A Transdiagnostic Perspective on Youth Irritability

Abstract

Irritability is increasingly recognized as a significant clinical problem in youth. It is a criterion for multiple diagnoses and predicts the development of a wide range of disorders. Research on etiopathogenesis suggests that genetic and family environmental factors play a role, as do abnormalities in reward and cognitive control circuitry. However, many of these effects are age dependent. Threat-responsive self-regulatory systems and the degree to which irritability manifests as tonic or phasic influence whether irritable youth exhibit more internalizing versus externalizing outcomes.

Irritability is often defined as a low threshold for experiencing anger in response to frustration. It can manifest as a persistently angry or grouchy mood, and/or as temper outbursts, and is one of the most common reasons for referring children and adolescents for mental health services (Carlson & Klein, 2018). Irritability has recently emerged as a major research focus, particularly in children and adolescents. In 2015, a recurring meeting, the Congress on Pediatric Irritability and Dysregulation, was established to bring together investigators working in this area.

The current literature on irritability has been strongly influenced by the work of a small handful of pioneering investigators. These include Leibenluft (2011, 2017), who has spearheaded an extensive clinical and mechanistic research program on severe mood dysregulation (i.e., irritability and outbursts) and its relationship to bipolar disorder and major depression. Stringaris’ work on the course and outcome of irritability and its relationship to other forms of psychopathology, particularly depression, has also been very influential (Stringaris, Vidal-Ribas, Brotman, & Leibenluft, 2018; Vidal-Ribas et al., 2016). Finally, Wakschlag has conducted exemplary developmental research delineating typical and atypical forms of irritability (particularly temper tantrums) in young children (Wakschlag et al., 2012).

However, there continue to be significant challenges in conceptualizing, assessing, treating, and understanding the etiopathogenesis of irritability (Carlson & Klein, 2018; Stringaris et al., 2018). In this paper, we review selected findings on youth irritability, highlighting our own work. We particularly emphasize the transdiagnostic nature of irritability, its “multifinal” (i.e., diverse) outcomes, and several factors that might, in part, account for this multifinality. Interested readers are encouraged to consult more comprehensive sources for greater detail on other areas of research on irritability (e.g., Brotman et al., 2017; Carlson & Singh, in press; Evans et al., 2017; Roy et al., 2019; Roy & Comer, 2020; Stringaris et al., 2018).

Current work on youth irritability is the product of at least four historical developments and streams of research (Carlson & Klein, 2018). First, emotion dysregulation is a common feature of attention deficit hyperactivity disorder but has never been included in the diagnostic criteria for this condition. This has led to confusion about where to classify children with hyperactivity and impulsivity who also display prominent irritable mood and temper outbursts (Faraone et al., 2019). Second, imprecision in the wording of the diagnostic criteria for bipolar disorder and in the structured diagnostic interviews designed to elicit these criteria opened the door to classifying children with prominent irritability and outbursts but who did not exhibit the elated mood and episodic course characteristic of classical mania as having an atypical form of bipolar disorder. Third, to stem the resulting “epidemic” of pediatric bipolar disorder, the diagnosis of Disruptive Mood Dysregulation Disorder (DMDD) was introduced in the 5th edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-5; American Psychiatric Association, 2013) to provide an alternative diagnostic “home” for highly dysregulated, explosive and chronically irritable children (Carlson & Klein, 2014). Finally, research on childhood disruptive behavior disorders showed that oppositional defiant disorder (ODD) can be parsed into irritable mood and externalizing behavior dimensions with different prognostic implications (Stringaris & Goodman, 2009).

Irritability as a transdiagnostic construct

Although irritability is the core of the new and controversial diagnosis of DMDD (Evans et al., 2017), it is also a criterion for numerous other diagnoses in DSM-5, including major depression (in youth), mania an hypomania, generalized anxiety disorder (GAD), posttraumatic stress disorder, ODD, intermittent explosive disorder, and borderline and antisocial personality disorder. As irritability cuts across so many diagnostic categories, it is arguably the most transdiagnostic symptom in the DSM. Interestingly, however, there is no definition of irritability in the DSM.

Irritability can also be viewed from the perspectives of the two major transdiagnostic classification systems. In the Research Domains Criteria (RDoC; Cuthbert, 2020), irritability is best represented by the biobehavioral construct of frustrative non-reward in the negative valence systems domain. In the Hierarchical Taxonomy of Psychopathology (HiTOP; Kotov et al., 2017), irritability is included in the distress subfactor of the internalizing spectrum.

A growing body of literature indicates that, at least in youth, the transdiagnostic nature of irritability is reflected in an exceptional degree of multifinality, as irritability predicts a wide variety of internalizing and externalizing outcomes, as well as suicidality, in children, adolescents, and adults (Evans et al., 2017; Orri et al., 2019; Vidal-Ribas et al., 2016). For example, in a large community sample of preschoolers, Dougherty et al. (2015) found that persistent irritability, assessed via a semi-structured interview with a parent, predicted higher levels of depression, anxiety, and disruptive behavior symptoms, functional impairment, and mental health services at age 9. Similar findings were again evident in follow-ups at ages 12 and 15, with most associations persisting after adjusting for preschool psychopathology and a variety of other early risk factors (Sorcher et al., 2021). Extending these between-persons findings, Chad-Friedman et al. (2021) examined within-person associations in this sample. Across 5 assessment waves spanning ages 3–15, children who experienced increased levels of irritability at a wave, relative to their own baseline, experienced heightened depression, anxiety, and ODD symptoms both at that wave and in the subsequent wave, relative to their typical levels. Taken together, these findings indicate that irritability exhibits similar patterns of multifinality across a wide developmental span, and suggest that early intervention could reduce rates of multiple forms of psychopathology.

Development and antecedents

The developmental course of temper outbursts is better established than for irritable mood. Tantrums are common during the preschool years but decrease through childhood and adolescence (Wakschlag et al. 2012). Irritability shows moderate rank-order (relative) stability over time using dimensional measures, but stability is much lower using categorical measures (Kessel, Dougherty et al., 2021; Vidal-Ribas et al., 2016). Several studies have used growth mixture modelling and related techniques to identify classes of children and adolescents with varying trajectories of irritability (e.g., Riglin et al., 2019; Wiggins, Mitchell, Stringaris, & Leibenluft, 2014). For example, in a community cohort followed from age 7 to 15, Riglin and colleagues (2019) identified 5 trajectories of irritability: low (81.2%), decreasing (5.6%), increasing (5.5%), late-childhood limited (5.2%), and high-persistent (2.4%).

While there is a substantial literature on outcomes of youth irritability, less is known about antecedents. This is important because irritability is not just a risk factor for adverse outcomes, but a significant clinical problem in its own right.

Paralleling its multifinality of outcomes, irritability is characterized by an equifinality of antecedents, with multiple influences contributing to its development. A small number of family and twin studies indicate that irritability is moderately heritable and has strong familial/genetic links with depression, and, less consistently, with anxiety and substance use disorders (Kessel, Dougherty et al., 2021; Riglin et al., 2019: Wiggins et al., 2014). Notably, the genetic influences contributing to irritability vary as a function of development and gender, with heritability diminishing with age for females but remaining stable or increasing for males (Roberson-Nay et al., 2015). Moreover, early irritability may be more genetically related to neurodevelopmental conditions, while later irritability may be more closely linked to disruptions in mood (Riglin et al., 2019). This highlights the importance of examining irritability at specific stages of development. However, when comparing different ages, it is also necessary to consider the influence of assessment methods and informants which are confounded with development (e.g., young children cannot provide self-reports; observational methods are more challenging with older youth).

Temperament is another likely risk factor, although it has received surprisingly little attention (Beauchaine & Tackett, 2020). There is considerable conceptual overlap between irritability and the temperament trait of negative emotionality (NE), the precursor of the personality trait of neuroticism. Like irritability, NE and neuroticism are transdiagnostic traits that are associated to varying degrees with all forms of psychopathology (Klein et al., 2011). Importantly, NE and neuroticism both include multiple facets, one of which is trait anger. Trait anger may be the component of NE/neuroticism that drives the symptom of irritability and accounts for its associations with both internalizing and externalizing disorders.

Early environmental factors are another potentially important, but understudied, domain. Wiggins et al. (2014) reported that exposure to family violence predicted higher trajectories of irritability throughout childhood. In a prospective study, Kessel, Dougherty and colleagues (2021) found that, adjusting for concurrent irritability, greater marital discord and permissive and authoritarian parenting at age 3 predicted higher levels of irritability at age 15, although the results varied somewhat depending on the child’s sex and the informant at age 15 (youth or parent).

Mechanisms

Mechanistic accounts of irritability have focused on neural circuitry implicated in reward and threat processing and cognitive control (Brotman et al., 2017; Leibenluft, 2017). These circuits have also been associated with a variety of other psychiatric disturbances, which may help explain why irritable youth are vulnerable to such a broad range of psychopathology.

Much of the research on neural correlates has focused on aberrant responding to non-reward (when rewards are blocked/withheld), reflected by amygdala-fronto-striatal dysfunction (Brotman et al., 2017). For instance, Dougherty et al. (2018) reported associations between irritability and functional alterations in regions supporting reward processing (i.e., striatum) and their connections with inhibition-related regions (i.e., prefrontal cortex). Importantly, these associations varied as a function of the age at which irritability was evident.

Irritable youths’ reward-related aberrations may extend beyond non-reward contexts. Kessel, Dougherty et al. (2016) found that preadolescents with a history of preschool irritability exhibited an enhanced Reward Positivity, an electrocortical response thought to reflect reward sensitivity, in a gambling task.

Cognitive control may also play an important role. For instance, cognitive flexibility is the ability to appropriately adjust one’s behavior to a changing environment. Using a developmentally modified Wisconsin Card Sorting Task, Kessel, Nelson et al. (2021) found that irritability was associated with poor behavioral performance and aberrant neural reactivity when required to shift set. Taken together, these findings suggest that irritable children may have difficulty modulating approach/reward reactivity, particularly in frustrating and dynamic contexts, and highlight the importance of jointly considering “bottom up” reward and “top down” control processes, and how they change across development.

Understanding multifinality

Why do some irritable children develop internalizing outcomes and others exhibit externalizing disorders? To identify factors that might moderate the developmental trajectories of early irritability, we focused on variables meeting two criteria: (1) they are associated with both internalizing and externalizing psychopathology, but in opposite directions; and (2) they are not correlated with irritability. We identified two candidates: (1) the error-related negativity (ERN), an electrocortical response that appears after participants make an error; and (2) patterns of cortisol secretion over the course of the day. Although the ERN and cortisol are products of different biological systems, they both reflect self-regulatory systems that can be activated by external and internal threats. Internalizing disorders tend to be associated with increased ERN and cortisol levels, whereas externalizing disorders are often characterized by blunted ERN and cortisol.

Indeed, we found that ERN and daily patterns of cortisol secretion were uncorrelated with irritability. However, both moderated the effects of irritability at age 3 on internalizing and externalizing symptoms at ages 9 and 12, adjusting both for baseline levels of these symptoms and the level of the other set of symptoms at follow-up. Thus, higher levels of irritability at age 3, in conjunction with an elevated ERN (Kessel et al., 2016) or a greater diurnal cortisol slope (with high levels in the morning) (Kessel, Frost et al., in press), predicted increased levels of internalizing symptoms in late childhood/early adolescence. In contrast, higher irritability, together with a reduced ERN or blunted diurnal cortisol slope (low levels throughout the day), predicted increased levels of externalizing symptoms. These findings suggest that the sensitivity of threat-responsive self-regulatory systems may shift the trajectory of early irritability in relatively more internalizing or externalizing directions.

Irritability can be experienced and expressed in a variety of ways. Hence, another approach to understanding its multifinality involves parsing irritability into components that may have different associations with internalizing and externalizing psychopathology. Carlson has distinguished between “how one feels” (e.g., irritable mood) and “what one does” (e.g., temper outbursts or rages) (e.g., Carlson & Klein, 2018). Similarly, Leibenluft has distinguished tonic (persistently angry, easily annoyed) from phasic (temper outbursts) irritability (e.g., Leibenluft, 2017). In a large community sample of older adolescents, Hawes et al. (2020) reported that, adjusting for the diagnosis at baseline, tonic irritability independently predicted depressive and anxiety disorders in adulthood, whereas phasic irritability uniquely predicted substance use disorder and antisocial behavior. Similarly, in a community sample of early adolescent girls, Silver, Carlson et al. (2021) found that, controlling for baseline diagnosis, tonic irritability independently predicted higher rates of depressive disorders and GAD 3 years later, whereas phasic irritability uniquely predicted a lower rate of GAD, and higher rates of ODD, conduct disorder, and substance use disorders, as well as higher levels of risky sexual behavior and relational aggression. More recently, Silver, Bufferd et al (2021) extended this work to young children, showing that tonic and phasic irritability in 6-year olds are empirically separable dimensions. Moreover, adjusting for baseline diagnosis, tonic irritability predicted depressive disorders, social and specific phobia, and suicidal ideation and behavior in adolescence, while phasic irritability predicted subsequent ADHD and disruptive behavior disorders.

These findings suggest that youth with high levels of irritability vary in how their irritability is manifested and in threat-responsive self-regulatory system functioning, which may predict whether they are more likely to develop internalizing or externalizing psychopathology. However, further parsing is needed. For example, irritability may differ in the context of other impairments (autism, learning/language problems) and settings (community vs. clinical) (Carlson & Klein, 2018; Stringaris et al., 2018). For example, in clinical settings it may be particularly important to focus on what children do during outbursts (e.g., verbal versus physical aggression), as the latter is more likely to lead to hospitalization, PRN (i.e., as necessary) medication, and seclusion/restraints (Carlson & Klein, 2018).

Conclusions and future directions

Irritability in youth is highly transdiagnostic. It is the product of equifinal pathways and results in multifinal outcomes. While its manifestations are highly correlated, irritability can be parsed into separable dimensions that predict different outcomes and may reflect different causal processes.

A high priority for future work is sharpening the conceptualization of irritability and improving its measurement. With a few notable exceptions (Stringaris et al, 2018; Wakschlag et al., 2012), irritability has been assessed using measures developed for other purposes. Moreover, there are no instruments designed to distinguish phasic and tonic irritability, or to assess the severity of the phasic/behavioral component which is often the basis for clinical referral and critical treatment decisions (Carlson & Klein, 2018). It is also necessary to include multiple vantage points – children, parents, teachers, and behavioral observations – as each appears to provide unique information (Ali et al., in press; Kessel, Dougherty et al., in press), and some methods may provide more valid information about some aspects of irritability than others (e.g., youth may be better reporters on tonic, and parents on phasic, irritability). Most importantly, measures must be designed and norms established to distinguish typical from atypical levels and forms of irritability throughout development (see Wakschlag et al., 2012 for an exemplar).

Research on etiology and mechanisms is still in the early stages. Most studies use case-control designs comparing individuals with and without a specific diagnosis. This approach is poorly suited to elucidate transdiagnostic processes. Future studies should be designed to facilitate comparisons across a range of diagnoses and symptom dimensions to determine similarities and differences between irritability and other forms of psychopathology, and to explore whether the mechanisms associated with irritability differ in the context of different disorders. In addition, there is a need for further research on the environmental and interpersonal influences on irritability, as well as on the relationship between irritability and relevant trait dispositions (e.g., negative emotionality and its facets, effortful control/disinhibition). It is also critical to explore differences in the etiology and mechanisms underlying the tonic and phasic forms of irritability, which could lead to the development of more precisely targeted treatments. Importantly, genetic and environmental influences on irritability appear to differ at different ages (Roberson-Nay et al., 2015), highlighting the importance of attending to developmental continuities and discontinuities in research on etiology and mechanisms.

Finally, it may be useful to integrate the literatures on anger and irritability in youth and adults (e.g., intermittent explosive disorder), as each body of work may provide insights for the other. In conclusion, youth irritability warrants increased clinical and scientific attention as it can inform our nosology, shed light on transdiagnostic processes, enhance prognostication, and hopefully lead to more effective treatment of youth psychopathology.