Youth Experiences of and Views on Un-locked Coercive Placement in Residential Child Care

This article explores Norwegian youth experiences of and views on coercive placement in un-locked residential child care institutions. Inspired by Antonovsky’s salutogenic theory, the article discusses factors that make placement an opportunity for development among youth with serious drug and behavioural problems. The empirical material comes from interviews with 34 youth under and after coercive placement. The findings reveal that coercive placement in un-locked institutions can be helpful and necessary, provided that the institutions have the means available to protect the residents and provide supportive and meaningful treatment content. Factors such as treatment structure, the content of everyday life, clear expectations, and boundaries are discussed as important factors that help the placement to be an opportunity for development among youth with serious drug and behavioural problems.


Introduction
Coercive placement in residential child care institutions is a serious intervention that is seen as a last resort to be used when other measures have failed (Backe-Hansen et al., 2011;Kvalø & Köhler-Olsen, 2016). A long-lasting discussion on the appropriateness and effects of residential child care for youth with serious behavioural problems has dominated the research field (Andreassen, 2003;De Swart et al., 2012;Knorth et al., 2008;Whittaker et al., 2016). In Norway, this has led to a decline in placement from 250 per year in 2012 to approximately 150 in 2018 (Statistics Norway, 2020). Youths with serious behavioural problems are usually placed in secure locked residential care institutions, and important questions are being raised about the risks associated with the confinement of minors (Bengtsson, 2012;Vogel, 2018). In Norway, youths with serious behavioural problems are placed in un-locked residential care institutions, unlike in Sweden and Denmark (Bengtsson & Böcker Jacobsen, 2009). This article is based on interviews with 34 Norwegian youths with experiences of coercive placement in un-locked residential child care institutions. The aim of the article is twofold. Firstly, contribute to the knowledge on youth experiences of and views on placements in un-locked institutions. Secondly, to elucidate factors that help the placement to be an opportunity for development among youth with serious drug and behavioural problems.
Coercive placement and the use of restrictions in Norway, as in other countries, have been discussed. Reports show that the restrictions vary between different institutions and questions have been raised about the expediency of the restrictions (Barneombudet, 2015). Ulset and Tjelflot (2012) explored the use of restrictions, control, and physical restraints in child care institutions and found that they can be experienced as both scary and insulting. In this study, youths' experiences with coercive placement will be elucidated from a salutogenic perspective (Antonovsky, 2000). Antonovsky (2000) developed his salutogenic theory on the basis of his theory of general resistance resources, which focuses on the ability of an individual or group to successfully cope with stress (Antonovsky, 2000;Lindström & Eriksson, 2006). The salutogenic perspective relates to health and development in a wide sense, and the factors that cause people to move toward good health are among the key issues.
The concept of a 'sense of coherence' is central to Antonovsky's (2000) theory and includes three components: comprehensibility, manageability, and meaningfulness. Due to its focus on individual capabilities and resources, the perception of coherence can be regarded as a perspective on coping. Comprehensibility refers to the extent to which the external world is perceived as comprehensible. Manageability refers to the perception of being able, as an individual or through accessible networks or other people, to access resources that enable the individual to handle/cope with demands that he or she encounters. Meaningfulness adds a more emotional aspect to the perception of coherence, in focusing on the importance of perceiving one's life situation as comprehensible and manageable. What gives meaning may help involve us emotionally, and linked activities tend to be perceived as challenges and as worthwhile engaging in (Antonovsky, 2000, pp. 35-36). Antonovsky's theory adds to the discussions of residential care concepts that make it possible to analyze the factors that make the placement into an opportunity for development.
In Norway, coercive placement is governed by the Child Welfare Act section 4-24 (Act of 17 July 1992, no. 100). A child who has shown serious behavioural problems defined as 'serious or repeated criminality', 'persistent abuse of intoxicants or drugs', or 'other ways' can be placed in a treatment or training institution for up to 12 months. Decisions are made by the County Welfare Board and must be passed before the youth is 18 years old. The length of the stay differs among the Nordic countries. In Sweden, the average length of placement in secure residential care in 2016 was three to six months (Vinnerljung et al., 2018), while in Denmark it was two months (Henriksen, 2017). In Norway, as in the other Nordic countries, youths from 13 to 17 years are those most frequently placed in residential care institutions (Bengtsson & Böcker Jacobsen, 2009).
Norwegian institutions vary in size, from units for one to three residents to units with space for up to approximately ten residents. The institutions are un-locked and no physical barriers prevent the youths from leaving. Still, the residential staff are authorized to withhold or return a youth if they leave the institution without permission or fail to return at an agreed time. In addition, the Rights Regulation (2011) allows for the restriction of individual autonomy in certain cases, such as urine tests, room searches, confiscation of telephones/computers, and so on. These regulations emphasize the institutions' legal obligation to ensure that residents are provided with appropriate care and treatment. Children and youths' rights to participation, protection, and care are also highlighted in the UN Convention on the Rights of the Child (UNICEF UK, 1989).

Research on Residential Child Care Institutions
Research on residential child care institutions can largely be linked to two traditions. One focuses on the effect of residential treatment and the development of therapeutic methods, while the other focuses on the content and experiences of residential care. Given the complexity of residential care institutions, and service system differences across different national context, we will in the following mostly focus on European and Nordic research. The first tradition is linked to discussions on the effects of treatment methods and the development of therapeutic methods (De Swart et al., 2012;Knorth et al., 2008;Whittaker et al., 2016). In a review of 110 studies, Knorth et al. (2008) found that treatment was described in general terms. Studies of the effects tend to show mixed results, and it is difficult to demonstrate conclusively what factors and methods can predict positive outcomes from residential treatment (Elliott et al., 2005;Hart et al., 2015;James, 2017). Results are also contradictory. One frequently debated factor is time. While some studies claim that treatment programs should be as short as possible to avoid 'social contagion' (Andreassen, 2003), others state that the time spent in treatment can be a positive factor in achieving a good treatment outcome (Huefner et al., 2018;Vinnerljung & Sallnäs, 2008).
When and how the treatment outcomes are measured is another factor that varies. A longitudinal study from Norway showed that children placed in residential care have a higher likelihood of marginalization in terms of education, employment, paid work, and mortality compared to children with no such experience (Backe-Hansen et al., 2014;Clausen & Kristofersen, 2008). Furthermore, comparing different types of intervention is difficult because the youth in care face different challenges and have different treatment needs (Backe-Hansen et al., 2011, 2014Hassel Kristoffersen, Holth, & Ogden, 2011). Research indicates that the same intervention can be positive for some youth but harmful to others (Andreassen, 2015). Hence, studies of treatment effects have not produced clear evidence but rather questions about what works for whom.
Research on residential care for children and youth is frequently referred to as a 'black box' because of the lack of knowledge about the care content (Harder & Knorth, 2015). The second research tradition has highlighted the need for more knowledge, both about what happens during residential care and about the youths' own experiences (Harder & Knorth, 2015;Jakobsen, 2010;Libby et al., 2005;Tysnes, 2014). Some recent studies have contributed to this tradition by discussing the importance of the therapeutic alliance and group climate as key to treatment motivation and outcomes (Duppong Hurley et al., 2017;Roest et al., 2016;Strijbosch et al., 2019). Other studies have focused on how children and youth assign meaning to and perceive everyday life in different types of confinement. Common to these studies is a focus on relational practices and identity formation (Bengtsson, 2012;Franzén, 2015;Jansen, 2011;Reime, 2017;Tysnes, 2014;Vogel, 2018). While Reime (2017Reime ( , 2018 and Tysnes (2014) explored coercive placement in un-locked residential child care institutions, Bengtsson (2012), Franzén (2015) and Vogel (2018) studied experiences of coercive placement in secure institutions.

Methodology
The study is based on a reanalysis of selected data from two major qualitative studies that explored youth experiences of coercive placement in un-locked residential child care. In Tysnes (2014), 17 youths (12 boys and 5 girls) were interviewed in-depth about their experience of placement and leaving care. At the time of their placement, they were 14 to 17.5-years-old (average age: 16.3 years). The youths were recruited with help from the child welfare services and residential care institutions and were interviewed from 2008 to 2010, one to six years after placement. In Reime (2018), 17 youths (10 girls and 7 boys) aged 16-18 were interviewed about their experience of being coercively placed in residential care. The informants were recruited through direct contact with relevant residential care institutions and were interviewed during their placement (2012)(2013). Although the empirical data are a few years old, the relevant laws and practices have not changed.
The reanalysis provided a unique possibility to compound experiences under and after treatment, and to explore common themes across the two data sets. The analysis was rooted in research questions concerning the youths' experiences of and views on the appropriateness of coercive placement for their personal development. These questions were richly discussed in both studies and were covered using the following interview questions: 'How did you experience being coercively placed in residential child care?', 'How did you experience life in residential child care?' and 'What do you think of the fact that youth can be coercively placed in residential child care for up to one year?' (Tysnes, 2014) and the questions 'What is it like to be coercively placed in residential child care?' and 'Can you tell me about a normal day in residential child care?' (Reime, 2018).
We conducted the reanalysis as a theme-centred analysis within an interpretative hermeneutic tradition (Haavind, 2000). In the first round of analysis, we organized the materials based on the relevant questions in the interview guides. In the second round of analysis, we searched for key topics that recurred in both studies. These were narrowed down to two main categories of experiences of coercive placement: 'It worked out' and 'it didn't work out'. The category 'it worked out' refers to stories about positive experiences of coercive placement, while the category 'it didn't work out' refers to stories about negative experiences of coercive placement. We then used Antonovsky's (2000) salutogenic theory as an analytical approach and searched for experiences or the lack of experiences of comprehensibility, manageability, and meaningfulness within the two categories. Antonovsky's theory enabled us to explore factors important to youths' experiences of a 'sense of coherence,' which is set out as an important precondition of development. The results were grouped into two main categories: (a) the content of everyday life and the treatment structure and (b) clear expectations and boundaries. We also analysed a question pertaining to the youths' views on coercive placement using the same procedure as in the analysis of experiences.
The studies' quality and limitations must be considered in relation to their methodological and theoretical frameworks. The two studies build on an explorative methodology with relatively few informants, so findings from the studies should not be perceived as objective or general. However, a reanalysis can contribute to knowledge on youth experiences and supportive factors in a field that has previously been understudied. The theoretical framework has enabled us to explore youths' experiences of coercive placement. Structural conditions such as marginalization, poverty, and exclusion are also important aspects that are relevant in attempting to understand youth experiences of coercive placement, but they are outside the scope of this article.
The research projects complied with the relevant ethical guidelines, and both were approved by the Norwegian Centre for Research Data. The youths received written and oral information about the research project and were told of their rights as participants. All data were transcribed verbatim and de-identified. The informants presented in this article are given fictitious names. In the presentation of the results, quotes from youths who were interviewed during their placement are marked with a 'U' ('under'), while those interviewed after discharge are marked with an 'A' ('after').

Perception of Coercive Placement in Un-locked Residential Child Care
In general, the youths did not wish to be in residential care. The placement meant that they were deprived of the opportunity to make decisions regarding their own lives. In particular, it was hard for the youths if their parents agreed with the child welfare services and supported the coercive placement. During the processing of their case by the County Welfare Board, the youths could not imagine that they were going to be placed in coercive care. Some of them also chose to appeal the board's decision on coercive placement. For example, Lasse (A) said the following: Well, I felt that it was bloody awful then, and that everybody was against me, because I didn't think that I had a problem. I felt misunderstood, and that everybody, the state, the police, the child welfare service and my family really wanted to prevent me from living the way I wanted.
The initial period in residential care was a turbulent and challenging time for the youths themselves as well as for the residential staff. The experience of not being understood and the anger against the child welfare system and their own family recurred in many narratives about the decision to use coercive placement. Many of the youths described an initial stage characterized by resistance and opposition to the residential care institution. This could be expressed through resistance to participate in shared tasks during the day, taking illegal substances inside or outside of the residential care institution, or leaving without permission and staying away for several days. The majority of the participants described how they ran away to take drugs and/or made plans to run away to take drugs. For many of them, the initial period of their stay was characterized by an attraction to drugs, and some even had withdrawal symptoms, such as Ulrik (A) One important finding was that for most of the youths, this resistance and attraction to drugs seemed to abate during their stay, and they gradually took on a more positive view of the situation. Of the 17 youths who were interviewed after completion of their treatment, 14 chose to remain in residential care when the coercive period ended.
In the following, we will focus on the factors that appear to have contributed to a change in the youths' attitudes over time and helped set off an incipient process of change toward personal development. We will elucidate the two main categories of results: (1) the content of everyday life and the treatment structure and (2) clear expectations and boundaries.

The Content of Everyday Life and the Treatment Structure
Maia (U) described how she strongly resisted being in residential care at the start of her stay. When she was interviewed seven months later, she was sensitive to the positive change that had occurred, from the first feelings of desperation to a perception of everyday life as easier and with less of a craving for drugs: At first, I was like, to heck with the residential care, and everything was just lame. But now I have more things to do every day; I've found a job and go to school for two days per week and such. I would never have bothered with that at the time when I entered the residential care. I feel that I don't need drugs so much anymore.
The quote reflects a general tendency in the material, associated with the importance of having a meaningful daily life in the form of school or vocational training, inside the residential care institution or outside.
A general finding in both studies was that the youth value having a clear treatment structure. Youth with experiences of staged systems highlighted the benefits of a fixed structure. In a staged system, the treatment sequence is divided into four or five stages, from admission to discharge and aftercare (Koltveit, 2013;Tysnes, 2014). Many of the youths explained how the staged system motivated them to work on their YOUNG 29 (1) own development, such as Erik (A), who discovered that in order to be permitted to go home for a visit he had to avoid drugs. In the youths' descriptions, the staged models in the residential care institutions appeared to be relatively similar. Each stage focused on part of a treatment pathway, often visualized through treatment objectives, but also the development of individual goals for each youth. The youth's own development helped determine how soon he or she could enter a new stage.
For 33 of the 34 youths included in the study, drug abuse was a contributory cause of their placement. Those youths who felt that they had benefited from the treatment emphasized that the residential care institution had a clear therapeutic focus on addiction, as described by Lasse (A): First and foremost, they helped me understand my addiction problem and work hard on it. I had support around me when I started school and was gradually granted more and more freedom. That helped me quite a lot, actually. Even though I didn't see it as helpful in the beginning, I can see now that it was very positive for me.
The therapeutic focus appears to have helped the youths gradually adopt a more positive view of their placement in residential care. Dina (U) highlighted all she had learned about the effects of drugs as well as about her own mindset, and that the placement changed her views on drugs. In the interview, she talked about her past, which was characterized by risky behaviour and serious crime. Dina described her placement as representing a turning point that contributed to changes in her identity, self-confidence, and interactions with others. Victoria (A), like Diana, emphasized how her coercive placement in residential care helped her develop as a person. She described the change as a result of the work she had done on herself during her placement. For many of the youth in the study, the decision to quit drugs appeared to be a turning point in their perception of the coercive placement. Thus, the results of the study showed that when residential care succeeds in providing a youth with effective help for his or her drug problem, opportunities also open up for the coercive placement in residential care to have a greater impact on the youth's own development project.
Youths who reported having gained little benefit from the placement in residential care emphasized its lack of content and the scant help they received for their drug problem. Carl (A), whose life after the stay in the institution has been characterized by drug abuse, was doubtful about what a residential care institution can expect to achieve with a 16 to 17-year-old youth with a drug problem.

Clear Expectations and Boundaries
Another factor that appeared to be crucial for initializing possibilities for development in residential care is the maintenance of meaningful boundaries and rules. These structures help provide the youth with an understanding and overview of what is expected of them. Ada (A) told us that initially, she had kept to the rules only because she wanted to go and visit her family alone and therefore did not want to be relegated to a lower stage: 'But it changes, though […] You just do as you're told, and then you're rewarded for it […]. But then you find out that some of them are there because they care, and that helps too.' Ada described a process of change that gradually increased her desire to quit drugs. For the youth to perceive the rules as appropriate, it also seems crucial that they perceive the staff as caring about them, as Ada described in the above quote.
A common feature in many of the youths' narratives was that after some time in coercive residential care they gradually started to accept their situation and adapt to the residential care institution's requirements and expectations. Like Ada, Hanne (U) also started a process of change because of her wish to obtain the rewards that came with a promotion to a higher stage.
Clear boundaries appear to foster a sense of security and predictability, but a vital precondition is that all staff members apply the same rules to the youth and that the rules are meaningful. If the youth are met with varying rules, this will cause a sense of insecurity. Ola (A) felt that the residential staff handled the rules and expectations unequally. He felt that this was difficult to relate to and made the situation unpredictable. Others told us about meaningless rules; for example, concerning detailed regulations of daily life in residential care, not being permitted to eat whenever they wanted or wear certain items of clothing was perceived as degrading. Striking a balance between rules that are necessary for the working of the residential care institution on the one hand and the development of each youth on the other appeared to represent a problem for some of the residential care institutions included in the study. Rules that are perceived as meaningless give rise to frustration and discouragement, help sustain a negative perception of one's time in residential care and diminish the motivation to initiate a process of change toward development. Detailed regulation of everyday life may also challenge youths' selfworth and integrity.
Rules and structures for preventing the youth from taking drugs or bringing drugs into the residential care institution were largely accepted by the youth, and many pointed out that this was necessary. Siri (U) told us that strict boundaries are some of the best features of the coercive placement: Being surround by strict boundaries, that you […] have someone there to push you all the time, who can back you up, can talk about your feelings […] om krav til kvalitet og internkontroll and have everybody just sit there and listen only to you. Lots of clever youth and adults here, who see you and can guide you. Yeah, these are some of the best things, I believe.
Siri described a positive relationship between the imposition of boundaries, care, and opportunities for development. The youth linked a strict framework and clear boundaries to the experience of being seen and taken seriously. The absence of rules and structures may leave the impression that nobody cares. Some of the youths recounted such experiences from previous spells in residential care, including coercive placements. One of them was Dina (U), who in a previous residential care placement felt that nobody attempted to stop her from taking drugs or leaving the residential care institution, and she interpreted this as indicating that they did not care about her: 'They didn't care about me there at all. I stayed out all night several times per week. I took drugs while in the building … they didn't even notice.' Knowing that someone sees you and will protect you from harm helps establish predictability and a sense of security during the stay. Dina felt she was taken seriously when the residential care where she was placed stopped her when she wanted to use heroin.

Youth Views on the Coercive Placement
A recurrent finding was that several of the youths who had experienced coercive placement appreciated the opportunity that Norway provides for placing youth in coercive residential care. Many of the youths who were interviewed after discharge from residential care, such as Nora (A), pointed out that they needed help over time and that it took time before they were able to see the extent of their problem and need for help: I think that it's much better than the sort of assembly-line type of treatment, for one to three months, perhaps six months at most. After six months, I had barely started thinking about not taking drugs. That's just the start of the process, you need a lot more time. Especially when you've been at it for a while, you're really stuck in it. So, I'm really quite in favor of it. Of course, you're not, there and then. But now later, I'm really happy about it.
Based on her own experience, Nora was skeptical of residential care institutions that set out to treat youth with drug problems for less than six months. For her, time was an essential factor in starting a positive process of change. Coercive placement also means that the youth are forced to take a break and distance themselves from drugs and the drug scene. Several youths included in the study highlighted this as a positive and important element of coercive placement. Sara (U) described distance from drugs as one of the key therapeutic tools. She used the term 'brainwashed' to illustrate her experience of the effects of drugs on the brain and reported that it takes time before one can start thinking again. Sara's statement can be interpreted as saying that coercive placement may be necessary to provide youth with a sufficiently long break from drugs and to ensure that they have the cognitive and physiological preconditions needed to start working on a process of change. Some of the youths also pointed out another advantage of being under coercive care: they were free from the ambivalence inherent in being able to choose whether or not to stay in residential care on a dayto-day basis.
The youths also reflected on what the alternatives to coercive placement might have been, and many stated that if not for the coercion, they might have died. This is mainly because their life situation prior to the placement was characterized by comprehensive and uncritical use of intoxicants, as described by Stig (A): 'Looking back, I think that it was really a good thing. If I hadn't been placed under coercive care, I actually believe that it could have got so bad that I might have died.' Also, youths who felt that they had benefited little from their coercive placement thought that coercive residential care ought to be an option for youth. Youths with both positive and negative experiences of coercive care pointed out that one precondition for using coercive placement is that the residential care institution must help the youth with their problems. As expressed by Lasse (A) in this excerpt, 'Well, I think it's OK in that sense, because then they will be able to see what they are getting up to. But they need to be placed in an institution which can cope with their problems'. Here, Lasse pointed to a key challenge for institutional child welfare services: the need to ensure the quality of the institutions. Coercive placement may create distance from drugs, but that alone is insufficient to provide appropriate residential care. For the institutions to play a role that goes beyond that of a storage facility, it is essential that their content fulfils the youths' needs and motivates them to undertake a positive process of change. At the same time, the youths pointed out that coercion may not be suitable for everybody, and coercive placement may generate a lot of resistance in some young people.
In this review, we have presented some of the factors that the youths themselves highlighted as crucial for the placement to be an opportunity for personal development.

Coercive Placement as an Opportunity for Development
The findings showed that coercive residential care in un-locked institutions can become a starting point for a positive process of change, provided that the residential care institution can meet the youths' need for secure boundaries and generate a sense of meaning and coherence. The youths' experiences of coercive placement and how their attitudes gradually changed during the stay were discussed in light of Antonovsky's (2000) theory on the 'sense of coherence' and its components of comprehensibility, manageability, and meaningfulness.
At the time of their arrival and during the initial period in residential care, the youths felt little comprehension of why they were placed. They tended to believe that they were in little need of help with their drug problem or criminality. They referred to the placement as a kind of punishment that prevented them from living the life they wanted. However, during their year in coercive care, this changed and several of them gradually realized that they needed help. Some highlighted how the coercive placement freed them from the ambivalence inherent in having to make choices for themselves. The youths pointed out how the coercive placement helped them stay alive. A clear therapeutic focus, predictable boundaries, and a transparent structure appear to be essential. In this study, the youths were generally concerned with quality. A residential care institution must be able to help the youth with their drug issues and assistance in coping with their emotions, as well as other needs for treatment. It is also crucial that residential care institutions prioritize helping the youth cope with daily life; for example, through opportunities for schooling, vocational training, and leisure activities. These factors stand as important for the youth perception of manageability and correspond with matters that are deemed important in Norwegian policy documents (Rights Regulation, 2011;Kvalitetsforskriften, 2008).
Based on the youths' narratives, the youth must perceive the stay in residential care as meaningful. The importance of this perception of meaningfulness is also highlighted in Bengtsson's (2012) study of boys in secure care units in Denmark through their descriptions of boredom. In our study, youths who perceived the coercive placement as meaningful gradually came to see it as part of their development project. On the other hand, those who described the placement as void of content tended to focus more on the negative aspects of the coercive placement and were generally less motivated toward the treatment. The importance of regarding oneself as an actor in a personal process of change was underscored by the findings of Vogel's (2017) study of girls in secure care in Sweden, which accentuated the importance of the youths' perception of autonomy.

YOUNG 29(1)
The coercive placement meant that the youths were disentangled from a life characterized by drug abuse, and to some extent also criminal activity. In this study, distance from the drug scene and time spent in residential care were factors highlighted as conducive to change and development. The findings in this study are thus not consistent with the trend that has characterized the developments in residential care in Norway in recent years; for example, the establishment of MultifunC institutions that provide residential treatment programs with a duration of six months. The trend toward shorter stays in residential care has been justified concerning the treatment effects and reduced risk of 'social contagion' (Andreassen, 2003), while others have raised the question of whether this development in residential care can be interpreted as financially motivated (Huefner et al., 2018). An evaluation of the MultifunC treatment model in Norway failed to demonstrate that it ensures better outcomes than other types of residential care (Fossum, Babaii, & Handegård, 2018). The importance of the time span in framing a space for a positive developmental process found in this study can also be seen in contrast to confinement practices in Sweden and Denmark which are of a shorter length. The youth in this study underscored that it takes time to quit drugs and recognize the need to change one's situation. Some of them reported having spent more than six months in residential care before they were ready to address their developmental process. Thus, this study helped corroborate findings made by other studies documenting that drug rehabilitation is a time-consuming process that requires long periods in therapy to produce a successful outcome (Huefner et al., 2018;Vinnerljung & Sallnäs, 2008).

Coercive Placement as a Necessary Protection
In recent years, there has been a trend in Norway toward significantly fewer coercive placements, and the political aim has been a reduction of residential placements for children in general (Backe-Hansen et al., 2011;Barneombudet, 2015). One finding in the present study was that the youths themselves pointed to the extent of their drug problems and their need to be protected against themselves and their drug habit. In Norway, as well as in Sweden and Finland, youth are coercively placed due to substance use problems (Manninen et al., 2015;Tysnes, 2014;Vinnerljung, 2018). Many of the youths in this study highlighted that the coercive placement may have saved them from dying of an overdose. This risk of death is corroborated by other studies, which have found that youths with behavioural or drug problems also have an increased mortality risk (Backe-Hansen et al., 2014, pp. 170-172;Manninen et al., 2015).
The residential care institutions for coercive placements in Norway are un-locked institutions, in contrast to those in other countries such as Denmark and Sweden (Bengtsson & Böcker Jacobsen, 2009). This study's findings must be interpreted in this particular context and may have implications for how to reason about the use of coercive placement and the content of residential care. Firstly, un-locked institutions can be regarded as representing a higher risk because there are no closed doors or physical barriers to prevent the youth from using drugs or running away. This risk was also documented by stories in this study, particularly relating to the initial stage of placement. Youths described themselves being in opposition to the placement; they mentioned taking illegal substances inside or outside the residential care or leaving without permission and staying away for several days. Closed doors and other physical barriers could possibly have prevented some of these activities and contributed to protecting the youth from potential harm.
In this context, it seems important that residential care can provide a safe and predictable environment involving clear and meaningful boundaries and even the use of physical restraints if necessary to protect the youth from harm. The placements in the study that did not work out underscored this finding and were characterized by stories of residential care that had no clear rules and did not manage to provide meaningful content in everyday life. These youths told us about running away, using drugs, and pursuing other potentially harmful behaviours. The county governors have nevertheless indicated that the use of restrictions should be reduced and highlighted children's right to participate (the County Governors of Aust-Agder and Vest-Agder counties, 2017; the County Governors of Hordaland, Rogaland and Troms counties, 2016). Use of restriction and physical restraints in coercive residential care has thereby been made less relevant as a possible measure. The findings of this study question how to deal with the balance between the rights to participation, protection, and provision when the youth in question has severe drug problems and an increased mortality risk. These equally important principles are laid down in the UN Convention on the Rights of the Child (UNICEF UK, 1989).
Secondly, most of the youth in this study perceived coercive placement as largely positive in their developmental process and as necessary to prevent serious risk and motivate change. It might be relevant to the question if this finding is also related to the fact that coercive placement in Norway takes place in un-locked institutions. The absence of fences and locked doors gradually allows the youth more freedom to take part in activities outside the residential care institution if they adjust to the treatment program and manage without drugs. This stands as an important motivator for a positive change process. Unlike Sweden and Denmark, which generally have shorter spells in secure care and more variation in treatment time (Henriksen, 2017;Vinnerljung et al., 2018), coercive placements in Norway are done for one year. Based on the findings of this study, we consider the length of placement to be an essential factor in promoting a positive development process. Our findings support those of other studies that recommended long-term treatment (Huefner et al., 2018;Vinnerljung & Sallnäs, 2008).

Conclusion
This article presented a reanalysis of two previous studies on Norwegian youths' experiences of and views on coercive placement in un-locked residential child care. Even though the youths included in this study initially did not want to be placed in residential care, several of them highlighted how the coercive placement was helpful and necessary. The youths themselves drew attention to the seriousness of their own drug problems and their need to be shielded and protected from drugs. The findings of this study showed that coercive placement in an un-locked institution can have the potential to reduce youth risk behaviours and even be lifesaving if the residential care institutions have the means to protect the residents. The residential care YOUNG 29 (1) institutions' ability to provide the treatment and time the youth need is a precondition for the perception of the placement as representing the possibility for personal development. If the residential care institutions fail to provide a safe and meaningful environment, un-locked placements can potentially represent a risk for further criminalization and marginalization. To further explore and broaden the perspectives provided in this article, research on the structural factors affecting youths' experiences with un-locked residential child care would be fruitful.
This article can contribute to raising questions on different aspects of coercive placement in residential care. The results from the present study will not be directly transferable to other countries because of the differences in service systems across national contexts; for example, due to un-locked/secure institutions, the length of placement, and other regulations. Nevertheless, the results can stimulate reflections on different ways of organizing coercive care and their impact on youths' experiences and developmental processes. We point to the need for further research, preferably in the form of comparative studies on the role of the length and type of placement in the effectiveness of coercive placements.