminology of exploitation is commonly used in the UK to describe a population of young people who may be ‘gang involved’ but ultimately are being criminally or sexually exploited for the benefit of others (Home Office, 2018). There are strong links between child exploitation and antisocial behaviour (NSPCC, 2024). Victims of grooming and exploitation may offend either as a consequence of their abuse or due to coercion and threats from those exploiting them (Cockbain & Brayley, 2012).

In 2019, the Home Office funded the development of Violence Reduction Units to provide a coordinated strategic response to reduce violence, including gang-related crime. The approach is based on the work of the Scottish Violence Reduction Unit, which adopts the World Health Organisation's definition of a Public Health Approach (Craston et al., 2020). In 2021, new services were commissioned to deliver a multisystem, trauma-informed approach to tackling youth violence (Fox & Thakordas-Desai, 2024).

MST, gangs and exploitation

Despite a large evidence base supporting the effectiveness of MST, there has been less research focussing specifically on MST amongst gang-involved youth. The Home Office's Ending Gang and Youth Violence strategy document (HM Government, 2011) advocates for the promotion of intensive family work with the most troubled families, including gang members, and highlights the potential usefulness of MST. The Serious Violence Strategy (HM Government, 2018) echoed this guidance and endorsed MST as an intervention of choice. Boxer (2011) and Boxer, Veysey, et al. (2015) evaluated the impact of gang involvement on the effectiveness of MST treatment outcomes. These studies showed that gang-involved youths were significantly less likely to have successful case closures after engaging in MST in comparison with uninvolved youths. However, utilising the same sample as the Boxer, Kubik, et al. (2015) paper, Boxer et al. (2017) showed that there were no significant differences in rates of re-arrest between gang- and non-gang-involved youth, suggesting that MST was equally as effective in reducing offending for gang involved and uninvolved youths. This is consistent with findings in the broader MST treatment literature (that some dose of MST still lowers recidivism compared with usual care for juvenile justice-involved youth) (Schaeffer & Borduin, 2005).

Regular negative peer contact has been shown to be a predictor of treatment drop-out amongst young people (De Haan et al., 2013), and less negative peer associations were predictive of positive outcomes in MST (Tiernan et al., 2015). Meta-analysis has shown that MST was less successful in making changes to peer relationships than it was to making changes to family relations or individual adjustment (Curtis et al., 2004). Caregivers further report that the most difficult area in which to intervene was negative peer associations (Tighe et al., 2012).

One study exploring MST therapists' experience of working with gang-involved young people hypothesised that the gang acted as a rival to the intervention because the gang resources to draw the young person in outweighed the resources of the parents (Packer, 2014). Viewing young people as part of a separate and more powerful system led to a sense of hopelessness for therapists and that gang-involved young people presented with an increased risk of violence and criminal behaviour. At the time of conducting this research no studies had explicitly focussed on MST for child exploitation.

The present study builds on initial findings from Packer's (2014) research. This is the first qualitative exploration of young people and caregiver experiences of MST where there are exploitation or gang involvement concerns; the research question sought to understand their experience of MST, including the barriers and facilitators to reducing young people's criminal behaviour and contact with others involved in antisocial behaviour.

METHOD

Design and materials

Two semi-structured interview schedules were developed, for caregivers and young people, respectively, through a review of the literature and following the chronology and aims of MST. Open-ended questions were themed around pre-, during- and post-MST intervention, with additional questions specific to the referral behaviour relevant to the participant and written in developmentally appropriate language. For example, questions included: What was helpful and unhelpful in addressing your child's involvement with criminal activity?, What did you do with your MST therapist or parents/caregivers about the people you were hanging out with?, What was helpful/unhelpful about this? and What parts of this were difficult and why? Expert-by-experience feedback was obtained on the two interview schedules from a caregiver and young person, which included advice on how best to conduct the interview and improving clarity of questions, prior to minor amendment and data collection.

Recruitment and sampling

Participants were recruited from three MST sites within charities and local authorities across three geographical locations in England. Inclusion criteria were young people aged 10–17 years, or their caregivers, who had completed MST intervention within the previous 2 years. The criteria below were developed in conjunction with a separate feasibility trial which explored using MST to reduce the risk of criminal exploitation (Langdon et al., 2023). It was agreed that meeting at least two behaviours in the list would indicate that the young person was at risk of being criminally exploited or gang involved based on risk predictors in the exploitation literature. The focus was on community behaviours rather than home behaviours.

The young person needed to have been referred with at least two of the following behaviours:

• criminal behaviour (conviction, final warning, reprimand within the last year)
• frequently missing from home/absconding
• substance misuse
• school exclusion/attendance concerns
• association with negative peers or unknown adults
• aggressive behaviour outside of the home

MST therapists identified people who met the inclusion criteria and informed them of the research, and if potential participants were interested, then they were referred on to the researcher for further information prior to participation consent. Thus, an opportunistic sampling approach was used.

Participants

A total of thirty-seven participants were identified by MST therapists (twenty-one caregivers and sixteen young people): six did not want to take part, twenty were unable to be contacted and one person consented but then withdrew. Reasons for non-participation were not explored and participant right to refuse was respected. A total of 10 participants were recruited into the study (six caregivers and four young people; three of whom were caregiver–young person dyads). Demographic information about each participant is provided in Table 1.

Procedure

Ethical approval was granted by Royal Holloway University of London Ethics' Committee. In addition, each recruitment site approved the research within their local area. Informed consent was obtained before conducting the interviews (including caregiver consent if participant was 15 years of age or younger). Interviews were conducted using Zoom video calling platform or over the telephone. The duration of the interviews ranged from 40 min to 90 min depending on how focussed participants were in responding to questions.

Analytic approach

Interviews were transcribed verbatim and analysed using an inductive, latent approach to thematic analysis (TA) to understand participants' experiences within their broader social context. TA provided the opportunity for patterns and meaning in the data to be identified and analysed, following Braun and Clarke's (2006) six steps of familiarisation, initial coding, identifying themes, reviewing and refining codes and themes, iteratively clarifying and defining theme meanings and reporting.

Elliott et al.'s (1999) six guidelines for good quality qualitative research were followed, including credibility checks within the research team. For example, a transcription of the first interview was reviewed by co-authors to establish data quality and coded by two authors, with good agreement. Initial themes were discussed by all co-authors, with reference to all ten transcripts, and final themes and sub-themes agreed upon. Thereafter, in line with Elliott et al. (1999), a participant was consulted on the findings and thematic map to ensure themes resonated with their experience. Although there is no definitive sample size for the application of thematic analysis, there is some consensus that data saturation is usually achieved within the first twelve interviews (Guest et al., 2006). With the sample of ten for the present study and with the additional attention to ensuring Elliott et al.'s (1999) six principles for good qualitative research, the authors optimised integrity of the data whilst also attending to reflexivity and subjectivity embedded in thematic analysis (Braun & Clarke, 2019).

Reflexivity

All interviews were conducted by one researcher, who kept a reflective journal throughout the research process. This enabled personal reflections, assumptions and biases to be detailed during the data collection and analysis phases. Reflections were shared with the other two authors, one of whom had extensive experience of MST and the other of the research process. All researchers were post-doctoral educated, white and female, with experience of working within the criminal justice system. There were regular meetings of the research team to discuss how this impacted the interviews, analysis and interpretation of the data with consideration of issues around power and difference. There was active engagement in the research team of mitigating bias in the understanding of the data through this process of triangulation.

RESULTS

Four main themes (changes experienced, improvements in caregiver–young person relationship, facilitators and barriers to change), and thirteen subthemes emerged through thematic analysis (see Figure 1) in relation to the research question. The second theme, although connected to both changes and facilitators, was viewed as a significant theme in its own right. Although not every sub-theme was explicit for both caregivers and young people, the views of each are equally important.

Changes experienced

As a result of engaging in MST, participants reported changes across two areas: referral behaviours and young person's perspectives.

Referral behaviours

Participants reported reductions, and sometimes cessation, in the referral behaviours (aggression, missing, criminal behaviour, and substance use).

‘She wasn't disappearing in the night, I wasn't having to call the police … she was letting me know where she was’ (CG2).

‘I started behaving a lot more in school … stopped getting involved with like the wrong crowd, getting involved with the police’ (YP1).

Young person's perspective

MST involvement with the family increased young people's insight into the impact that their behaviour, particularly going missing, was having on the family system, which motivated them to change their behaviour.

‘The fact that … it were putting them through that much stress, that's what really … stopped me going missing because … I know when my mum's going through stress … she can have epileptic fits … and that's the main … reason I stopped’ (YP1).

Young people developed insight into the potential impact that their behaviour may have on future opportunities and the life they wanted to build for themselves. They became more future oriented after engaging in MST, a change which they noticed had occurred after the intervention had finished: ‘before I was like “no, I don't care” but now I've actually thought I do want a future because I love money … I've realised “why would I hang out with those [unhelpful peers]” … I don't think they [MST] help at the time but like after you start to think, and you do realise it's had an effect’ (YP2).

Young people noticed a change in their perspective with regard to their peer relationships, becoming more aware of the impact that their negative peers were having on their behaviour. ‘They were a bad influence and I never realised that … getting me to do things I shouldn't be doing … doing things with me that I shouldn't be doing at my age’ (YP4).

Improved caregiver–young person relationship

Both caregivers and young people noticed improvements in their relationship with each other. Caregivers described trying to encourage the young person to want to be at home more by increasing the pulls into the family home and increasing their warmth towards their child. Young people valued the relationship they had with their caregiver; noticing that the relationship had become more positive and that their caregivers were more understanding of their situation and difficulties.

Increased pulls into the home

‘We just started doing more stuff together, activities and that, whatever really … I started bowling … and my mum and dad used to come and support me’ (YP1).

‘With MST … we planned what I was going to do when she absconded and … they made me think about things. So, I know that [young person] loves food, so it was kind of ways I could entice [her] to come back home …. So, we went through all those kinds of things that X likes about being at home’ (CG4).

Caregivers encouraged their child to socialise with their peers in the family home as an alternative to them being out on the streets. This allowed some caregivers to supervise their child's whereabouts and behaviour, and enabled greater parental insight in their child's peer relationships:

‘I wasn't keen on these friendships, at least I could keep an eye on it and then make my own judgements. Then I could say to [young person] “I think that person is quite rude, I think that person's really disrespectful”, so I was able to gauge the kind of people she was around’ (CG4).

Caregiver warmth

Caregivers offered support to their child by shifting the focus of their interactions from the negative to the positive:

‘I kind of stopped the … “why are you late?” to … “I'm really glad that you are home”’ (CG5).

The shift was recognised and valued by young people: ‘My mum is better now, like she's nicer to me … my mum used to be a bitch and like wrap me out every time … now so she's listening to [the therapist] and every time we argued she was like “stay calm, this and that”’ (YP2).

Adjustment to parenting practices

Caregivers recognised the importance of containing their reactions to their young person going missing and not responding with anger.

‘I remember the first time she absconded, and she come back, I was like “how could you do this to me? I've been worried sick”. I absolutely went mad and she went again, and she was gone again for another 3 days’ (CG4).

Caregivers described a tension between balancing the safety of their child and implementing clear and consistent boundaries: ‘It was really difficult … because every time the police brought her back, they would say “well don't have a go at her about it because she'll leave again”, so then I felt as a parent “well, you're getting away with it”’ (CG4). Caregivers felt that by not reacting negatively to their child going missing they may have been ‘just letting her do what she wants to do’ (CG2).

Facilitators of change

Specific subthemes around setting clearer boundaries and expectations, creating a network of other caregivers, power of multi-agency involvement and changing peer group were identified as facilitators of change.

Setting clearer boundaries and expectations to address going missing

Caregivers spoke about setting up clear expectations for the young person about staying in contact when they were out of the house to ensure their safety:

‘It wasn't about telling her not to do something, it was like “if you're going to do it you've got to be safe when you're out and about”. It was more like “we need to know where you're going, that you're safe and you need to answer your phone and you need to speak to us”. …So, in that respect that was good because it did work’ (CG2).

In response to clear boundaries being put in place, young people were more open to communicating their whereabouts with their parents: ‘If I'm staying somewhere, I'll tell her now instead of just not telling her and making her worried’ (YP2).

Participants talked about a curfew being introduced to set explicit boundaries about when the young person was expected to be home at night:

‘I had to be back by half 10 … I think it's when she [MST therapist] started speaking to my mum that they tried to put a time in’ (YP2).

‘… if I'd phone her, she needed to answer … if she was going to be late, she'd phone me up … Sounds like a small thing, but it really was quite massive’ (CG2).

Network of other caregivers in addressing going missing

Caregivers described benefits to creating a peer support network, which they could use when their child went missing. ‘We built up a bit of a kind of a contact base for his friends and friends’ parents, where they all live and stuff like that. So, if for any instance he was to go missing or we didn't know where he was, they were our first kind of point of call before calling the police’ (CG1).

Power of multi-agency involvement in addressing going missing

Participants spoke about how MST brought together multiple agencies including the police, social services, specialist exploitation teams, support workers and schools:

‘Through MST we got other people involved … there was … the police and I was working with one of the community police workers as well, and as a group [of agencies] they put the ban from the local area, and the harbouring notices ¹ and the ban from the peers he was hanging around with. So, it was a group effort, but they [MST] did help to get those things into place’ (CG3).

Similarly, caregivers acknowledged the importance of professional teamwork in managing a young person being missing: ‘I can't do this by myself. Everybody was involved and helped me. MST, police, school teacher, so everybody helped little, little, little. It was very constructive and good’ (CG6). Good communication and information sharing was particularly important: ‘I think it helped also for everyone to know what was going on. For everyone to share information so I don't have to notify school that he's gone missing, they already know. MST already know so we can work on that in the next session, so it's easy for everyone to know’ (CG5).

For young people, multi-agency professional involvement was experienced as intrusive and motivated young people to change their behaviour in the hope that this would reduce involvement with their family: ‘I just don't like feds … I don't want feds on my back, I don't want to be known like “on the register” or whatever it is they do … the only thing now to get rid of my social workers is going to school and she was like “if you just go to school and try to do it, like they will be gone”’ (YP2).

Changing peer group changed behaviour

Caregivers described encouraging their child to build new relationships with positive peers and creating opportunities for their child and positive peers to socialise.

‘I introduced him to … his old group of friends which were better peers and … if we were having a BBQ or something like that, to invite some of his older friends over… I certainly noticed changes in [his] behaviour. I think because he was hanging around with those better peers, his anger was less and his violence was certainly less because I think he kind of thought “if I'm really angry and nasty to these people, they are probably not going to want to hang around with me anymore”’ (CG5).

Young people recognised that their parents were encouraging them to spend more time with friends of their own age and that changes to their peer group resulted in them participating in more age-appropriate activities: ‘they were more around my age group and more wanted to do stuff that I wanted to do … going out playing, riding bikes, playing football, all stuff along those lines’ (YP4).

Barriers to change

Caregivers identified barriers around young people's associations with negative adults and frequently changing peer groups. These were not identified by the young people themselves.

Associations with negative adults

Caregivers described young people associating with older peers, often adults in the local area. They described the strong pulls and techniques used by adults to groom and criminally exploit their child into selling drugs.

‘She was 13 the first time she went missing … When she was coming back after disappearing she was coming back with new clothes … two guys, 24 and 27, … were trying to get [her] to run the county lines and then started to buy her clothes and say “look you can have the latest phone, you can have this” and at 14, she was obviously quite drawn in by the idea of a new phone’ (CG4).

Caregivers described becoming involved in dangerous situations when they confronted these adults whilst out looking for their missing child:

‘My dad went with my brother to try to get [young person] back from his friends’ property and my Dad was chased by some … men in a car and they tried to drive him off the road at 100 miles an hour’ (CG3).

Other negative adults included the caregivers of negative peers, who were barriers to stopping their child from going missing and engaging in criminal behaviour. These adults encouraged antisocial behaviour by providing their child with alcohol, drugs and food, despite harbouring notices being placed on the property, and were encouraging the young person to sell drugs on their behalf:

‘Her mother selling the drugs and make a party for the kids … when I go near the house, I already smell the drug … and I say “my daughter is in your house?” and she was “oh yes [she] is here” … I was saying “please give me my daughter” and she say “no”’ (CG6).

Frequent changes in the young person's peer group

Frequent changes in friendship groups made it difficult for caregivers to keep track of who their child was associating with and where:

‘It lasts for about three or four weeks and then she'd move onto another group of people, so we never really knew where she was or the people she was with. She changed her friendship groups all the time, so it was very difficult to track her’ (CG2).

DISCUSSION

This research explored young people's and caregiver's experience of MST where there were concerns around gang involvement or child exploitation.

Changes experienced

Participants reported reductions in substance use, associations with negative peers, missing episodes and use of violence. This is consistent with previous meta-analytic research showing significant treatment effects for MST in reducing substance use and delinquency (van der Stouwe et al., 2014). The findings of this research suggest that MST may be effective in reducing problematic behaviour amongst young people at risk of exploitation or gang involvement, consistent with Boxer et al. (2017).

For young people, greater consideration of the impact of their behaviour on others and ideas for a future life were shifts in perspective following MST. Previous research has found that positive goals and aspirations for the future were important processes of change in MST (Tighe et al., 2012), and young people were increasingly thinking about their goals for the future and their life direction after MST (Conroy et al., 2021; Paradisopoulos et al., 2015).

Improved caregiver–young person relationship

Participants described increased warmth and spending more time together doing activities as a family that the young person enjoyed and reported improvements in communication and understanding of each other. The subtheme of ‘increased pulls into the home’ relates to existing literature on the push–pull framework of gang involvement (Decker & Van Winkle, 1996). Specifically, caregiver knowledge of a young person's whereabouts and the young person's willingness to disclose this information was predicted by adolescents' perceptions of parental warmth. Parental warmth has been shown to have a direct effect on youth delinquency (Klevens & Hall, 2014). Caregivers reported experiencing difficulties balancing parental warmth with the implementation of clear boundaries. Authoritative parenting style, defined as high control alongside high parental warmth, is where caregivers are responsive to the needs of the child whilst developing and maintaining clear and well-defined expectations for the young person (Baumrind, 2005). This parenting style is associated with more prosocial behaviour in comparison with authoritarian, permissive or neglectful parenting styles (Mensah & Kuranchie, 2013).

Facilitators of change

Both caregivers and young people discussed that clearer boundaries and expectations were put in place regarding when they were expected to be home by introducing a curfew. Young people reported being more open with their parents about their whereabouts and kept in better contact with them when they were out of the home. Parental knowledge, which is the extent to which parents are aware of the young person's whereabouts and activities, has been shown to be a key predictor of antisocial behaviour (Trentacosta et al., 2009). Interventions which specifically target parents' skills in monitoring their child's behaviour are effective in improving the young person's behaviour (Dishion & McMahon, 1998). Within the context of MST, studies have found that improved parental monitoring decreased negative peer associations (Huey et al., 2000). This study highlighted the importance of peer support networks (in line with Child Safeguarding Review Panel, 2020) and a consistent multi-agency approach to addressing young people going missing. The involvement of multiple agencies was a common experience for families, which acted as a motivator for young people to change their behaviour to reduce the number of professionals involved. Caregivers talked about the value of consistency in the professionals supporting them and the importance of shared communication between agencies. A joined-up response and multi-agency working has been highlighted in various reports to ensure a co-ordinated response to mapping, identifying and safeguarding children at risk of exploitation (Children's Society, 2019; Ofsted, 2014). Participants recognised the value of the development of new relationships with positive peers to engage in more prosocial, age-appropriate behaviour. Decreasing association with negative peers and increasing affiliations with prosocial peers is a key aspect of the MST model (Henggeler et al., 2009).

Barriers to change

Caregivers identified their child's association with negative adults as a significant barrier to reducing their involvement in criminal behaviour. Relationships with controlling adults or older peers are known vulnerabilities for child exploitation (Home Office, 2018). Evidence of families and caregivers colluding with drug dealing and criminal exploitation as a means to earn extra money for the household has been shown previously (Violence & Vulnerability Unit, 2018). Caregivers felt that their child had complex relationships with multiple different peer and adult groups who were a negative influence on them. Caregivers described peer relationships as being transient and continually changing, which made it difficult to completely reduce the impact of negative peers. Tighe et al. (2012) found that parents believed their child's contact with antisocial peers was difficult to change and they felt powerless in addressing this.

Implications

Setting clearer expectations, adjustments to more authoritative parenting style, improving the relationship between children and caregivers and weakening association with negative peers by introducing more positive, prosocial peers were effective facilitators of behaviour change for young people. These are strategies MST therapists regularly use to address antisocial behaviour and negative peer associations, and these findings suggest that these approaches may also be useful to families where there are exploitation concerns.

Association with negative adults is a key risk factor for exploitation and may represent a unique challenge for this population, which is difficult and complicated to intervene with. The power and influence of the parents of young people's peers in encouraging and normalising antisocial behaviour may offset and challenge the work done between parents and therapists in MST to intervene with young people's problematic behaviour and negative peer associations. The frequent changes in peer relationships and associations with negative adults presents a complex problem for therapists, which may be difficult to intervene effectively. There may be a need for enhancements to the intervention to address the unique challenges with which those at risk of exploitation present. MST therapists working with these families should direct particular attention to the multiple negative peers and adult associations to make effective and sustainable changes to behaviour and to reduce risk of exploitation.

Study strengths and limitations

This is the first study to qualitatively explore young people at risk of exploitation and caregivers' perspectives of MST. Highlighting the voices of both caregivers and young people has provided an important contribution to the limited existing evidence base in this area. Despite best efforts to recruit from multiple sites, the final sample was drawn from three MST sites in England, which may not be representative of the wider UK population. There was a lack of ethnic diversity in the sample as the majority of participants were white British, which is not reflective of the over-representation of ethnic minority young people in the youth justice system (YJB, 2021). The caregiver sample were all mothers, and a lack of fathers as participants in this research project mirrors previous research which shows that fathers are under-represented in child and family orientated research (Phares, 1996). Due to recruitment challenges, a relatively small sample of young people agreed to take part in the study. There was also not an objective measure of peer relationships in the inclusion criteria.

There is potential for selection bias as an opportunistic recruitment method was used and those participants who were identified by therapists and agreed to participate may have been more motivated to do so. Therapists may have been more likely to contact families with whom they had positive treatment outcomes or established good therapeutic rapport. This selection bias may have led to a more favourable interpretation of the MST experience than those who did not engage. However, opportunistic sampling in qualitative research can be beneficial as it can identify ‘information-rich cases’ that are able to provide richly textured information about experiences that supports the expansion of knowledge about a research topic (Vasileiou et al., 2018). The exclusive use of treatment completers in the study meant that potential barriers to engagement in MST for this population were not explored.

CONCLUSION

This study used a semi-structured interview approach to provide an initial exploration of young people at risk of exploitation and their caregivers' perspectives of MST. Whilst the findings highlight further considerations for therapists, such as an increased emphasis on understanding multiple peer and adult association, the utility of existing interventions widely used in MST are highlighted. These findings add to the existing and expanding UK evidence base on the efficacy of MST for the at risk of exploitation population. It is hoped that the clinical implications of the study findings can enhance the practice of MST therapists working with families in which there are exploitation concerns.