Introduction
Mechanisms of persistent self-destructive behaviour in patients with internalizing behaviour problems are still poorly understood. Although a vast body of research exists with results at a variable level, research regarding pathway approaches seems to be relatively sparse, yet important to understand individual development of psychopathology [
1]. This type of research may be hampered by current diagnostic systems, such as DSM-5 [
2] or ICD-10 [
3] that aim to classify taxonic conditions and aim to be a-theoretical regarding origins. However, many psychiatric disorders do not show to be clearly demarcated taxonic entities [
4,
5], and may be criticized because of it’s a-theoretical nature [
6]. Hence, different systems are being developed to diagnose and investigate dimensional entities. One of these is the Hierarchical Taxonomy of Psychopathology (HiTOP; [
7]). HiTOP aims to combine individual psychopathological phenomena into homogeneous domains, and grouping them into psychopathology spectra (e.g., internalizing, externalizing, and thought disorder). As such, HiTOP has a strong focus on quantitative nosology, clinical phenomena and a hierarchical structure. It supports a transdiagnostic dimensional approach [
8] and aims to facilitate reconceptualization of psychopathology by accentuating their interdependence and relevance for one another [
9].
HiTOP is highly driven by grouping clinical psychopathological phenomena into a meaningful hierarchical structure. The HiTOP spectra correlate with the overall liability of a person to develop psychopathology, the
p factor [
10]. This liability may lead to psychopathology in one of the spectra. The
p factor theory underscores the importance of genetic underpinnings for the general liability to develop psychopathology, although transdiagnostic mechanisms may also be found at different domains, such as poor emotion control, or different levels of thought disorder [
11]. A psychopathologic phenomenon in which poor emotion regulation and disturbed thinking may be explicitly present is suicidality. Moreover, it has been related to a range of psychiatric disorders as a transdiagnostic factor based on a general psychopathology liability [
12]. However, the
p factor is a
general construct which may be difficult to translate to the individual patient, because variance at a group level needs to be critically studied at an individual level for better understanding of diversity in process and outcome [
1]. As such, HiTOP does not yet incorporate aetiology [
13], neither on a group level nor on an individual level.
Aetiology is an important topic in understanding disorders and, as such, diagnosing psychopathology. As HiTOP seems to lack a clear statement on aetiology, the question remains on how to understand aetiology of apparently distinct disorders yet showing overlap in their presentation. Suicidal behaviour, including suicide ideation as well as parasuicidal acts, has been associated with complex developmental pathways [
14], and covers a broad spectrum of psychiatric conditions [
15‐
17]. In adolescence it is often associated with depressive disorder [
18] and with borderline personality disorder (BPD), and may be challenging to treat [
19].
Aiming to address aetiology as well as emotion dysregulation in these disorders, several psychotherapeutic interventions have been developed. Two interventions that seem to be the most important of these are Mentalisation Based Therapy (MBT; [
20]) and Dialectical Behaviour Therapy (DBT; [
21]). MBT is a psychoanalytically oriented intervention that makes mentalizing a core focus of therapy [
22]. Mentalizing is the cognitive process by which we make sense of each other and ourselves, in terms of intentional states [
23,
24]. Traditional MBT provides individual and group sessions, crisis planning and integrated psychiatric care. Individual sessions focus on developing a therapeutic alliance through a close attachment relationship and on maintaining an optimal level of arousal during interactions with others by addressing the details of the mentalizing process. Group sessions focus on mentalizing in a more complex interactional process [
23]. The importance of addressing attachment, and epistemic trust has been stressed by MBT [
25]. In contrast, DBT has emerged from standard behaviour therapy and developed to a transdiagnostic therapy for clients with complex, high-risk disorders [
26]. It is a modular and hierarchical treatment consisting of a combination of individual psychotherapy, group skills, training, telephone coaching, and a therapist consultation team. Although DBT has a strong focus on acceptance, it provides a hierarchy of what to treat and when to treat it for a particular patient. DBT consists of four stages: (a) decrease dysfunctional behaviour, (b) experience emotion, (c) reduce ordinary problems in life, and (d) increase a sense of completeness. Treatment is split into two domains: (a) skills training and (b) problem solving and motivation [
26].
For adolescents, the evidence of efficacy for both DBT and MBT is promising, although somewhat limited [
27‐
29]. Other psychotherapeutic interventions, such as emotion regulation therapy, cognitive analytic therapy, and cognitive behaviour therapy show at best modest improvement [
27,
28]. Furthermore, improvement during psychotherapeutic interventions seem to decrease at follow-up [
30]. As such, a substantial group of these adolescents show to be treatment refractory [
30,
31], and may even show to be determined not to respond to treatment [
32].
As non-response to psychotherapeutic interventions may be due to neurobiological factors, pharmacotherapy may be described to patients with emotion dysregulation. However, pharmacological interventions in adults appear at best adjunctive to psychotherapeutic interventions [
33], and the evidence regarding the efficacy of pharmacotherapy is sparse. Therefore, it has been recommended to avoid pharmacotherapy in BPD. Whereas many scholars focus on improving efficacy of current treatment modalities, information regarding the underpinnings of treatment refractoriness remains scarce [
34].
Another reason for non-response may be that the underlying condition of treatment refractory suicidal behaviour can be difficult to diagnose, and one may diagnose a disorder that is not actually present [
32]. From a perspective of emotion dysregulation (e.g., borderline personality disorder) it may be difficult to diagnose because of transient changes in symptomatology [
35]. From a perspective of thought disorder (e.g., autism), it may be difficult to diagnose because of difficulties in identifying behaviour problems correctly [
36]. This may further be complicated because of the presence of symptomatology that, at a behavioural level, seems to correspond with different kinds of psychopathology, such as internalizing and externalizing behaviour problems (e.g., [
37,
38]), as well as well as to disorders in the thought disorder spectrum (e.g., [
39]). Difficulty to diagnose specific disorders may also be due to the cross-disorder presence of different types of internalizing behaviour, such as anxiety, depressive symptoms, somatization, eating disorder, and emotion regulation disorders [
40].
From a DSM perspective, treatment refractory suicidal behaviour may present itself in a range of disorders with or without comorbidity. HiTOP aims to take a different perspective. The HiTOP framework describes developmental factors that moderate development of psychopathology, and aims to gather data to inform regarding natural course and treatment efficacy. It does not describe specifically how these factors interact in order to develop a specific disorder or comorbidity. Furthermore, it does not seem to address a possible time line in the development of disorders that may precede or follow each other. HiTOP implies that comorbidity may possibly reflect higher order defect, implying genetic differences more prominently than environmental differences.
Treatment refractory suicidal behaviour can be found in a range of disorders in which suicidal behaviour predominantly has been labelled as internalizing behaviour problem. Therefore, we broaden our focus from treatment refractory suicidal behaviour to treatment refractory internalizing behaviour (TRIB), which we define as internalizing behaviour based on emotional developmental pathology, having shown treatment refractoriness to guidelines informed treatment in previous mental health services. The reasons for this treatment refractoriness seem to be poorly understood. The current paper will discuss clinical difficulties in TRIB, with respect to diagnostic considerations and classification, and with respect to therapeutic issues. We performed literature searches in the PubMed database regarding the specific subject in this essay. In our literature searches we aimed to find meta-analyses, structured reviews, papers written by leading scholars to found our theory. As such, we based our theory on best available evidence.
We argue that a focus on ‘diagnosis by classification’ may distract attention of the underlying cause of the core problem, and consequently lead to inadequate choices for interventions. We will introduce a theoretical framework on TRIB. In this model, we reconsider social anxiety as a broader concept and state that the development of early developmental social anxiety may be seen as a central pathway to consecutive internalizing behaviour disorders in adolescence. In line with MBT and DBT, we address issues as trust and motivation for treatment that may be explicitly lacking. However, where MBT and DBT predominantly seem to focus on the individual, we stress the importance of intensively involving parents for improvement of emotion regulation and further development. The TRIB model implies similarities between juvenile emotional disorders and juvenile disruptive behaviour disorders (DBDs). With this model, we aim to provide a hypothetical explanation for TRIB, thus providing a guiding framework for interventions and future research.
Discussion
This essay aims to provide an integrative conceptual framework for TRIB as a guiding framework for future research and interventions. Along the HiTOP framework we outlined an aetiological model for TRIB from a transdiagnostic dimensional approach, addressing the interdependence of biologic and psychological developmental, and specific environmental aspects in relation to the development of psychopathology. The genetic and neurobiological make-up of a child and the interaction with (i.e., reaction of) parents may moderate the expression and reinforcement of emotion regulation, moral development and thus also attachment of a child. Disturbances in these interdependent developmental processes, specific attachment patterns, emotion regulation patterns and moral developmental level may be seen as higher order processes that become dominant in a way that leads to psychopathology, and as such, to an increased load on the p factor. Furthermore, we argue that these early developmental disturbances may lead to early social anxiety, and from thereon, to more severe psychopathology. In line with HiTOP, higher order therapeutic interventions should focus on these higher order factors that moderate psychopathology in fundamental ways. Therefore, we argue that initial therapeutic interventions in TRIB should primarily focus on restoring a secure base script and emotion regulation through intensive systemic interventions. This may help to increase adherence to therapeutic interventions in order to increase exposure and resume societal activities.
The HiTOP model identifies higher-order dimensions that reflect associations among lower-order dimensions. [
13]. Although we primarily take a clinical perspective, the TRIB model seems to fit in empirical evidence regarding hierarchical dimensions for general psychopathology [
182]. As such, the highest order dimension may be a single factor, the
p factor, which is seen as the overall liability to mental disorder [
10,
11], consisting of three sub-dimensions, that is, a psychotic, an internalizing and an externalizing experience dimension [
52]. However, the TRIB model implies the existence of two sub-dimension, that is, a psychotic and an emotion dysregulation dimension. This is in line with previous research showing high correlations among the fears, distress, and externalizing factors [
183]. Moreover, we hypothesize that an important common factor for both internalizing and externalizing behaviour may be found in common parenting practices and attachment processes. Thus, the three-factor model of psychopathology, as described by Krueger, is further condensed in the TRIB model, staying in line with the contention that mental disorders correlate because they are moderated by the same set of genetic and environmental factors [
183]. Nevertheless, DSM classification may help to further tailor therapeutic approach, especially when patients are motivated.
We like to stress that the TRIB model describes
interactional sequences between biological, psychological and social factors, and
not a linear causal relationship in which parents are to be blamed. Nevertheless, we focus on parenting because this seems to open opportunities for positive change and increased well-being for both youths and their parents. Even though genetic and neurobiological influences may be strong [
184], major interventions in the treatment of internalizing as well as externalizing behaviour primarily focus on more efficacious parenting in order to decrease externalizing behaviour. Hence, if internalizing and externalizing behaviour show similar underpinnings, and if the provision of a limit setting intervention is more effective in externalizing behaviour, what then is the best way to provide this intervention? Part of the adolescents with TRIB do not accept limit setting behaviour from their parents. Therefore, a major challenge remains in how to provide an optimal equilibrium between autonomous behaviour and restrictions on this autonomy.
The relationship with trauma, especially recurrent small-t trauma, needs further investigation, in relation to both diagnosing TRIB and treatment. Childhood rejection appears to be linked to rejection sensitivity, and rejection sensitivity has been linked to BPD [
185], while high agreeableness and conscientiousness have been found to predict suicidality in relation to interpersonal trauma [
186]. These findings are in line that being sensitive to rejection is related to increased levels of aggression and victimization [
187], and that attachment organization has been found to be related to suicidal behaviour [
188]. However, assessing developmental trauma is difficult, yet important because of increased comorbidity [
189]. Hence, though small-t trauma may play an important role in the development of TRIB, further research is needed to disentangle its relationship and means for interventions.
Therapeutic alliance needs further investigation. If indeed lack of motivation for treatment is the result of lack of trust, how to increase trust? Restoration of trust through mentalization comprises an important factor in therapeutic alliance [
190]. For this, solution-focused therapy (SFT) seems to provide a helpful paradigm, that is, assessing whether the patient and parents are intrinsically committed to involve and invest in treatment [
191]. If this is the case, the therapeutic alliance can be labelled as a
client relationship. If not, there are two remaining possibilities: in a
visitor relationship the patient does not even wants to be here, in a
complainant relationship the patient does not feel part of the problem. Both types of relationship need to be addressed accordingly to increase the chances for a positive treatment outcome. Co-construction of meaning, together with strength-oriented techniques are important in SFT [
192]. Research implies SFT to be efficacious [
193], with increased confidence, increased self-efficacy and increased community participation [
194]. However, further research is needed on this specific labelling method regarding alliance.
The TRIB model has its limitations. As it represents a theoretical diagnostic model, it is important to investigate its diagnostic validity, such as the Robins and Guze criteria [
195]. Hence, further research is needed to identify more specifically the important signs and symptoms of youth with TRIB, natural course and response to treatment. More specifically, which youths with severe emotion dysregulation are endangered to develop therapy refractoriness? As evidence is growing that environmental and contextual characteristics are inextricably linked to the underlying biological characteristics of psychopathology [
196], it is important to investigate whether in this group of patients parental engagement indeed is of crucial importance. From a HiTOP perspective, it may be important to investigate whether developmental processes as emotion regulation, moral reasoning and attachment are indeed higher order dimensions moderating the
p factor. Furthermore, is seems important to investigate whether social anxiety should be seen as higher order pathology, because it seems to represent a higher order thought disorder, that is, inflexibility in thinking, that may lead to more severe disorders.
Summary
Auto-aggressive behaviour, especially treatment refractory suicidality in adolescents with psychiatric disorders is challenging to clinicians, especially in case there is a need for clinical treatment. This may be due to several factors in which current classification systems, based on the presence or absence of diagnostic entities, and lack of therapeutic efficacy may be important ones. In attempting to overcome these limitations, we have integrated current knowledge regarding causality and interdependency of suicidality and auto-aggressive behaviour across disorders within the HiTOP framework, in order to propose a coherent hypothetical transdiagnostic developmental framework regarding these unsettling behaviours in youths.
We argued that the interdependent development of biologic factors, attachment, moral reasoning and emotion regulation in an overprotective environment may lead to social anxiety disorder and thus to emotion dysregulation and severe internalizing behaviour disorders. Loss of trust and trauma may further moderate the development of psychopathology as well as treatment refractoriness. Avoidance behaviour and lack of motivation for treatment may appear to be the most prominent, which also may show itself in severe internalizing as well as externalizing behaviour problems. To optimize treatment efficacy for both internalizing and externalizing behaviour, it seems important to create a shift in mind-set by de-emphasizing DSM-diagnoses, addressing non-compliance, emphasizing the importance of restoration of trust between parents and their child, and limitation of avoidance behaviour.
We discussed that our model describes interactional—not linear—sequences between biological, psychological and social factors. Furthermore, it seems to be in line with the HiTOP framework, by describing higher order dimensions regarding development of psychopathology and therapy. Although this model is hypothetical, it may be helpful to increase therapeutic efficacy of both biological and psychotherapeutic interventions. Also, this model may help to provide directions for further research.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.