Background
Juvenile offenders in residential care are a population marked by highly elevated rates of trauma, psychopathology and other psychosocial problems [
1,
2], while child welfare youth often show delinquent behavior in addition to similarly elevated rates of psychosocial treatment needs [
3,
4]. Furthermore, a substantial number of juveniles involved with the juvenile justice system are so called crossover youth, meaning they have also been involved with child welfare authorities [
5]. There have thus been political discussions to prioritize treatment needs and rehabilitation of juvenile offenders over punishment and not place them in juvenile correction facilities, but in the most appropriate residential treatment setting (e.g., [
6]). This would involve sharing resources and institutions with child welfare youth who have been placed in out-of-home care because of, for example, maltreatment or neglect.
The Swiss juvenile justice system has an explicit focus on rehabilitation, education and treatment of delinquent juveniles [
7]. In general, juvenile delinquency is viewed as a symptom for developmental adjustment problems and juvenile delinquents are viewed as a population in need of protection and guidance more than, and above, punishment alone. As in the adult system, there is a two-pronged approach separating punishment (“
Strafen”) and interventions based on treatment needs (“
Massnahmen”). For example, following a delinquent act, the court can order a juvenile to restitution or another punishment and/or, given the personal and social circumstances that might have contributed to the delinquent act (e.g., ongoing lack of parental supervision, mental health issues, developmental problems), an open-ended foster family or residential placement to address these issues [
7]. This means minors can be placed in child welfare and juvenile justice institutions because of delinquent behavior (juvenile justice measure), child protection reasons (civil law measure, e.g., maltreatment, neglect, or parental absence, psychopathology or drug abuse) or other reasons (e.g., special needs, special education) [
7,
8].
In Switzerland, out-of-home placement of children and juveniles is usually a measure of last resort, after other interventions within the family of origin have failed or a placement is deemed necessary to protect the child’s wellbeing and development [
7,
9]. Children and adolescents are usually placed based on their age, gender and treatment needs and thus, unlike in many other countries, child welfare youth and juvenile delinquents can reside in the same facilities [
8]. Switzerland thus offers an opportunity to study potential effects of shared placement of child welfare and juvenile justice youth. However, despite this ongoing practice, very little is known about minors placed in care in Switzerland in general [
9], and there is no systematic knowledge on any effects of shared placement of juvenile offenders with child welfare youth to date.
The aim of the current paper is to map the demographic, crime-related and psychosocial characteristics of child welfare and juvenile justice youths in residential care in Switzerland, and subsequently examine its relationship with offending behavior in adulthood. This knowledge will not only inform us which factors to emphasize on in the assessment and treatment of these youngsters, but could also help better match the adolescents’ needs with the institution’s treatment options.
Adolescents in out-of-home care in Switzerland
To this day, there are no official statistics on the total number of minors in foster care families or other out-of-home placement facilities in Switzerland, but estimates range from 22,000 to 30,000 children and adolescents [
10]. Child protection is regulated by local authorities. Foster sector case management is less regulated and influenced by local structures, availability of local treatment options, and the individual qualifications of the case workers [
9]. However, since the beginning of this century, steps have been taken to improve reporting, professionalism and quality control. In 2007, for example, a new juvenile criminal code (
Jugendstrafgesetz, JStG) has entered into effect. All adolescents placed in a child welfare or juvenile justice institution through juvenile justice authorities have to be placed in an institution approved by the Swiss Federal Office of Justice (
Bundesamt für Justiz [BJ]). To be approved, the institution have to fulfill certain quality and reporting standards [
11] which, under new juvenile criminal law, includes regular assessments to document ongoing appropriateness of the placement [
7]. In 2013, around 200 newly regionally consolidated professionalized and interdisciplinary Authorities for Child and Adult Protection (
Kindes- und Erwachsenschutzbehörden [KESB]) have replaced the 1420 lay authorities which were organized on a municipal level [
12]. At the same time, new federal legal regulations for foster placements (
Pflegekindverordnung [PAVO]) have taken effect and the BJ has begun to collect and share best practices, general information and statistical analyses on youth placed in institutions or foster families on a new online platform (
www.casadata.ch). The Swiss Federal Office of Statistics (
Bundesamt für Statistik [BFS]) yearly publishes the number, age and gender of sentenced minors (e.g., [
13]. Similarly, the National Conference for Child and Adult Protection (
Konferenz für Kindes- und Erwachsenenschutz [KOKES]) now publish yearly numbers of child protection articles in court rulings (e.g., termination of parental rights or removal of custody), giving an idea of the number of new residential placements based on civil law. However, to date no information on the exact nature of the civil law intervention or any information on the children it aims to protect has been released.
Mental health problems of youth in the child welfare system
A recent meta-analysis of studies from the U.S. and Europe showed that among children and adolescents in the child welfare system, 49% met criteria for a current mental disorder [
3]. More in detail, 27% were diagnosed with a conduct (CD) or oppositional defiant disorder (ODD) and 11% met the criteria for an attention-deficit/hyperactivity disorder (ADHD). The population also had high rates of internalizing problems, with prevalence estimates for anxiety at 18%, depressive disorders at 11% and posttraumatic stress disorder (PTSD) at 4%, with higher prevalence of internalizing disorders among girls and more externalizing disorders among boys. In comparison, worldwide-pooled prevalence of mental health disorders within adolescents in the general population is estimated at 13%, with anxiety disorder at 7%, any disruptive disorder (i.e., CD or ODD) at 6%, ADHD at 3%, and any depressive disorder at 3% across geographic location [
14].
These numbers have been even higher among youth placed in residential care, ranging from 49 to 76% [
15‐
17], with high rates of comorbidity in all studies. Additionally, these adolescents show elevated rates of chronic illness [
18], and childhood trauma [
19], which in turn have been associated with worse mental health status into adulthood [
20,
21]. For example, in a study with a Norwegian sample of youth in residential placement, the 71% of adolescents who had experienced maltreatment were even more likely to show CD, general anxiety disorder, dysthymia and major depressive disorder as well as more attempted suicides [
22]. In addition, substance use as well as depression have been associated with increased rates of juvenile delinquency in youth in the child welfare system [
4] and there is a substantial body of research linking past maltreatment experiences with delinquent behavior in adolescence [
23,
24]; within the child welfare system, delinquency rates for youth with a history of maltreatment are approximately 47% greater than their non-maltreated counterparts [
25]. In one recent U.S. study, for example, a history of maltreatment increased the risk of arrest by 55% and of committing a violent crime by 96% [
26]. Finally, it is estimated that more than a third of youth in child welfare are known to the juvenile justice system [
5,
24].
The limited data on Swiss samples of youth in out-of-home placements have not differentiated between child welfare and juvenile justice youth but have shown similar high rates of mental health treatment needs and high comorbidity, with overall 74% of children and adolescents fulfilling criteria for one, and 60% fulfilling criteria for more than one mental disorder in residential care populations [
27]. Furthermore, 25% of them suffered from complex psychiatric disorders with emotional and behavioral symptoms and elevated rates of delinquency.
Mental health problems of youth in the juvenile justice system
Similar to their counterparts in the child welfare system, youth in the juvenile justice system often come from backgrounds of poverty, family dysfunction, and maltreatment [
28,
29]. Between 70 and 95% of detained youth were found to have mental health problems [
1,
30,
31]. A meta-analysis of youth in juvenile detention and correctional facilities has shown that they are about 10 times more likely to suffer from psychosis than the general adolescent population [
1], and there is a high prevalence of previous trauma and PTSD [
32,
33]. Rates of substance use are extremely high, with dependence and abuse affecting between 40 and 70% of juvenile offenders in custody [
31]. A systematic review among detained male adolescents found mean prevalence estimates of 70% for any mental disorders, with CD and substance use disorders (SUDs) being the most frequent [
30]. Although their numbers are much smaller, girls in the juvenile justice system tend to be younger and have more severe mental health problems than their male counterparts [
1,
34]. Fazel and colleagues [
1] found that almost 30% of girls in detention qualified for a diagnosis of major depression compared to 11% of boys, almost 20% presented with ADHD, (12% among boys). Both girls and boys shared similar elevated rates of CD at 53% of the sample, but girls present with higher rates of comorbid externalizing and internalizing disorders. These findings are troubling, since meta-analytic results have shown that presenting with an externalizing disorder or comorbidity increases the risk of recidivism for juvenile delinquents by around 20% [
35].
Risk of adult criminal behavior
A substantial body of research links past maltreatment and neglect with juvenile delinquency and (violent) offending in adulthood [
36]. Simultaneously, studies show an elevated risk of adult criminal involvement among former foster care youth [
37,
38]. To date, the evidence regarding the influence of out-of-home placement on future delinquency is inconclusive [
39,
40]. For example, it has been shown that any out-of-home placement increases the likelihood of delinquency in adolescence and into young adulthood, especially if there is a history of placement instability [
25,
41,
42]. Type of placement matters as well, and for some adolescents, kinship or family foster care has been associated with better outcomes compared to residential placement in group homes [
36]. However, all those findings are confounded by the fact that placement outside the home is usually reserved for the most severe cases of detected maltreatment, while more severe mental health, substance use and conduct problems are associated with both residential placement over family foster care and more placement instability, all of which independently influence likelihood of future delinquency [
24,
38,
43‐
45]. Furthermore, a Swedish population-based study only found negative effects of placement in care on adult criminality for boys first placed between ages 13–18, but not for girls or younger boys [
40].
Recent national data showed that 8% of adolescents born in Switzerland in 1992 were convicted as adolescents, and that having a juvenile conviction was associated with a six-fold increased odds of an adult conviction [
13]. Among convicted Swiss adolescents, 26% were re-convicted as young adults. Additional analyses showed that being male, having multiple juvenile convictions, being over 16 years of age at first conviction as well as having been convicted for more severe crimes increased the risk of adult criminal conviction. However, the study did not differentiate between type of adult criminal conviction (violent or non-violent) and there is no knowledge on risk of adult criminal conviction among young adults who were in residential care as adolescents.
Based on the aforementioned information, the aim of the current study threefold. First, to examine similarities and differences in demographic markers, previous offending (self-report and official conviction) and treatment needs (psychiatric profile, substance use) between adolescents placed in residential care by either child protection or juvenile justice authority. Second, to investigate whether the adolescents committed by child protection or juvenile justice authority differ in their long-term risk for any, violent, and non-violent young adulthood criminal convictions. Third, to examine if this relationship persists after controlling for well-known risk factors for adult criminal conviction (gender, age at beginning of placement, trauma, past self-reported delinquency, past convictions) or mental health treatment needs in adolescent residential care.
Discussion
Adolescents in residential care are marked by multiple disadvantages before and during placement, as well as consequently in young adulthood. They show elevated rates of trauma, psychopathology and other psychosocial problems, and an elevated risk of involvement in both juvenile delinquency and adult criminal behavior [
1‐
4]. At least one third of youth in child welfare are also known to the juvenile justice system [
5,
24].
In Switzerland, adolescents are placed in residential care because of delinquent behavior (juvenile justice measure) or for child protection reasons (civil law measure, e.g., maltreatment, neglect, or parental absence, psychopathology or drug abuse), meaning they might reside in the same institutions based on their educational or treatment needs [
7,
8]. The current study capitalized on this opportunity to examine effects of shared placement of juvenile offenders with child welfare youth in Switzerland and investigated long-term adjustment in the form of adult criminal conviction. Similarities and differences in demographic markers, current treatment needs (trauma, psychiatric profile, substance use) and previous offending behavior (self-report and official conviction) between both groups were examined, and it was investigated if these demographic and crime-related risk factors, and mental health treatment needs while in residential care influenced risk for adult criminal conviction. Especially investigating the influence of mental health treatment needs on risk of adult offending is of high practical relevance, since it might present an important avenue for intervention.
Results of the current study showed overall few differences in mental health treatment needs between child welfare and juvenile justice youth, and no association between placement authority and risk of adult criminal conviction after accounting for other risk factors and mental health treatment needs. Univariate analyses of group differences showed that, while juvenile justice youth had higher levels of substance use, there were no differences in past traumatic experiences, angry-irritable or depressed-anxious problems on the MAYSI-2. As expected, both groups differed in previous offending behavior as well as in previous convictions, with juvenile justice youth scoring higher on all indicators. It is important to note however, that even among juvenile justice youth, 22.5% had no previous conviction. This indicates that juvenile justice authorities do mandate placements based on educational or treatment needs independent of substantiated delinquent behavior, as intended by Swiss law. The lack of differences in mental health problems or planned duration of placements between the two groups is also an indicator that placement decisions are based mainly on treatment needs and not as a means to discipline juvenile justice youth.
In terms of demographic factors, we found age, gender and nationality differences, as well as some regional differences. While the increased proportion of males among the juvenile justice youth corresponds to international samples of juvenile offenders [
1,
34], the age differences and differences in nationality merit some closer attention. While groups did not differ in their number of placements or age at first conviction, juvenile justice youth were older at their first placement as well as at the time of the study. Furthermore, although the results regarding nationality are difficult to interpret,
1 juveniles with a non-Swiss nationality were more prevalent in the juvenile justice than in the child welfare sample. This could indicate that older adolescents and adolescents with a non-Swiss nationality represent a subgroup where the consequences of dysfunction at home or treatment needs appear later compared to welfare youth whose needs seems to get noticed by authorities earlier. Whether this is associated with the severity of the situation or represents a different reaction of the adolescents to similar scenarios has to be the focus of future research.
Furthermore, the results of the logistic regressions showed several associations between risk factors and adult criminal conviction, as well as differing associations by type of adult criminal conviction. As in the Swiss national data [
13], gender was the strongest predictor of adult criminal conviction, with males showing between 4.6 to 7.1 times increased odds for an adult conviction. Same as in the national sample, in the present high-risk population previous juvenile convictions were associated with an increased likelihood of general, violent and non-violent conviction. However, these findings were not repeated in self-reports of juvenile delinquency, with self-reported severity of previous delinquency having no association with any of the outcomes when other factors were taken into account. While our results showed that placement authority did not influence likelihood of adult conviction after accounting for other risk factors, this difference indicates that there might be labeling processes by contact with the juvenile justice system that results in legal convictions. However, they do not seem to influence reasons for institutionalization. Further investigation of mechanisms behind these results should be the focus of future research.
Lastly, our results showed an association between traumatic experiences and non-violent adult conviction up to ten years later, as well as an association between alcohol and drug use and general adult conviction. The latter is consistent with the results of previous studies in which substance use problems were found to be related to (adult) (re)offending [
57‐
60]. Despite a small effect size, it could still be important to screen for alcohol and drug problems in adolescents upon entering the residential care institution, so these problems can be taken into account in treatment in order to prevent long-term negative outcomes, such as delinquency in adulthood [
59,
61,
62]. Regarding trauma, though the effect sizes were also comparatively small, our findings correspond to the body of research showing a connection between childhood trauma, delinquency and adult criminal involvement [
63‐
65], whilst noting that this does not apply to all forms of childhood traumatic experiences [
66]. Hence traumatic experiences and psychosocial stress should also be included into standard screening and assessment, and taken into account in the treatment of juvenile and adult offenders [
66]. Evidence-based trauma-therapeutic interventions as well as trauma-pedagogic care concepts should be embedded into child welfare and juvenile justice settings. Trauma-informed care is a conceptual framework and milieu therapeutic approach that relates to the understanding of and responsiveness to trauma exposure [
67]. It conceptualizes problem behaviour in the context of an individual’s traumatic exposure and contains anticipating and avoiding practices which increase the risk of traumatic re-enactment [
68,
69]. Guiding principles of trauma informed care include: safety, trustworthiness and transparency, peer support, collaboration and mutuality, empowerment and choice, and cultural, historical and gender issues (see also the infographic on the website of the Office of Public Health Preparedness and Response [OPHPR] of the Center for Disease and Control Prevention [CDC]:
https://www.cdc.gov/cpr/infographics/6_principles_trauma_info.htm).
2 This trauma informed approach has also been examined in the more specific context of the juvenile justice system [
70,
71]. Although the combination of evidence-based trauma-therapeutic interventions in combination with trauma informed care concepts is highly promising for the treatment of, for example juvenile offenders (but also adolescents in the child welfare system), more research is warranted to examine its impact on offending behavior/recidivism as well as other adolescent/adult functional outcomes.
Limitations
The current study must be seen in the light of several limitations. A first set of limitations relate to the research design of the larger MAZ. study from which this sample was drawn [
8]. First, the classification child welfare versus juvenile justice youth was based on the placement ground in the institution at baseline assessment of the study. However, research in the field of crossover youth has shown that a percentage of adolescents appear in both systems during their childhood/young adulthood [
5]. Second, in our sample, we know that a part of the adolescents was in out-of-home care before and could still be found in both systems after the study. Unfortunately, we were unable to carry out a comprehensive, accurate residential care trajectory analysis, on the one hand because the adolescents are not always fully aware of their history and on the other hand because this information is not collected in a structured manner by a centralized organization in Switzerland. Finally, by design, participants were interviewed at varying time points after the beginning of their institutional stay. The MAYSI-2 however is designed to be administered at intake into a juvenile justice facility. Given the time limited nature of the anchoring questions in this screening measure, it can therefore not be excluded that the results have been influenced by the varying time spent already in an institution. We tried to offset this limitation by controlling for time since intake in our analyses.
A second set of limitations concerns the assessment used in the current study. An important point is that many of the tools we used in this study were self-report instruments (MAYSI-2, self-reported delinquency). The use of self-report instruments entails a risk of both overestimation and underestimation. On the other hand, it offers the opportunity to gain more insight into certain aspects (often relating to internalizing mental health) that may have been overlooked when using only third-party assessments. Notably, we used official registered criminal convictions for the outcome variables. However, future research and analyses should include information from multiple sources. Finally, trauma is a broad and multi-faceted concept with often no clear definition leading to an exponential use. We used the traumatic experience scale of the MAYSI-2, which is a very rudimentary screening scale only consisting of a limited number of items. This approach takes little account of the number, duration or effect of a certain (potentially) traumatic experience and is supposed to be a quick screening tool that needs further enhanced clarification and more sophisticated measurement tolls. Nevertheless, it is a short and feasible indicator for possible trauma exposure.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.