How Do Forensic Clinicians Decide? A Delphi Approach to Identify Domains Commonly Used in Forensic Juvenile Treatment Planning

Abstract

Adolescents in forensic care display wide varieties of complex psychiatric, psychological, and contextual problems. Based on large amounts of information, clinicians need to consider, integrate, and prioritize these offender characteristics to construct individual treatment trajectories. So far, scientific knowledge on how these treatment decisions take place is scarce. Current study uses a qualitative design for detecting implicit knowledge of clinicians on which offender characteristics they consider while making treatment decisions. Using the Delphi method, 34 experienced clinicians working in forensic care facilities in the Netherlands were asked about the most important domains of offender characteristics that influence their treatment planning. Eight domains were identified as being crucial in treatment planning: Mental health problems, Personal characteristics, Family, Offense, Motivation, Treatment, School/Work/Housing, and Peers/Spare time. Based on current results, focus on a broad spectrum of individual and contextual characteristics is recommended. Moreover, protective factors and comorbid problems on multiple domains should be considered.

Keywords

Delphi technique decision support treatment planning juvenile delinquency

Introduction

Knowledge about the complexity of (psychiatric) problems of juveniles admitted in forensic care facilities and best practice treatment interventions have grown substantially over the past decades (e.g., Colins et al., 2010; Fazel, Doll, & Långström, 2008; Lipsey, 2009). Forensic clinicians, however, are still challenged in their process of decision making concerning treatment planning for these adolescents (Kazdin, 2008). During the process from diagnostics toward treatment allocation, clinicians need to consider the interaction, interdependence, malleability, and urgency of problems in the individual adolescent and his context (e.g., Barnao & Ward, 2015; Kazdin, 2008). Clinicians, thereby, need to integrate knowledge about various factors, as the problems of these adolescents are not limited to one area. This can be seen in the variety of risk factors that predict recidivism in assessment instruments like the Structured Assessment of Violence Risk in Youth (SAVRY; Borum, Bartel, & Forth, 2002) or the Level of Service Inventory–Revised (LSI-R; Andrews, Bonta, & Wormith, 2000). Furthermore, to guide adolescents toward a positive and meaningful future, treatment should also include focus on protective factors (Ward, Yates, & Willis, 2011).

This intricate process of decision making has not yet been explicitly described and documented (Kazdin, 2008). Research on clinical decision making is mainly focused on the medical field (e.g., Frost, Cook, Heyland, & Fowler, 2011; van Hagen et al., 2011), the development of support systems (Lutz et al., 2006; Spreen, Timmerman, Ter Horst, & Schuringa, 2010), or on the use of existing risk assessment instruments (Maloney & Miller, 2014). Clinicians in forensic settings, however, often make treatment decisions based on implicit knowledge they have collected during their training and practical experience (Stewart & Chambless, 2007). This makes the individual clinician in the institution unintentionally an expert in decision making. The use of case formulation in forensic practice makes use of this expertise and helps to organize and integrate information about a person, point toward treatment, and anticipate outcome (Sim, Gwee, & Bateman, 2005), which is especially suitable for complex patients (see Hart, Sturmey, Logan, & McMurran, 2011). This case formulation, however, is a methodology that is still in its infancy (Davies, Black, Bentley, & Nagi, 2013; Hart et al., 2011; Kazdin, 2008). Therefore, to structure and improve the process of decision making in the field of adolescent forensic treatment, the first step would be to convert the implicit clinical knowledge from experts to more explicit and mutual knowledge. This communal knowledge is likely to be used, shared, and evaluated by other clinicians, and can improve future decision making processes and quality of care.

The joint effect of multiple factors has to be considered during the process of decision making, of which a large number is supported by scientific literature. Recent studies have shown a high prevalence (40% to 74%) of psychiatric problems in juvenile offender populations (Colins et al., 2010; Fazel et al., 2008; Teplin, Abram, McClelland, Dulcan, & Mericle, 2002; Vermeiren, Jespers, & Moffitt, 2006). Moreover, comorbidity rates of psychiatric disorders (the presence of two or more psychiatric disorders) are extremely high (50%) in incarcerated adolescents (Colins et al., 2010; Teplin et al., 2002; Vermeiren, 2003; Vreugdenhil, Doreleijers, Vermeiren, Wouters, & van den Brink, 2004). These prevalence rates have been linked to future delinquent behavior (Hoeve, McReynolds, & Wasserman, 2013; McReynolds, Schwalbe, & Wasserman, 2010) and therefore seem clinically relevant treatment indicators. Studies on the psychosocial functioning of juveniles provide additional useful information on the development of juvenile delinquent behavior and offer similarly starting points for treatment. For example, strong links have been demonstrated between juvenile delinquency and difficulties with peers (e.g., Chung & Steinberg, 2006; Haynie & Osgood, 2005), educational problems (e.g., Chitsabesan et al., 2006), and family problems (e.g., Farrington, Coid, & Murray, 2009; Hoeve et al., 2009). Other important treatment indicators within the treatment of juvenile delinquents focus on the responsivity of the adolescent (Andrews & Bonta, 2003; Andrews, Bonta, & Wormith, 2006). According to the risk–needs–responsivity model (RNR; Andrews & Bonta, 2003), elements such as level of motivation and learning style are necessary to consider to adjust treatment to the individual to promote the changes for positive treatment outcomes.

This knowledge cannot easily be translated into practical implications in the institutions and leaves much space for interpretation by the clinician. The clinician, who works day to day with the adolescent, is confronted with a wide variety of individual and contextual problems as well as strengths of the adolescent (Kazdin, 2008). These are not all underpinned in instruments for risk assessment or evidence-based treatment manuals and theories, such as the SAVRY (Borum et al., 2002). Despite extended studies on risk factors in relation to recidivism (e.g., Borum et al., 2002; de Vries Robbé, de Vogel, & Douglas, 2013; Olver, Stockdale, & Wormith, 2011) and efforts to develop evidence-based treatment, the bridge from the applicability of research findings toward treatment allocation is still lacking (Kazdin, 2008). Four important reasons illustrate the difficulties clinicians face during the process of decision making.

The first is called the “scientist–practitioner gap” and points out that results of studies are based on populations and provide groupwise differences (Kazdin, 2008; Lutz et al., 2006). Therefore, findings cannot automatically be applied in clinical practice for individual adolescents. As a result, treatment decisions are mainly based on previous clinical experience above scientific information (Stewart & Chambless, 2007).

Second, although specific problems such as aggression, conduct disorder, and family problems have been extensively studied separately, the influence of their co-occurrence in treatment progress has been investigated poorly. The problems of these adolescents, however, are mostly a mix of factors in the psychological, contextual, and social domain. Clinicians are forced to gather and combine the available information for the particular adolescent in their treatment facility, whereas the influence of comorbid problems on treatment progress has not been studied. Research does not yet provide the information to clinicians to overcome the challenge to look at the combination of factors that influence the individual (Kazdin, 2008).

Third, the majority of evidence-based interventions implemented in juvenile justice institutions (JJIs) target predominantly externalizing behavioral problems (e.g., Andrews, Zinger, & Hoge, Bonta, Gendreau, & Cullen, 1990; Garland, Hawley, Brookman-Frazee, & Hurlburt, 2008; Lipsey, 2009). These externalizing problems are more visible, more easy to identify, and thus easy to transform into treatment goals. It has been stated that ignoring less obvious mental health problems, like anxiety or mood disorders, leaves these problems undetected and untreated (Espelage et al., 2003). This might increase the possible risk of future recidivism as research has found a relation between mental health problems and recidivism (Hoeve et al., 2014; McReynolds et al., 2010).

Fourth, the majority of interventions focus on risk factors. Effective interventions should include protective factors and strengths of the adolescent as well. This is described in the good lives model (GLM; Ward & Brown, 2004). Key point of this strength-based rehabilitation theory is the promotion of important personal goals of the adolescent toward a good or better life. This has a positive effect on the reduction and management of future offending behavior (Ward, Mann, & Gannon, 2007). The GLM is helpful in transforming theoretical elements into daily practice but does not provide clear instructions for the clinician as to what to do in a particular case of a juvenile offender. This makes that the GLM model is not yet implemented in the clinical practice of the JJIs, although useful treatment guidelines have been developed for specific subgroups of (juvenile) offenders, such as sexual offenders (Wylie & Griffin, 2013) and offenders with mental health problems (Barnao, 2013).

With the considerations described above and the challenges clinicians face during the process of treatment planning, it is not surprising that therapy indication and experienced improvement rates for the same patient differ, depending on therapists’ background, training, and experience (Eells & Lombart, 2003; Okiishi et al., 2006; Witteman & Kunst, 1997). More knowledge about this implicit decision making process is needed to build support models and improve clinical decision making, so clinicians can learn from the knowledge and expertise of their colleagues, especially because there is already much information about risk factors but less on how to influence the complex forensic juveniles. As the expertise about decision making in treatment for juvenile delinquents can be found within the field itself and the fields of clinical practice and research should contribute more to each other’s field (Kazdin, 2008), a systematic study on the actual process of decision making of clinicians is the first step to build more explicit and communal knowledge for treatment planning.

Conceptual issues in health care services, like decision making processes, are best investigated by multiple research methods using the knowledge from researchers as well as clinicians (Tashakkori & Teddlie, 1998). Recent studies in health care services have demonstrated the importance of qualitative techniques where practice models are developed from and matched to daily clinical practice and the collaboration of practitioners and researchers (Fiander & Burns, 2000). The Delphi technique is an example of a qualitative study method for a “reliable and creative exploration of ideas or the production of suitable information for decision making” (Adler & Ziglio, 1996, p. 3). Other studies in (forensic) mental health searching for consensus on difficult topics have successfully used the Delphi technique to work toward conformity within a panel of experts (e.g., Sharkey & Sharples, 2001; To et al., 2014).

Therefore, the current study used the Delphi technique to systematically investigate the implicit knowledge used by clinicians in the process of decision making to find domains that are important to consider during treatment planning in the adolescent forensic field. First, using a qualitative design (Delphi), domains clinicians consider important for allocating treatment will be investigated. Next, after having analyzed the domains, there will be searched for consensus on the most important domains one has to take into account before the start of treatment planning. We aim to find a series of well-defined domains that are found important and are suitable to transfer to other clinicians working within the adolescent forensic population to support the process of treatment planning. This is to support clinicians during the process of decision making as an addition to their own knowledge, expertise, and use of requested assessment instruments.

Method

Design

The present study is a qualitative study among clinicians in the forensic field, performed using the Delphi technique, because of its specific characteristics that suited the rationale of our study. This structured research method has been developed around 1940 to gain consensus within a group of experts and has been the object of different social assignments for complex problems for which current science does not have the answers yet (Jenkins & Smith, 1994; Keeny, Hasson, & McKenna, 2011). It applies structured communication by using a series of questionnaire rounds to achieve consensus within a panel of experts and has been used in many academic domains (e.g., Rowe & Wright, 2011). Experts independently answer questions in different rounds, which are collected and analyzed by researchers. Each round of questions builds on the results of the previous one and the experts are challenged to provide their input from another perspective each time. This iteration allows experts to adapt or sharpen their opinions in subsequent rounds. In this way, a common vision and consensus is developed between the different experts (Keeny et al., 2011). The average expert panel for the Delphi technique includes approximately 20 experts with a maximum of 50 experts. When a panel includes too little or too many experts, drawbacks as an unrepresentative pool of experts or large dropout may influence the findings (see Hsu & Sandford, 2007). Over the years, many forms and interpretations of the technique have been developed (e.g., Sharkey & Sharples, 2001); therefore, current study makes use of the guidelines described by Keeny and colleagues (2011) in their book The Delphi Technique in Nursing and Health Research. The Delphi technique has been described as a good research method (Keeny et al., 2011); however, there is a debate about the reliability and validity of the technique. It can be discussed whether two different panels would produce the same results and a low response rate would influence the validity of the answers (In Sharkey & Sharples, 2001). Unfortunately, there is no guideline yet on how the Delphi technique should be performed or set a minimum response rate (e.g., Keeny et al., 2011; Walker & Selfe, 1996). In the present study, we aimed for a response rate of 80%.

The current study started with an open first question and is therefore known as the Classical Delphi (Keeny et al., 2011). By using the first open question, implicit knowledge was retrieved from the experts to gain practice-based information instead of their theoretical knowledge. The second and third question rounds were used to gain consensus in the answers of the experts from the first round, through categorizing and prioritizing the earlier answers. A final focus group was organized to make the input from the previous Delphi round definite. It has been suggested that a Classical Delphi consists of four rounds, but more recently, two or three rounds have been preferred because participant fatigue and the stability of consensus may shift after two or three rounds (In Sharkey & Sharples, 2001).

Participants

In the Netherlands, there are currently seven JJIs, which are either private or public, all controlled by the Dutch Ministry of Safety and Justice. At the time this study was performed, nine JJIs were operational. These centers are treatment facilities for adolescents between the age of 12 and, approximately, 23 years who are incarcerated because of their delinquent behavior. The duration of stay varies from 1 day to 6 years, depending on the placement order. This can be pending judicial verdict, detention, or a mandatory treatment order for the duration between 2 and 6 years.

The board of directors of every JJI in the Netherlands as well as collaborating mental health institutions and the Dutch Institute for Forensic Psychiatry and Psychology have been asked to provide at least two clinicians to participate in the Delphi study. In this request, the importance of an experienced, but also diverse expert panel was addressed. Suggestions were made to ask the clinicians with the longest and shortest clinical career. The directors either directly asked a clinician to participate in this study or asked their staff who would volunteer to participate in this study. As these directors were aware of the importance of the study, it was assumed the participating clinicians could be considered as experts.

Initially, 36 experts were sent the questionnaires. Two of them did not react in the first round, and were excluded for further analysis in this study. As a result, a total of 34 experts (male = 13, female = 21) participated in the analysis of current Delphi study. Most participants were between the age of 40 and 44 years (10 categories ranged between 25 and 29 to older than 65 years), and the mean years of working experience was 12.5 years (range = 2-39, SD = 10.3). The group of participants consisted of (mental health) psychologists (n = 15), psychotherapists (n = 8), child and adolescent psychiatrists (n = 6), and other therapists (n = 5). All JJIs were represented as well as some mental health services with forensic expertise (9%) and the Dutch Institute of Forensic Psychiatry and Psychology (9%).

Procedure

All experts have been visited by the main researcher. During this visit, the principles of the current study and the purpose of their contribution were discussed. Experts were explicitly asked for their participation for the whole Delphi study, regardless of changes in job descriptions or work setting. At the end of this visit, the first round of questions as well as an informed consent and background form were delivered to the participants on paper and by email. All subsequent questionnaires were posted and emailed to the participants to be sure they got the information in time and in a way convenient to them. After every question round, the participants received a small present thanking them for their effort. The Delphi study was finished with an expert meeting for all participating clinicians to finalize the results. The response rate of the first round was 86%, second round 79%, and the third round 59%. Twelve participants attended the expert meeting.

Delphi Study

Questionnaires and other materials used for the Delphi study were developed by researchers and participants of the project team established for this project. This project team consisted of nine experts working in participating organizations of the Academic Workplace Forensic Care for Youth, which facilitates current study. Collaborating organizations are two universities, two universities of applied science, two JJIs, and two institutions for child and adolescent psychiatry.

Round 1

The Delphi study started with an initial open question: “What are, according to you, important domains of characteristics of adolescents and their context to take into account before you start a treatment trajectory or to decide to indicate someone for a specific intervention in a juvenile justice institution?” The experts were asked to name at least five domains and explain them. These explanations were indicated as the underlying factors for the domains.

Coding

Based on the answers from the experts in the first round, domains of offender characteristics were formed. During a process of “open coding” (Boeije, 2005), all answers and explanations from the experts were read and related underlying factors were defined and grouped together in coherent domains. During this process of open coding, all answers were read independently by the two researchers; the main message was taken out, discussed, and formulated in a uniform manner. Subsequently, the answers were analyzed based on corresponding domains. Initially, two researchers read all the answers, labeled similar answers with colors, sorted them by domain and labeled them with relevant domain titles. Then, this process was repeated by the project team, using different colored post-its per domain. This open coding procedure took 2 hr and led to eight domains and was the official coding moment. This process was performed once again for a time period of 45 min with clinical experts from a JJI, who were not in the panel of experts from the Delphi study. During this meeting, the title names of the eight domains were presented to the clinicians beforehand, while room was left for changes. This procedure was chosen to get the most objective grouping of factors in domains and check the domains for relevance in clinical practice. Finally, all coded answers (domain title and explanation) were imported in the software program Atlas.ti Version 7 to explore frequencies and connections between domain descriptions. No answers were left out, and double answers were imported multiple times.

Round 2

In the second questionnaire, the experts were asked to list the domains in order of importance. Subsequently, the experts were asked to complete the list of domains in case they found something was missing in the domain or in an explanation of the underlying factors.

In addition, they were explicitly asked about the role of protective and promotive factors as these were barely mentioned in the first round. Clinicians could indicate on a 10-point scale how important protective factors were to them during the constructing of treatment trajectories.

Based on the experts’ answers in the second Delphi round, the list of domains and underlying factors was modified by the project team. This modified list was used as part of the questionnaire for the third Delphi round.

Round 3

In the last Delphi round, the experts were asked to think about two cases from their current clinical practice. In the first case, treatment went as planned, the goals formulated during the residential stay were reached, and there were (moderate) successes. The second case should be a case where treatment had failed because treatment goals were not reached despite the efforts of treatment staff. The experts were asked to consider the most important factors for treatment success or failure, choosing from the list of domains and underlying factors resulting in Round 2. Finally, the participants were asked to reflect retrospectively on missing elements in the residential treatment of the juvenile delinquents. These questions were added to evaluate to what extent the domains are used in actual daily practice and what underlying factors were seen as important per domain.

Expert Meeting

Finally, all experts participating in the Delphi rounds were invited for an expert meeting. During this meeting, the final set of eight domains with their underlying factors was presented and the prioritization and descriptions of the domains were checked for the last time and moderated in three focus groups.

Results

Round 1

The first open Delphi round resulted in a total of 229 domains (M = 7.4, SD = 2.3, range = 4-11) from the experts on the first question. The domains experts mentioned were mainly focused on problem behavior and negative behavior from the adolescent and his context, such as psychopathology, poor parenting, or difficulties with peers. Within this focus on problematic factors, seven (of 34) experts mentioned domains with offense characteristics as important in treatment planning. Only three experts also mentioned protective factors as important for treatment planning.

The results from the first Delphi round were read and labeled according to the “open coding” process. Analysis of these answers resulted in the eight separate domains: Family, Mental health problems, Motivation, Offense, Peers/Spare time, Personal characteristics, Treatment, and School/Work/Housing. The explanations the experts gave for each domain were used for defining the underlying factors of the domains. All underlying factors were distributed to a domain. Table 1 provides an overview of the eight domains and their conceptualization after the entire Delphi study and is presented after the results of the focus group.

Table 1.

The Eight Domains and Their Most Important Underlying Factors Resulted From the Third Delphi Round.

Domain	Most important underlying factor stimulating factors
Mental health problems	Psychiatric symptoms Adjustment of psychiatric symptoms Comorbidity Acknowledgment of problems Medication
Personal characteristics	Consciousness of problems/learnability Personality development/attachment Core cognitions Emotion regulation Intellectual/social/verbal capacities Strengths Culture
Family	Important person(s) in system Availability of emotional support Involvement in treatment Assessability Recognize problems System interactions Pedagogic capacities Role in future Strengths
Offense	Attitude toward the offense Openness in offense history Context of the offense Substance (ab)use Coping during offense(s) Consequences of offense
Motivation	Attitude toward treatment Request for help from adolescent Motivation for treatment Influenced by peers
Treatment	Connection between the core problems and treatment Time in the institution Previous treatment and interventions Therapists outside the institution
School/work/housing	Possibilities from the adolescent Own wishes for school/work/housing Motivation for school/work Finances
Friends/spare time	Received support from peers Presence of a positive role model Network outside the institution Romantic relationship Contact delinquent peers Influenced by peers

The eight domains partially overlap as the experts gave similar explanations, although appointing them to a different domain. An example of these interrelations can be seen in Figure 1.

Figure 1.

Underlying factors of the domain Treatment and its connections with other domains.

Round 2

In the second round, the eight domains that had been formulated based on what clinicians mentioned in Round 1 were checked by the participating experts. It was asked if the eight domains corresponded with their earlier answers and if something was missing. All participating clinicians responded that the domains corresponded almost completely with their previous response, resulting in no adjustments to the division or labeling of the eight domains. They provided just a short extension of the explanations they gave with each domain.

Next, the participants were asked to prioritize the domains according to importance for treatment planning. Mental health problems appeared as the most important domain for planning treatment trajectories, followed by Personal characteristics, Family, Offense, Motivation, Treatment, School/Work/Housing, and Peers/Spare time.

The experts indicated that protective factors in the lives of the adolescents were very important to evaluate when starting treatment planning. This specific question was answered by the experts on a 10-point scale, with the mean of 8 points (M = 7.96, SD = .92, range = 6-10).

Round 3

The third Delphi round resulted in a final consensus on the eight domains and underlying factors based on clinical cases from the current practice of the experts. Results from this round demonstrated that important factors, which are neglected in treatment planning, mainly focus on the network of supportive people in the adolescent’s network. The most often mentioned factors that were missed, but retrospectively found very important, were “Knowledge about the problem/disorder by adolescent and his family” (Mental health problems), “Strengths of family members” (Family), “Cooperation with therapists outside the institution” (Treatment), and “Knowledge about network of friends outside the institution” (Peers/Spare time).

Expert Meeting

The expert meeting resulted in the final description of the domains and most important underlying factors. Table 1 gives an overview of the final results of the Delphi study. In the domain of Mental health problems, factors as “Adjustment of psychiatric symptoms” and “Acknowledgment of problems” were mentioned often as very important and were therefore included in the domain description. Describing factors from the Personal characteristic domain, based on the experts opinions, included “Consciousness of problems,” “Learnability,” and “(negative) Cognitions.” The “Evolvement in treatment” and “Availability of emotional support” were included in the description of the Family domain. In the domain Offense, the experts found the “Attitude toward the offense” and the “Context of the offense” important describing factors based on their experience. “Attitude toward treatment” and “Request for help from juvenile” were mentioned most often by the experts as important factors in the Motivation domain and included in domain description. In the Treatment domain, the factors “Connection between the core problems and treatment” and “Time in the institution” were pointed out as important describing factors by the experts. In the domain of School/Work/Housing “Possibilities from the juvenile,” his “Own wishes” and “Motivation,” toward these facets were mentioned as the most important factors and therefore included in the domain description. Finally, concerning the domain Peers-Spare time, the “Received support from peers,” “Network outside the institution,” and the “Presence of a positive role model” are important stimulating factors according to the experts. Regarding the prioritization of the domains, the addition was made that factors from the domain Motivation were found to be important at all times. They should, according to the experts, be taken into account through the whole trajectory of treatment.

Discussion

The current qualitative study, based on the Delphi technique, focused on the implicit knowledge of 34 forensic clinical experts with regard to domains relevant for clinical decision making in treatment planning for incarcerated juvenile offenders. Eight separate domains were identified: Mental health problems, Personal characteristics, Family, Offense, Motivation, Treatment, School/Work/Housing, and Peers/Spare time. The experts indicated that Mental health problems and Personal characteristics were the most important domains in treatment planning. However, Motivation has the most influence on the priorities and interrelation of the other domains. These practice-based domains provide guidance in treatment planning when facing youth with a complex combination of co-occurring disorders and risk factors (e.g., Andrews et al., 2006; Lipsey, 2009). Therefore, these results are an addition to currently used (risk) factors derived from relevant quantitative studies. Our Delphi study has yielded unique directives on how to approach treatment planning, for example, to use a broader scope on what factors to consider instead of a focus on what risk factors to treat. That is because the domains, which are pointed out by the experts to be important for treatment of juveniles, are explained by a variety of factors that increase the motivation, treatment circumstances, and involvement of family members during treatment. The list of domains and their underlying factors gives clinicians direction toward information that needs to be gathered and targets for a broad overview of the individual. Based on the current findings, treatment of adolescents in JJIs should contain three elements, which will be further discussed: (a) a wide scope on the offender characteristics, represented by multiple domains; (b) the value of motivating the juvenile for treatment; and (c) focus on protective factors and future perspectives in the life of the adolescent.

First, the eight domains pointed out by the clinicians to be important for treatment contain more factors than domains included in the SAVRY, a risk assessment instrument for future violence and delinquency that is frequently used in JJIs (Borum et al., 2002). Nevertheless, most criminogenic factors from the SAVRY (Borum et al., 2002) can be found in the eight domains. For example, the historic items are all related to either the Family or the Offense domain. The contextual items are related to the domains, Family, School/Work/Housing, and Peers/Spare time. The clinicians from current study gave less value to factors directly related to the offenses and externalizing behavior during their process of treatment planning. Instead, they focused on the importance of the personal characteristics of the offender. These characteristics describe a more extensive and divergent spectrum of factors (e.g., coping and attachment style, acknowledgment of problems, verbal capacities) than the individual items from the SAVRY and other risk assessment instruments. It is important to emphasize the incorporation of a wider scope of mental health problems and specific characteristics of the adolescent, when deciding on the allocation of specific treatment trajectories. Studies signify the importance of identification and treatment of mental health problems in adolescents in juvenile JJIs (Espelage et al., 2003; Hoeve et al., 2014). Without the assessment of internalizing problems, externalizing behavior appears unpredictable (Espelage et al., 2003). Therefore, it seems necessary for the treatment of externalizing problems to detect and treat internalizing problems. This wider scope and focus on personal characteristics corresponds partly with theoretical frameworks underlying most treatment interventions used in JJIs such as the RNR model and GLM. These frameworks provide guidelines to consider the heterogeneity of the population in general terms, such as the need to adjust the intensity of treatment to the level of risk for an individual to reoffend. However, these models mainly corroborate what to treat, pointed out in risk factors, and therefore current findings, focusing on how to approach individualized treatment planning, expand, and complete these models.

Second, the present results demonstrate the importance of the motivation of the adolescent for development and treatment. Although it is difficult to make guidelines about the prioritization of the eight domains, as each individual is different, clinicians point out the domain of Motivation as a crucial factor influencing the process of treatment for every juvenile in treatment. Therefore, the motivation of the adolescent to change should be used as a starting point for treatment. This is on par with research that has found motivation a key factor in the positive proceeding of the treatment trajectory (e.g., Salekin, Lee, Schrum Dillard, & Kubak, 2010) and in reducing the risk of reoffending (e.g., Mulder, Brand, Bullens, & van Marle, 2010; Olver et al., 2011). A focus on motivation can also be used to overcome the difficulty of the interrelationship between the different domains as motivation is connected to all other domains. When adjusting treatment to the responsiveness of the juvenile, the chances are bigger that the adolescent will cooperate, which will lead to positive treatment outcome. Clinicians, however, still need to be aware of the co-occurrence and interaction of mental health problems, as well as challenges and strengths in the psychological, social, and contextual field when constructing treatment pathways.

Third, attention from clinicians should also be guided toward positive characteristics of, and around, the adolescent toward future perspectives of life. Clinicians in the current study initially put a prominent focus on problem characteristics and risk factors of the adolescent. Only three experts mentioned protective factors as important for treatment planning. When asked specifically, they point toward the importance of the strengths and qualities of the adolescent and his context during treatment. Although clinicians confirm the importance of these protective and promoting factors in daily practice, these factors seem to fade to the background during treatment as positive factors are sparsely mentioned in Round 1 of the Delphi study. In a work environment where clinicians are faced with considerable negativity, stress, and responsibility and where safety is of concern, one has a tendency to focus on problems instead of opportunities and strengths. However, studies on the role of protective factors in the assessment of risk for future violence in adults demonstrate the added value of the combined focus on risk as well as protective factors (de Vries Robbé et al., 2013). This undoubtedly applies to adolescents as well. Clinicians in the current study also highlight the lack of focus on cooperating during residential treatment with significant others (e.g., parents, peers, social workers) outside the JJI. They mention that this is particularly important when working toward successful rehabilitation. In current clinical forensic practice, engaging parents, peers, and other important people outside the JJIs is relatively new. Moreover, to support clinicians in their decision making, the protective and promoting factors, such as motivation for treatment and constructive relations with prosocial peers, as well as a view toward positive future perspectives need to have a more prominent role.

The current Delphi study among forensic clinicians resulted in clear and important domains for guidance in clinical decision making during treatment planning, however, it also has some limitations. First, during the different Delphi rounds experts dropped out. Although dropout was due to understandable reasons, such as job loss following budget cuts, the results and validity of current Delphi study would be of greater value when comprehending the view and experience of all experts who initially started this Delphi study. This dropout led to a lower response rate (59%) for the third round in relation to the first (86%) and second (79%) rounds of questionnaires. Although a gold standard for the response rate of Delphi is lacking (e.g., Keeny et al., 2011; Walker & Selfe, 1996), the first two rounds can be considered as very good and thus reliable. Moreover, the main aim of current study was addressed in the first two rounds: to find a series of well-defined domains that are considered important and suitable to transfer to other clinicians working within the adolescent forensic population to support the process of treatment planning. The third round was added to test the developed domains in daily practice of the expert panel, and therefore less crucial for the main aim of the present study. Second, most of the clinicians in this study are working in residential settings, which gives a unique perspective on treatment for this specific highly problematic population. However, to be able to generalize the results to forensic care for adolescents, information from different forensic settings is needed.

To conclude, the current study makes implicit knowledge from forensic clinicians concerning the process of treatment decisions for juvenile delinquents more explicit. Mental health problems, Personal characteristics, Family, Offense, Motivation, Treatment, School/Work/Housing, and Peers/Spare time are considered the main domains. These eight domains give guidance to clinicians for treatment allocation in forensic settings and emphasize on how to treat an adolescent in addition to what to treat. With current findings, the first step in bridging the “science-practitioners gap” is made as these findings are based on clinical experience of individual cases instead of group differences in large populations of adolescents. Moreover, this study makes the implicit process of treatment planning more tangible and is a valuable addition to outcomes of scientific studies and theoretical frameworks that promote individual suited treatment based on a combined set of domains. Future research should keep focus on combining the practical and scientific field as both worlds can learn from each other. With this knowledge in mind, a future qualitative study on effective treatment interventions for each of the eight domains, combined with recent literature, is strongly recommended, and will be executed. Also, to support clinicians in weighing and prioritizing different problem fields, future studies should focus on the effectiveness of treatment interventions in juvenile delinquents with comorbidity.

Footnotes

Acknowledgements

The authors would like to acknowledge ZonMw for funding this project, as well as thank all the experts from JJIs in the Netherlands that gave their contribution to this study.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: All authors are affiliated with the Academic Workplace Forensic Care for Youth, funded by The Netherlands Organization for Health Research and Development (ZonMw, The Hague; grant nr 159010002). These funding sources had no involvement in the study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.

References

Adler

Ziglio

(1996). Gazing into the Oracle: The Delphi method and its application to social policy and public health. London, England: Jessica Kingsley Publishers.

Andrews

D. A.

Bonta

S. J.

(2003). The psychology of criminal conduct (3rd ed.). Cincinnati, OH: Anderson Publishing.

Andrews

D. A.

Bonta

S. J.

Wormith

J. S.

(2006). The recent past and near future of risk/or need assessment. Crime & Delinquency, 52, 7-27.

Andrews

D. A.

Bonta

S. J.

Wormith

S. J.

(2000). Level of Service/Case Management Inventory: LS/CMI. Toronto, Canada: Multi-Health Systems.

Andrews

D. A.

Zinger

Hoge

R. D.

Bonta

S. J.

Gendreau

Cullen

F. T.

(1990). Does correctional treatment work? A clinically relevant and psychologically informed meta-analysis. Criminology, 28, 369-404.

Barnao

(2013). The Good Lives Model tool kit for mentally disordered offenders. The Journal of Forensic Practice, 15, 157-170.

Barnao

Ward

(2015). Sailing uncharted seas without a compass: A review of interventions in forensic mental health. Agression and Violent Behavior, 22, 77-86.

Boeije

(2005). Analyseren in kwalitatief onderzoek: Denken en doen [Analyses in qualitative research: Thinking and Acting]. Den Haag, The Netherlands: Boom en Lemma Uitgevers.

Borum

Bartel

Forth

(2002). Manual for the Structured Assessment of Violence Risk in Youth (SAVRY). Tampa: University of South Florida.

10.

Chitsabesan

Bailey

Williams

Kroll

Kenning

Talbot

(2006). Learning disabilities and educational needs of juvenile offenders. Journal of Children’s Services, 2(4), 4-17.

11.

Chung

H. L.

Steinberg

(2006). Relations between neighborhood factors, parenting behaviors, peer deviance, and delinquency among serious juvenile offenders. Developmental Psychology, 42, 319-331.

12.

Colins

Vermeiren

Vreugdenhil

van den Brink

Doreleijers

Broekaert

(2010). Psychiatric disorders in detained male adolescents: A systematic literature review. The Canadian Journal of Psychiatry, 55, 255-263.

13.

Davies

Black

Bentley

Nagi

(2013). Forensic case formulation: Theoretical, ethical and practical issues. Criminal Behaviour and Mental Health, 23, 304-314.

14.

de Vries Robbé

de Vogel

Douglas

K. S

. (2013). Risk factors and protective factors: A two-sided dynamic approach to violence risk assessment. Journal of Forensic Psychiatry & Psychology, 24, 440-457.

15.

Eells

T. D.

Lombart

K. G.

(2003). Case formulation and treatment concepts among novice, experienced, and expert cognitive-behavioral and psychodynamic therapists. Psychotherapy Research, 13, 187-204.

16.

Espelage

D. L.

Cauffman

Broidy

Piquero

A. R.

Mazerolle

Steiner

(2003). A cluster-analytic investigation of MMPI profiles of serious male and female juvenile offenders. Journal of the American Academy of Child & Adolescent Psychiatry, 42, 770-777.

17.

Farrington

D. P.

Coid

J. W.

Murray

(2009). Family factors in the intergenerational transmission of offending. Criminal Behavior and Mental Health, 19, 109-124.

18.

Fazel

Doll

Långström

(2008). Mental disorders among adolescents in juvenile detention and correctional facilities: A systematic review and meta-regression analysis of 25 surveys. Journal of the American Academy of Child & Adolescent Psychiatry, 47, 1010-1019.

19.

Fiander

Burns

(2000). A Delphi approach to describing service models of community mental health practice. Psychiatric Services, 51, 656-658.

20.

Frost

D. W.

Cook

D. J.

Heyland

D. K.

Fowler

R. A.

(2011). Patient and health care professional factors influencing end-of-life decision-making during critical illness: A systematic review. Critical Care Medicine, 39, 1174-1189.

21.

Garland

A. F.

Hawley

K. M.

Brookman-Frazee

Hurlburt

M. S.

(2008). Identifying common elements of evidence-based psychosocial treatments for children’s disruptive behavior problems. Journal of the American Academy of Child & Adolescent Psychiatry, 47, 505-514.

22.

Hart

Sturmey

Logan

McMurran

(2011). Forensic case formulation. International Journal of Forensic Mental Health, 10, 118-126.

23.

Haynie

D. L.

Osgood

D. W.

(2005). Reconsidering peers and delinquency: How do peers matter? Social Forces, 84, 1109-1130.

24.

Hoeve

Dubas

J. S.

Eichelsheim

V. I.

van der Laan

P. H.

Smeenk

Gerris

J. R. M.

(2009). The relationship between parenting and delinquency: A meta-analysis. Journal of Abnormal Child Psychology, 37, 749-775.

25.

Hoeve

McReynolds

L. S.

Wasserman

G. A.

(2013). The influence of adolescent psychiatric disorder on young adult recidivism. Criminal Justice and Behavior. Advance online publication. doi:10.1177/0093854813488106.

26.

Hsu

C. C.

Sandford

B. A.

(2007). The Delphi technique: Making sense of consensus. Practical Assessment, Research & Evaluation, 12(10), 1-8.

27.

Jenkins

D. A.

Smith

T. E.

(1994). Applying Delphi methodology in family therapy research. Contemporary Family Therapy, 16, 411-430.

28.

Kazdin

A. E.

(2008). Evidence-based treatment and practice: New opportunities to bridge clinical research and practice, enhance the knowledge base, and improve patient care. American Psychologist, 63, 146-159.

29.

Keeny

Hasson

McKenna

(2011). The Delphi technique in nursing and health research. Oxford, UK: Wiley-Blackwell.

30.

Lipsey

M. W.

(2009). The primary factors that characterize effective interventions with juvenile offenders: A meta-Analytic overview. Victims & Offenders, 4, 124-147.

31.

Lutz

Lambert

M. J.

Harmon

S. C.

Tschitsaz

Schürch

Stulz

(2006). The probability of treatment success, failure and duration: What can be learned from empirical data to support decision making in clinical practice? Clinical Psychology & Psychotherapy, 13, 223-232.

32.

Maloney

Miller

(2014). The impact of a risk assessment instrument on juvenile detention decision-making: A check on “perceptual shorthand” and “going rates.” Justice Quarterly, 32, 900-927. doi:10.1080/07418825.2013.863961

33.

McReynolds

L. S.

Schwalbe

C. S.

Wasserman

G. A.

(2010). The contribution of psychiatric disorder to juvenile recidivism. Criminal Justice & Behavior, 37, 204-216.

34.

Mulder

Brand

Bullens

Van Marle

(2010). Profiling serious juvenile offenders in juvenile institutions: change in risk factors in their population. International Journal of Forensic Mental Health, 9(2), 93-100.

35.

Okiishi

J. C.

Lambert

M. J.

Eggett

Nielsen

Dayton

D. D.

Vermeersch

D. A.

(2006). An analysis of therapist treatment effects: Toward providing feedback to individual therapists on their clients’ psychotherapy outcome. Journal of Clinical Psychology, 62, 1157-1172.

36.

Olver

M. E.

Stockdale

K. C.

Wormith

J. S.

(2011). A meta-analysis of predictors of offender treatment attrition and its relationship to recidivism. Journal of Consulting and Clinical Psychology, 79, 6-21.

37.

Rowe

Wright

(2011). The Delphi technique: Past, present, and future prospects: Introduction to the special issue. Technological Forecasting & Social Change, 78, 1487-1490.

38.

Salekin

R. T.

Lee

Schrum Dillard

C. L.

Kubak

F. A.

(2010). Child psychopathy and protective factors: IQ and motivation to change. Psychology, Public Policy, and Law, 16, 158-176.

39.

Sharkey

S. B.

Sharples

A. Y.

(2001). An approach to consensus building using the Delphi technique: Developing a learning resource in mental health. Nurse Education Today, 21, 398-408.

40.

Sim

Gwee

K. P.

Bateman

(2005). Case formulation in psychotherapy: Revitalizing its usefulness as a clinical tool. Academic Psychiatry, 29, 289-292.

41.

Spreen

Timmerman

M. E.

Ter Horst

Schuringa

(2010). Formalizing clinical decisions in individual treatments: Some first steps. Journal of Forensic Psychology Practice, 10, 285-299.

42.

Stewart

R. E.

Chambless

D. L.

(2007). Does psychotherapy research inform treatment decisions in private practice? Journal of Clinical Psychology, 63, 267-281.

43.

Tashakkori

Teddlie

(1998). Mixed methodology: Combining qualitative and quantitative approaches. Thousand Oaks, CA: Sage.

44.

Teplin

L. A.

Abram

K. A.

McClelland

G. N.

Dulcan

M. K.

Mericle

A. A.

(2002). Psychiatric disorders in youth in juvenile detention. Archives of General Psychiatry, 59, 1133-1143.

45.

W. T.

Vandevelde

Soyez

De Smet

Boers

Vanheule

(2014). Treatment perspectives on interned mentally ill offenders in a forensic psychiatric center (FPC): A Delphi study on experts’ opinions. Psychology, Crime & Law, 20, 61-77.

46.

van Hagen

Spaander

M. C. W.

van der Gaast

van Rij

C. M.

Tilanus

H. W.

van Lanschot

J. J. B.

Wijnhoven

B. P. L

. (2011). Impact of a multidisciplinary tumour board meeting for upper-GI malignancies on clinical decision making: A prospective cohort study. International Journal of Clinical Oncology, 18, 214-219.

47.

Vermeiren

(2003). Psychopathology and delinquency in adolescents: A descriptive and developmental perspective. Clinical Psychology Review, 23, 277-318.

48.

Vermeiren

Jespers

Moffitt

(2006). Mental health problems in juvenile justice populations. Child and Adolescent Psychiatric Clinics of North America, 15, 333-351.

49.

Vreugdenhil

Doreleijers

T. A.

Vermeiren

Wouters

L. F.

van den Brink

(2004). Psychiatric disorders in a representative sample of incarcerated boys in the Netherlands. Journal of the American Academy of Child & Adolescent Psychiatry, 43, 97-104.

50.

Walker

A. M.

Selfe

(1996). The Delphi method: A useful tool for the allied health researcher. International Journal of Therapy and Rehabilitation, 3, 677-681.

51.

Ward

Brown

(2004). The good lives model and conceptual issues in offender rehabilitation. Psychology, Crime & Law, 10, 243-257.

52.

Ward

Mann

R. E.

Gannon

T. A.

(2007). The good lives model of offender rehabilitation: Clinical implications. Aggression and Violent Behavior, 12, 87-107.

53.

Ward

Yates

P. M.

Willis

G. M.

(2011). The good lives model and the risk need responsivity model: A critical response to Andrews, Bonta, and Wormit (2011). Criminal Justice and Behavior, 39, 94-110.

54.

Witteman

C. L. M.

Kunst

(1997). Planning the treatment of a depressed patient. Clinical Psychology & Psychotherapy, 3, 157-171.

55.

Wylie

L. A.

Griffin

H. L.

(2013). G-map’s application of the Good Lives Model to adolescent males who sexually harm: A case study. Journal of Sexual Aggression, 19, 345-356. doi:10.1080/13552600.2011.650715