‘I (really) know what you mean’. Mechanisms of experiential peer support for young people with criminal behavior: a qualitative study

ABSTRACT Individuals with a criminal background are increasingly involved in support for people with criminal behavior. However, research into what happens in the relationship between these experiential peers (EPs) and clients is scarce. This qualitative study investigates EPs’ perspectives on the mechanisms of experiential peer support and how this compares to regular support by care providers without lived experiences. We interviewed seventeen EPs who provided support to young people with criminal behavior. The results suggest that shared experiences between EPs and their clients play a central role. EPs identify with their clients, leading to empathy and a non-judgmental attitude. Clients seem to perceive EPs as credible role models who offer hope. EPs’ lived experiences seem to induce a realistic view of desistance and an emphasis on a humane relationship with their client, which is characterized by equality, reciprocity, trust and sincerity. This recovery-oriented approach towards criminal behavior and desistance could also be utilized by non-EPs. Future research should investigate others’ perspectives on and experiences with experiential peer support, in particular those of clients and co-workers.


Introduction
Individuals with lived experiences are increasingly involved in mental health care services offering support to several populations, including people with psychiatric disorders or substance abuse problems. So-called 'experiential peers' (EPs) are particularly active in recovery-oriented mental health services (Kortteisto, Laitila, and Pitkänen 2018), which is not surprising since the recovery movement places great value on empowerment of clients and recognition of their perspectives. The involvement of EPs is also becoming more common in the forensic field, which is concerned with individuals who display criminal behavior and have subsequently gone through the judicial system. Several large cities in the United States have seen an expansion of state-funded peer mentoring initiatives with youths involved in the criminal justice system (Lopez-Humphreys and Teater 2019) and in the United Kingdom peer mentoring was a central component of the 2012 government plans to transform rehabilitation of prisoners (Buck 2018). In the Netherlands, the importance of recovery-oriented interventions, including experiential peer support, is increasingly being recognized in forensic mental health care. Around one quarter of the organizations in this field indicate to work with experiential peers, and an even higher number have expressed the ambition to incorporate this in their treatment or are making plans accordingly (Bierbooms et al. 2017). In 1965, sociologist Donald R. Cressey already wrote about involving 'criminals in the rehabilitation of criminals', saying that individuals who have displayed criminal behavior, but who are now 'on the right track', are the best messengers of anticriminal verbalizations because they know both the feeling of guilt and the absence thereof regarding committing crimes (Cressey 1965). Maruna (2001) also argues that the self-narrative of those who desist from offending often involves transforming one's own history of offending into a source of knowledge, and that these people are often drawn to positions as 'wounded healers', for instance as youth workers or drug counselors (Maruna 2001). Experiential peers may be helpful in reaching the forensic population, which could promote more suitable care and favorable outcomes regarding individuals' behavior. The incorporation of peer mentoring fits with the risk-need-responsivity model, a guiding framework for treatment of individuals with criminal behavior (Andrews, Bonta, and Wormith 2011). The responsivity principle seems particularly relevant for these interventions, as it concerns tailoring the intervention to the individual, thereby paying attention to demographics, readiness to change and personality. More recently added principles such as respect for the person, providing a human service and staff relational skills, also form reasons for incorporating EPs. Research has shown that treatments following these principles are associated with significant reductions in recidivism, whereas interventions failing to adhere to them yield minimal reductions in recidivism (Andrews, Bonta, and Wormith 2011).
To achieve behavioral change, a strong relationship or alliance between the client and the practitioner is important (Burnett and McNeill 2005). A strong alliance has been emphasized as a critical element of psychotherapy, and research has shown a modest but significant positive relation with clinical outcomes in adults (Martin, Garske, and Davis 2000;McCabe and Priebe 2004) and in youths (Kazdin, Marciano, and Whitley 2005;Kazdin, Whitley, and Marciano 2006;McLeod 2011). This alliance is particularly important for youths with externalizing behavioral problems (McLeod 2011;Shirk and Karver 2003) and has shown to be highly predictive of probation success (Hart and Collins 2014). However, attachment difficulties, a risk factor for criminal behavior (Hoeve et al. 2012;Zegers et al. 2008), can hinder the building of a relationship (Smith, Msetfi, and Golding 2010). Establishing a good alliance is also more difficult for individuals who have had professional help before or who have negative expectations of help (Van Hattum et al. 2019). Our previous study on at-risk adolescents demonstrated such expectations: several participants indicated not wanting any help, especially from someone who did not experience what they were experiencing, as these care providers would not understand them (Lenkens et al. 2019a).
Establishing a strong alliance may be easier for individuals who have had similar experiences to the juveniles they aim to support. An experiential peer may have an advantage over other care providers, since people are more likely to connect with people similar to themselves (McPherson, Smith-Lovin, and Cook 2001) and deep-level similarities between individuals enhance the quality of the relationship (Ghosh 2014). Similarity may refer to experiences such as coping with problems, receiving treatment, or facing stigma (Baillergeau and Duyvendak 2016), or going through a major life transition (Suitor, Pillemer, and Keeton 1995), such as that from 'offender' to 'ex-offender'. There may also be more agreement on goals and tasks between EPs and juveniles. EPs may have a different focus and approach to the perceived problem than a care provider without these experiences. Being persons convicted of offenses themselves, they may have a better understanding of juveniles' actual needs, instead of focusing on needs prioritized by other actors.
In our previous paper, we presented a theoretical model regarding seven potential mechanisms of experiential peer support for individuals with criminal behavior (Lenkens et al. 2019b). First, it might be easier for EPs to have empathy for individuals with criminal behavior (Bagnall et al. 2015;South et al. 2014) and to make them feel accepted and included (Buck 2016a). Second, clients might learn specific skills, behaviors, attitudes or knowledge (Davidson and Rowe 2008) from the EP who is a credible and realistic role model (Bagnall et al. 2015;South et al. 2014). Third, it might be easier to establish a trusting and meaningful relationship with EPs (Bagnall et al. 2015;Eby et al. 2013;Laurenceau, Feldman Barrett, and Pietromonaco 1998;Dutton 2018), which might also influence clients' other relationships and their social capital (Rhodes, Grossman, and Resch 2000). Fourth, EPs might exert (in)direct control over the client's behavior, by detecting and correcting risky behavior (Davidson and Rowe 2008;De Jong 2013). Fifth, EPs might model and stimulate self-acceptance and positive identity development (Buck 2016a;Maruna 2001). Sixth, clients might gain hope and perspective by witnessing that recovery is attainable (Davidson and Rowe 2008;Walker and Bryant 2013;Bierbooms et al. 2017). Seventh, EPs might play a bridging role between clients and formal institutions and additional services (Davidson and Rowe 2008).
The area of mental health services has a longer history of peer support, and research in that field tells us something about potentially relevant mechanisms. Empowerment, empathy and acceptance, stigma reduction and hope seem to play an important role (Repper and Carter 2011). Although these mechanisms resemble our theoretical model, it is unclear whether they can be generalized to the forensic setting. Criminal behavior is not necessarily perceived as problematic by the offending person, and treatment or help is often mandated by court and therefore involuntary. Thus, motivation for treatment or behavioral change is not guaranteed. Specific research into the forensic setting is also necessary to rule out 'deviancy training', a negative social learning effect that can occur when bringing together individuals with criminal behavior (Dishion, McCord, and Poulin 1999).
Previous research investigating experiential peer support in the forensic field has found that EPs feel they have a deeper understanding of the client's situation and the pain they are dealing with (Barrenger, Stanhope, and Atterbury 2017;Barrenger, Hamovitch, and Rothman 2019) and clients view them as non-judgmental (Matthews et al. 2019;Thomas et al. 2019;Buck 2018;Collica-Cox 2018). Several studies have found that both EPs and clients see EPs as role models (Collica-Cox 2018;Portillo, Goldberg, and Taxman 2017;Barrenger, Stanhope, and Atterbury 2017), although it is unclear whether EPs' lived experiences contribute to their credibility (Reingle Gonzalez et al. 2019) or rather make them inauthentic (Buck 2016b). It has also been shown that witnessing someone who has succeeded despite a problematic history can be inspirational and provide hope (Kavanagh and Borrill 2013;Marlow et al. 2015;Portillo, Goldberg, and Taxman 2017;Buck 2014Buck , 2016bBuck , 2019Matthews et al. 2019;Barrenger, Hamovitch, and Rothman 2019). In addition, studies have found that building a trusting relationship is important (Kavanagh and Borrill 2013;Matthews et al. 2019;Barrenger, Hamovitch, and Rothman 2019;Thomas et al. 2019) and that it may be easier for EPs to achieve this with clients due to their shared identity (Barrenger, Stanhope, and Atterbury 2017;Portillo, Goldberg, and Taxman 2017;Reingle Gonzalez et al. 2019). Lastly, studies showed that EPs act as a bridge between clients and staff (Hodgson et al. 2019;Barrenger, Hamovitch, and Rothman 2019) and that they make referrals or connect clients to services related to housing, mental health, education and employment (Marlow et al. 2015;Portillo, Goldberg, and Taxman 2017;Thomas et al. 2019).
Although these studies provide insight into mechanisms of the relationship between EPs and their clients, this is the first study that specifically focuses on a predetermined set of mechanisms and examines these in a structured manner, while leaving space for additional mechanisms to arise from the data. In addition, whereas in many studies the sample consisted of support providers with and without a criminal background, making it difficult to disentangle specific mechanisms, we only included experiential peers with a criminal justice history. This study focuses specifically on support provided to adolescents and young adults. The mechanisms at play may be different than for the adult population. It is important to establish an empathic, collaborative and equal relationship with youngsters (Creaney 2018), perhaps even more since the power differences with care providers is larger for young as compared to older people. In addition, role models may be of particular importance in learning behavior and becoming more resilient during this stage of identity formation (Hurd, Zimmerman, and Xue 2009). Lastly, we contribute to the existing body of research by asking participants to compare their own role and the support they provide with that of care providers without lived experiences. In this study, when speaking about 'regular care providers' or 'care providers without lived experiences' we refer to those individuals who provide support, guidance and treatment to these youngsters in a professional context, varying from social workers to psychologists.
This paper studies the support of young individuals who display criminal behavior by individuals with a criminal background. We focus on experiential peers who provide individual support to clients in the context of a formalized relationship. Additionally, we focus on support in which the relationship is asymmetrical, meaning that there is a designated provider and a designated receiver of support (Davidson et al. 2006), thereby excluding mutual support interventions. By analyzing the relationship between EPs and their clients, we may better comprehend what works in this type of experiential peer support and why. The aim of this paper is therefore to examine the mechanisms of experiential peer support present in the relationship between EPs and clients that may lead to favorable outcomes for clients. We do this from the perspective of EPs, who have had multiple experiences supporting young people with criminal behavior, and are likely able to reflect on past personal and professional experiences with some distance. Our main research questions are: 1) What are mechanisms in the relationship between EPs and their clients? and 2) How do EPs compare their approach and support to that of professional caregivers without lived experiences?

Methods
We conducted a qualitative study in which we interviewed seventeen EPs who provided individual support to adolescents and young adults with delinquent behavior. We describe our methods according to the Consolidated Criteria for Reporting Qualitative Research (COREQ) (Tong, Sainsbury, and Craig 2007).

Main researcher
The interviewer is the first author of this paper (ML), who was 31-32 years old at the time of interviewing. She had prior experience in interviewing adolescents with criminal behavior and no personal experience with the judicial system herself. ML spoke to experts in this field and constructed a theoretical framework regarding experiential peer support (Lenkens et al. 2019b). This led to a positive-critical attitude towards experiential peer support. The interview was for most participants the first time they met the interviewer, after communicating through phone, e-mail or social media.

Participants
We recruited participants through purposeful sampling. Starting from an overview of organizations in The Netherlands known to work with experiential peers (Bierbooms et al. 2017), we contacted forensic mental health care institutions, (juvenile) detention centers, volunteer organizations, rehabilitation centers, and probation officers, asking them whether they worked with EPs. When they did, we received the EPs' contact information and we contacted them directly. In addition, we spread our recruitment message through LinkedIn, the IVO Research Institute newsletter and our personal contacts. We also actively approached experiential peers through e-mail and LinkedIn. EPs were invited to participate if they a) had been involved in criminal behavior, and b) provided one-on-one support to young people (16-30 years old) involved in criminal behavior. We initially intended to focus the study on 16-to 23-year-olds, following the Dutch criminal law for adolescents, but since experiential peer support is not as common in forensic youth care compared to adult care, we decided to expand our range, still encompassing the peak of the age-crime curve and the age at which most individuals desist from crime (Hirschi and Gottfredson 1983).
Our final sample consisted of seventeen experiential peers who were given pseudonyms. All approached EPs who fulfilled the inclusion criteria agreed to participate in the study. The mean age of our sample, consisting of fourteen men and three women, was 37.5 years (sd = 10.37). Twelve EPs were born in the Netherlands; of these, four EPs had at least one parent who was born abroad. All EPs had committed at least one criminal offense and twelve EPs had been incarcerated in a juvenile and/ or adult correctional facility. Offenses mentioned were fraud, drug trade and trafficking, robbery, possession of weapons and theft. Five EPs explicitly mentioned they committed crimes to support their drug addiction. Twelve EPs worked or volunteered for an organization, two had their own (counselling or coaching) business, and three EPs combined both. Although most organizations had close connections to the criminal justice system, only three EPs (Robert, Matthew and Richard) worked at an official criminal justice setting with mandated care, and two EPs (Mark and Melissa) worked at facilities where a proportion of clients received care mandated by court. Two EPs (James and Michael) worked for the same organization in the hospitality sector that provides coaching and employment for youngsters with multiple problems. At least fourteen EPs had provided support to more than one client. The nature of their roles and tasks varied due to the different settings they worked in. Fifteen EPs provided to some extent emotional support to their clients. They had conversations with their clients about how they feel, listened to them and tried to support them in their recovery process. Four of them also used sports or storytelling to facilitate these conversations. Nine EPs helped clients in finding a job and writing application letters. Seven EPs helped their clients with other practical matters, such as suitable housing, debt assistance applications, and transportation. Six EPs had a bridging position between clients and other staff, advocated for their clients or accompanied them to meetings. EPs' background in training and education varied widely (Table 1).

Procedure and setting
We conducted the interviews between February 2019 and July 2019. They took place at EPs' workplaces (n = 11), in a café (n = 4), at the researcher's office (n = 1) and at the EP's house (n = 1). A research intern was present during three interviews. We asked participants to sign for informed consent and they received €20 to compensate for their time. The interviews lasted between 47 and 128 minutes (m = 82, sd = 22) and were audio-recorded and transcribed verbatim.

Data collection
We conducted semi-structured interviews using an interview guide. First, we asked participants about demographic characteristics, their role as an EP, and whether they had received any relevant training. We also asked them about their criminal justice history, such as whether they had been incarcerated and when the last time was that they had been involved in the criminal justice system. We asked participants to describe the support they provided to clients, the extent to which they disclosed their backgrounds and the perceived effects of their support. We also asked them specifically about the seven proposed mechanisms of experiential peer support (empathy and acceptance, social learning, social bonding, social control, narrative and identity formation, hope and perspective, and translation and connection). For instance, regarding social control we asked whether they felt their own experiences play a role in how they approach clients' deviant behavior, and more explicitly asked them whether they are more likely to detect such behavior or say something about it to their clients. We also asked participants to reflect on whether they worked differently from professional caregivers without lived experiences, in general and with reference to the mechanisms. This could refer to (former) colleagues or professional caregivers they had been in contact with as clients. Lastly, we asked participants about the conditions under which experiential peer support works, prerequisites for both clients and experiential peers, and the feedback they had received on their role as EP. After the first four interviews we decided to slightly adjust the interview guide to make it shorter and less cumbersome.

Analysis
We used the adaptive theory approach, which combines the use of extant theories and theory developed from data collection (Layder 1998). In our analysis, we started with codes derived from the theoretical model with seven proposed mechanisms. In addition, our interviews contained many open-ended questions which yielded in-depth answers allowing for new mechanisms or themes.
Coding and analyses were a continuous iterative process in which we examined common themes and connected theoretical concepts to respondents' answers. We used the software program NVivo for coding, analysis and writing memos. We used thematic analysis to identify concepts or themes (e.g. empathy, judgment, role model) related to the relationship between client and experiential peer. Authors ML and LD independently assigned codes to fragments that were relevant for the research. After three interviews, the interviewers studied the fragments of each code to ensure they belonged to the same code. They compared and discussed the codes and developed a preliminary codebook with their definitions, and evaluated overlap in coding for the first eleven interviews. For codes with the least agreement, ML studied the codes more thoroughly, and adjusted, merged or subdivided them where necessary. This led to a final codebook with 90 codes covering sixteen main themes, with definitions supplemented by examples. We then used this codebook to adjust the coding of the first eleven interviews and to code the remaining interviews. Interviews 12 to 16 did not add any new codes, which indicates that saturation was reached. However, since we had only two female participants in our sample, we recruited another female EP. This last interview, coded by ML only as it took place later, did not yield any new codes. For all interviews, ML looked at the themes and subthemes based on both coders' judgment and eliminated those fragments that did not belong to a particular theme. Subsequently, ML looked at the connections between themes, focusing on the mechanisms and EPs' perspectives on whether their approach differed from that of non-EP colleagues. Lastly, GN and FL checked the categorization and interpretation of a part of the results.

Shared experiences
At the foundation of the relationship between experiential peers and their clients lie shared experiences of criminal behavior and its consequences, specifically the experience of imprisonment and the loss of freedom and autonomy. EPs also see similarities with their clients regarding personal circumstances, especially when growing up. Their clients often have to deal with difficult family circumstances, peer pressure and living in a deprived area. EPs also talked about shared struggles in life. When asked about the most important similarities to his clients, Jason responded: 'Hopelessness, sorrow, addiction, the victim role. (. . .) Loneliness, problems with family, low self-esteem, feelings, having problems they have never talked about, debts, etcetera.' All EPs at least partially disclosed their background to their clients and all but one said they were open about their involvement in criminal behavior, although the amount of detail they share with clients varied. They seemed to take conscious decisions about what they disclose and to whom, and for instance do not share details about their criminal history if the client is eager for sensation. The example of James clearly demonstrates this: If you're candid and sincere, others will follow that example. If you see it has a negative impact, you shouldn't do it [disclose]. I don't tell them everything. If I tell them how much money I made . . . I'm not going to tell them. That's not what I want to impart. I want to show them the other side of the story . . . the negativity that it has brought. You need to be very careful with what you tell and show these youngsters.
EPs mainly saw positive effects of their self-disclosure, such as increased disclosure by the client. Other reasons for self-disclosure were to inspire or motivate clients by setting an example or to shock or warn clients.
Shared experiences and mutual identification between EPs and their clients seemed to activate several mechanisms in our sample, which are displayed in Figure 1. First, EPs identify with their clients' personal characteristics and situation, which leads to empathy and a non-judgmental attitude. Second, clients may identify with EPs, perceiving them as credible role models who provide hope. Third, due to EPs lived experiences, they feel like they have insight in what works in the approach or treatment of the target group. They appear to have a more realistic view on desistance, and aim for a humane relationship with the client, characterized by sincerity, equality, reciprocity and trust.

Empathy and a non-judgmental attitude
Experiential peers identify with their clients, leading them to believe that they can understand how clients feel and that they are better able to empathize with them, since they have similar experiences. They, for instance, know what it feels like to be incarcerated. David: The feeling of the empty moment when the cell door closes, the feeling you get every time you hear those guards' boots close to your door . . . Some tap at your door on purpose, or drop the lid three or four times. The feeling that comes with that, you cannot feel that. You cannot understand that.
EPs understand how certain situations and circumstances may lead adolescents to become involved in criminal behavior, what it is like to be seen as a criminal, and that life after prison is difficult. This profound understanding of isolation, pain and rejection is considered different from that of formal care providers who have gained their knowledge through education. In that sense, the frequently used phrase 'I understand what you mean' can only be true for EPs. In addition to experiences of criminal justice involvement, EPs are also familiar with the judgment and stigma associated with these experiences, which seemed to influence their perception of clients. EPs felt they are less judgmental than their colleagues without lived experiences, and that they are more likely to look at the person instead of the criminal act or possible diagnosis. Richard: 'I was very resistant to reading clients files, not because I was not capable of doing so, but because I didn't want to. I always thought it would distort the view of the person in front of me.' EPs also warned for a selffulfilling prophecy; if we label these adolescents as problematic, dangerous or bad people, they may start to live up to these expectations. EPs felt it is important to look at deeper causes of the behavior, since this may help to prevent relapse. According to David, professional care providers focus heavily on the circumstances leading to a specific crime. He felt they are hesitant to touch upon a deeper level because of the emotions that it may evoke. Not all EPs were convinced of structural differences in judgment compared to regular care providers. According to them, someone who is very judgmental of this population would not work in this field. In addition, several EPs indicated that they themselves also have difficulties understanding certain criminal behavior, especially sex offenses.

A credible role model offering hope
In addition to identification of experiential peers with clients, EPs feel that their clients also identify with them, and seem as one of their own. EPs felt they are seen as role models by the clients they support. An important aspect to this position is that the EP's background gives legitimacy to what he or she says. Daniel: 'When you talk to youngsters you notice that they often say "You understand, you come from the streets, so I don't have to explain it to you." And: "What is she doing here? She just graduated, doesn't have any experience, I'm not going to take her seriously, what is she going to do for me?"' This credibility may also contribute to a feeling of hope. As Samantha explained, it matters who tells you that 'everything will be OK'. Said by someone with a privileged position in society, who has had a normal upbringing and drives an expensive car, this message will not be as motivating and inspiring as when it is conveyed by someone similar to you. Witnessing that someone with a troubled past can achieve things can stimulate hope and a belief within clients that they too can accomplish something in life.
This position as a role model may make clients more likely to cooperate or willing to learn from the EP. EPs talked about skills or attitudes they try to teach or stimulate in their clients, including being assertive and asking for help, expressing their emotions, dealing with substance cravings, adding structure to daily life, taking responsibility for oneself, creating a social network and being in time for appointments. EPs try to stimulate self-awareness in their clients, by discussing negative aspects or consequences of their behavior, digging more deeply into the causes of this behavior or by confronting them. In response to clients bragging about their criminal behavior, most EPs show their disapproval or emphasize the negative aspects. According to Matthew, this is more powerful than coming from a regular care provider, since the latter is already assumed to disapprove of such behavior.
When asked about the effects of their endeavors, EPs indicated that clients start acting more responsibly and become more motivated. For some clients, EPs had literally become a role model: these clients aspire to become EPs themselves. EPs also mentioned that some clients had become more confident. However, they also acknowledged that the increase in clients' self-esteem or selfefficacy may be limited. As Steven said, 'the faith that something will succeed, these boys don't have that. They really don't know any better than that everything fails. Things don't come easy to them.'

Realistic view on desistance
In addition to the mutual identification that takes place, EPs' lived experiences also contribute to their approach. Based on their own experiences of desistance as a non-linear and complex process, EPs may have a more realistic understanding of desistance from crime and the difficulty of pursuing a normative lifestyle. They indicated that the process of desistance takes time and that making mistakes is sometimes necessary for growth.
EPs indicated that they can see through their clients' behavior when they lie, manipulate, give socially desirable responses or dodge certain questions. Stephanie: 'I always recognize it when someone is beating around the bush. You just know that from yourself. And I definitely point that out to them. I cannot help them if they're not being honest.' EPs also believed they recognize and interpret certain risk situations more accurately than other professionals. Boris, for instance, mentioned that in the facility he works at, he sees things that his colleagues will not notice, such as clients displaying odd behavior. Chris indicated that he is better able to recognize the type of criminal behavior someone is involved in. Professional care providers may react differently to certain situations, by being more startled or disappointed when there is deterioration or insufficient progress. Steven explained: I can also become disappointed in a client; when I've done a lot and when we're back to square one, when he has committed another crime. But even then, you still don't give up. I know how hard it is to reach a turning point. And that youths keep making mistakes. You just need to take account of that. That threshold is a bit higher for me than for someone else.
Although EPs may support the process of desistance, this process is eventually considered the responsibility or the choice of the client. For clients who are not motivated or ready to change, or who have not yet hit rock bottom, there is little an external person can do. In those situations, EPs felt they cannot initiate or accelerate the desistance process. A few EPs mentioned that they try to plant a seed by making a comment that leaves an impression. Andrew: If they have been in touch with me once . . . I always plant a seed and when it comes out, I don't know, but at least it is planted. And they come back when they're ready for it, it can take a year or a few months.
When asked about the effect of their support to clients, EPs indicated that some clients may have become more motivated to desist and actually quit criminal behavior, but many seem to have a relapse or deliberately choose the criminal path due to limited alternatives. Chris: There are boys with a very low [intellectual] level for whom this whole situation is quite hopeless. It's either the criminal path, hoping they will score and make lots of money, or shut off their brains and go work in some factory for the rest of their lives for a minimum wage. That offers little prospect.

A humane relationship: sincerity, equality, reciprocity and trust
Lastly, experiential peers' experiences seem to contribute largely to the type of relationship they strive for with their clients, characterized by sincerity, equality, reciprocity, and trust. For EPs, this theme represented the main difference compared with regular care providers. First, EPs demonstrate a sincere drive to help these young people, which seems largely related to their own experiences with the formal care system. They go beyond what is expected of them, try to arrange things quickly for their clients and are often available outside working hours. Brian: Officially I work two days a week. But I always pick up my phone, I always call back, whether it's eleven in the evening or the morning, I'm always there. But also simply because I feel, if someone calls you and they need you, come on, I'm not going to tell them "Yeah you can call me back tomorrow, or Thursday, then I'm at work again".
EPs stand up for their clients, do not give up on their clients, and see opportunities for them. They focus on strengths and qualities, and support clients in the development of their talents. Several EPs mentioned that some care providers do not have sincere intentions with the target population, or are not genuinely committed to their job and only do it for the money. Most EPs discussing this did not work at regular care facilities. Therefore, it is possible that this sentiment stems more from previous experiences they had with their own care professionals. Daniel and Chris expressed how they feel that sincere interest in the wellbeing of clients is more important than any lived experiences a care provider may have. Daniel: Genuinely caring is the most important thing. That's something youths can impeccably detect. Do you care about how a boy is doing or what help he wants to receive? If you really care, they will feel it, and then it doesn't matter whether you have a criminal record or whether you come from a safe family. They are just going to see: "Hey, are you real with me and do you want to help me?" And then they will help you, you know, to help them.
Secondly, EPs emphasized the equality between them and their clients, sometimes expressed as being seen as 'one of them'. According to Chris, a sense of equality is necessary because clients who see you as an authority may not open up. Joshua also believed this 'outsider position' of the EP is important to establish a connection with clients: If you lump them all together, the probation officers, the child protectors, the group leaders, the behavioral experts . . . they chose this job and in principle they belong to the system. They are the ones that lock you up at night and open your door in the morning, and lock you back up if you do something wrong, and they [clients] feel that they are all part of the system. I am an outsider to this system. I had a similar past, I come from an outside organization, so I really am an outsider, I think that's the biggest difference.
EPs talked about how they try to change the power relations, by giving clients some control over the situation or sitting next to them instead of opposite each other. The way EPs interact with clients, for instance by greeting them the way youths greet each other, the language they use and their approachability, also contribute to this atmosphere of equality. In addition, the time EPs spend with clients outside of prison often has an informal character, such as eating pizza or working out in the gym. EPs emphasized the perceived unequal status of many professional caregivers. They described them as people who grew up in protective environments and who have had everything served to them on a silver platter.
A third aspect of the relationship, related to equality, is reciprocity, which means that EPs also open up about themselves instead of only asking this from the client. EPs comparing their own reciprocity to regular care providers' approach saw a clear difference. They felt that many professionals do not show the same level of vulnerability towards their clients. As David said, to connect with clients it may help to sit in front of the client as a person, not as a job title. Robert explained why this is important: There are indeed colleagues who share a bit about their home situation or whatever, but that is not even close to what we do (. . .). I am not above someone, that's also not what I convey, but I am for sure not beneath anyone. Equality, that's what my position depends on.
By sharing their own struggles, EPs show clients that they are not the only one, which normalizes the situation and reduces shame. EPs indicated that their self-disclosure leads to more openness on the clients' end, who share more about their personal stories, struggles and emotions, thus becoming more capable of expressing their needs. This is a major step for juveniles who have developed a street mentality in which demonstrating vulnerability means losing face. Daniel, however, pointed out that eliciting disclosure from the client should not be the objective.
I don't know if that should be the approach: if I tell something, I hope that he will tell me something. You share that [personal experiences] just because you want to share because you care, or you want to show: "I have also been through stuff". He needs to decide for himself what he wants to share or not. And that takes time.
The fourth element of the relationship that is highly valued by EPs is (mutual) trust. According to EPs, clients seem to trust them more than other care providers; clients share information with them that they will not share with others. EPs emphasized how important this is; many clients have long histories with formal care in which they have been disappointed and rejected repeatedly, leading to a strong sense of distrust. EPs strive to be trustworthy; they talk to clients about confidentiality and they make sure to keep their promises. Some EPs do not express any disapproval of their clients' behavior, to establish or retain the trusting bond with their client. In the first stage of the relationship, Chris is reluctant to show any disapproval of the delinquent behavior.
If you can talk with them about certain things on the street . . . it can be very practical: the cutting of drugs, the prices, how you can deal drugs in a profitable way . . . that gains respect from clients. I try to create an opening and gain their trust. The moment I have that, I can take on a more advising and brother-like role and say "Hey, that's not okay".
Daniel described how some clients are open about their criminal behavior to him, because they consider him as one of them. This can create a dilemma; if it is too damaging or dangerous, he must do something with the information, but this could also mean a breach of trust.

Discussion
This study investigated the mechanisms in the relationship between EPs and their clients, and how EPs compare their approach and support to that of care providers without lived experiences. At the core of this relationship lie their shared experiences, their mutual identification and EPs' lived experiences that influence the support they provide. This leads to four main mechanisms.
First, EPs recognize the situation their clients are in and the struggles they experience. They feel this makes them more capable to understand and empathize with clients, especially regarding experiences such as being incarcerated, being labelled as 'criminal' and trying to desist from crime. In addition, most EPs believe they are less judgmental than regular care providers. They look at the person and the deeper roots of the behavior, such as the need to belong, instead of focusing on a specific criminal act.
Second, clients may be more likely to identify with EPs than with normal care providers without similar experiences. The EP as someone who has successfully passed the desistance process can provide hope and perspective for one's own future. In addition, a role model with lived experiences is considered more credible than someone with a different (and more privileged) starting point in life.
Third, EPs' own experiences with desistance seem to shape their view on this process. They consider it a non-linear process and are willing to give clients multiple chances. They recognize risk factors and behaviors, but deem the desistance process as belonging to the client, which can only occur if the client is ready and motivated, and which they can support but on which their influence is limited.
Fourth, EPs draw on their own experiences, mainly as former clients, to give shape to the relationship with their own clients. They put a large emphasis on the 'humanity' of the relationships they pursue with their clients. This relationship is characterized by equality, reciprocity, trust and sincerity.
Although presented as separate themes, these mechanisms seem strongly related and appear to interact. Identifying with clients, for instance, not only increases empathy, but also stimulates a humane approach with an emphasis on strengths, as EPs also remember how they were treated as clients. In addition, not only does the recovery view on desistance make EPs less likely to overreact to slip-ups, but this is also influenced by the fact that they aim to build and maintain a trusting relationship with their clients.
Looking at our proposed mechanisms (Lenkens et al. 2019b), we see that these results largely confirm the importance of empathy and acceptance and hope and perspective, although the perceived effect on self-efficacy is limited. Regarding social learning, EPs seem to hold a position as role model, but the skills they try to convey to juveniles seem less directly aimed at non-recidivism and more at creating stability and strengthening protective factors. The most striking difference, regarding the proposed mechanism of social bonding, is the emphasis on the quality of the relationship and not the mere existence of a relationship. Social control seems less prominent; although EPs seem to detect risk behavior, their response to it varies. Narrative and identity formation was not a prominent theme in our data, which may be the case because this process takes place internally and is difficult to observe from an outsider point of view. We did learn that EPs display elements of translation and connection, but rather consider this one of their roles, as elaborated upon in the description of the sample, than a mechanism occurring in the relationship between EPs and clients.
The current study adds to our proposed model (Lenkens et al. 2019b) the emphasis on the quality and humanity of the relationship between EPs and clients, and EPs' realistic view of desistance. The focus on the person instead of his or her behavior reflects a shift also visible in mental health care; from a disease or disability centered model aimed at fixing or healing to a client-centered model in which 'recovery' has an important role (Slade 2010). The way EPs talked about desistance shows a large resemblance to the characteristics of personal recovery, which is described as an active and individual process, of gradual trial and error, aided by a supportive environment but also attainable occur without professional intervention (Leamy et al. 2011). This parallel has been previously drawn by Best, Irving, and Albertson (2016), who point to the importance of changing social networks and identities for both processes (Best, Irving, and Albertson 2016). Several recovery processes play a key role in the relationship between EPs and their clients, such as connectedness, hope and optimism about the future and empowerment (Leamy et al. 2011). To advance the individual's personal power, treatment partnerships are important (Corrigan 2002). Although originally described in a mental health care context, the way these partnerships arise resembles the approach of EPs in our sample. First, care providers' focus should be on endorsing recovery instead of on assuming poor outcomes (Corrigan 2002). This is in line with 'positive criminology', which looks at the individual's whole (including strengths and talents), rather than mainly looking at criminal behavior and risk factors (Ronel and Elisha 2011;Ronel and Segev 2014). EPs in our sample seemed focused on the future instead of past criminal behavior and described their non-EP colleagues as being more negative towards clients when they make mistakes, whereas EPs realize that the process of desistance takes time and that small steps matter. Second, care providers should strive for collaborative decision making instead of a unilateral treatment plan. The latter can be perceived as coercion, which can lead to reactance in the client (Corrigan 2002), while a sense of autonomy is important for motivation and behavioral change (Deci and Ryan 2008). In our sample, EPs collaborate with their clients by aiming for a more equal and reciprocal relationship and by supporting clients' autonomy. For non-EP care providers to become more recovery-or desistance-supportive in their approach to clients, there may be skills or attitudes to be learned from EPs.
Some mechanisms may be more natural to EPs, but they could still be utilized by care providers without such experiences. EPs for instance mentioned a similar 'pain' or 'struggle' as their clients. This suggests that the advantage EPs gain from (disclosing) personal experiences is not exclusive to those with criminal behavior. People with a history of addiction or mental illness for example, could also support individuals with criminal behavior. In addition, the relationships that regular care providers have with their clients may also benefit from care providers being more vulnerable and open about themselves and showing more of their personal side and struggles. EPs in our study indicated that formal care providers also have something to offer, since everyone has personal experiences that can be valuable for this work. Such an approach, however, might be at odds with what they have learned about professional distance and requires careful consideration of content, timing and goal of self-disclosure and the type of client in front of them (Knox and Hill 2003;Peterson 2002;Murphy and Ord 2013).

Strengths and limitations
This qualitative study has contributed to the knowledge regarding experiential peer support for adolescents and young adults with criminal behavior. In addition to the data on previously conceived mechanisms (Lenkens et al. 2019b), the interviews gave insight into how EPs compare their approach to that of non-EPs and the importance of a recovery-oriented mindset. As an additional positive aspect, some experiential peers indicated that through the interview they had become more aware of their work and how they use their own experiences to help other people.
Since we used purposive and convenience sampling and relied on gatekeepers of several organizations to provide us with contact details of the EP, our sample may be selective. Although there was a 100% response rate, it is plausible that we mainly interviewed experiential peers that were satisfied, confident and aware of the work they do. Further research should aim for a more random sample. In addition, future research should consider more perspectives than those of EPs. We need clients' perspectives to learn more about potential effects and mechanisms. In preparation of this article we interviewed five young adults about mechanisms, effects and contextual factors of their relationship with the EP. These interviews suggested agreement with the mechanisms as described by the EPs. More research and a larger sample are necessary to gather more information on how clients perceive this relationship. We also need employers' and co-workers' perspectives to learn more about conditions under which experiential peer support may flourish.
During the interviews, we noticed that some conceptual distinctions were not as clear-cut. The explicit focus on criminal behavior may be unwarranted, as shared experiences of for instance substance abuse seemed more important for some EPs. In addition, many EPs worked with both adults and youths. Although we asked them to keep the latter group in mind, we cannot rule out that some situations they have described may refer to their adult clients.
We asked interviewees about specific mechanisms. Without these prompts, participants may not have mentioned these mechanisms. However, we have no indication that this led participants to be untruthful as their answers were varied and they also sometimes answered that a mechanism was not present. We noticed that the proposed mechanism of 'narrative and identity formation' was difficult to explore during the interviews. Additional research should find new ways of investigating this mechanism, starting by asking clients about their perspective. It is also important to note that, in comparing themselves with non-EP care providers, some EPs seemed to base their opinion partially on their own experiences as clients. These are valid but may be outdated as forensic youth care is an everchanging field.

Implications
Research has not kept pace with the proliferation of experiential peer support, specifically in the forensic field. It is important that studies consider effects of experiential peer support and the conditions that stimulate potential positive effects, such as training or education, supervision and guidance by the organization, and an EP's distance from criminal behavior. Future research should study which personal characteristics and acquired skills are necessary to provide adequate experiential peer support without becoming overburdened. One important risk that should be considered is that some EPs do not denounce their clients' criminal behavior for the sake of establishing a trusting bond. This may give clients the impression that this behavior is acceptable. Another aspect that could be investigated is the timing of experiential peer support, since the effectiveness of such support might depend on the client's stage of desistance (Weijers 2015).
The role of experiential peers needs to be further clarified to know which effects to measure (Davidson 2015). If it is unclear what EPs can offer and how they should do that, it is difficult to determine their added value to the existing care. Furthermore, several aspects need to be addressed to improve organizational support for EPs and collaboration between EPs and their non-EP colleagues. While several EPs had received negative reactions from colleagues, some also seemed to have a one-sided view of care providers as privileged and insincere. For some EPs, their own negative experiences had led to a general distrust of formal care. These mutual prejudices seem to hinder optimal utilization of and interaction between the strengths of both. Non-EPs can learn from an EP's experiences how it feels to be incarcerated and how hard it is to get back on track. For the EP, who often does not have the same level of education, it can be helpful to learn from a non-EP-colleague the necessary professional skills and the theoretical knowledge about criminal behavior and desistance. By working together, EPs may pave the way for clients' receptiveness to more specialized help.
Respectful and appreciative collaboration may also lead to an overall more recovery-oriented perspective within the organization, in which the client perspective becomes more important. Collaboration may thus be able to improve the care for adolescents and young adults with criminal behavior. These youths often come from disadvantaged backgrounds in which recognition and appreciation were lacking. Regaining trust in one person that they can relate to, even if it is minimal, can be a first step away from criminal behavior and towards desistance.