Irritability, marked by low frustration tolerance and temper outbursts, has significant public health implications. It is a common reason for youth mental health treatment and predicts lasting impairment into adulthood (Brotman et al., 2017; Copeland et al., 2014; Evans et al., 2022; Stringaris et al., 2009; Vidal-Ribas et al., 2016). Existing literature highlights dynamic influences on irritability, particularly during vulnerable periods like adolescence (Caprara et al., 2007; Riglin et al., 2019; Roberson-Nay et al., 2015). However, little is known about individual differences in irritability across development, hindering effective assessment and intervention, and limiting efforts at targeted prevention.

Person-centered approaches, like growth curve and growth mixture models, are crucial for understanding individual differences in irritability trajectories. A majority of these studies have characterized irritability trajectories starting in early childhood (Wiggins et al., 2014; Zhang et al., 2023) or middle childhood (Orri et al., 2019; Riglin et al., 2019), with some that have extended into adolescence (Yu et al., 2023). Extant studies have generally identified four or five classes that follow the pattern of low and stable, decreasing, increasing, and high and stable irritability levels across ages 3–15 years (Orri et al., 2019; Riglin et al., 2019; Wiggins et al., 2014; Yu et al., 2023). While the majority of youth adhere to the normative trajectory of irritability (i.e., decreasing over time or low and stable), shifts in levels of irritability have been reported across early adolescence (Yu et al., 2023). Two studies to our knowledge have focused on irritability trajectories across the transition to young adulthood and found that irritability is typically stable over this time period (Caprara et al., 2007; Leadbeater and Homel, 2015). Ultimately, there is a dearth of research examining the development of irritability across the full span of adolescence despite that there may be important shifts in levels of irritability during this period (Leibenluft et al., 2023; Yu et al., 2023). Moreover, irritability is a symptom of several psychiatric disorders that emerge or increase in adolescence, which is also when problems often cascade into widespread and lasting difficulties in adulthood (Costello et al., 2011; Paus et al., 2008). Adolescence is further characterized by significant biological, physical, social, and affective changes that may contribute to and also be impacted by changes in irritability across this developmental period (Beauchaine and Tackett, 2020; Steinberg and Morris, 2001). Thus, further investigation is needed to characterize irritability trajectories in adolescence to inform interventions that prevent the negative sequalae of outcomes.

In addition, it is crucial to explore the psychopathological and functioning outcomes of trajectories of adolescent irritability. Again, however, research on these outcomes during this vulnerable period is limited. Existing studies, often in children, suggest associations between high or increasing irritability and later diagnoses of depression, disruptive behavior disorders, and attention-deficit/hyperactivity disorder (Ezpeleta et al., 2016; Orri et al., 2019; Pagliaccio et al., 2018; Riglin et al., 2019). There is less consistency regarding anxiety disorders, with some studies showing an increased likelihood and others showing no association (Ezpeleta et al., 2016; Orri et al., 2019; Pagliaccio et al., 2018; Riglin et al., 2019). Moreover, Orri et al. (2019) found that children with high irritability and heightened risk of depressive symptoms at 13 years were more prone to suicidality at 15 to 17 years.

Ensuring consistent measurement of constructs, such as irritability, across time is crucial for developmental research. Without establishing measurement invariance, developmental differences might be attributed to measurement variation rather than true age-related changes (Maciejewski et al., 2017; Widaman et al., 2010). It cannot be assumed that using the same scale at different time points captures the same construct (Widaman et al., 2010). Despite its importance, only a few studies have examined measurement invariance of the most commonly used irritability measures, and findings have varied depending on who the reporter is. For instance, strong measurement invariance was established for parent-reported irritability on the Affective Reactivity Index (ARI) and Child Behavior Checklist (CBCL) irritability scale irritability scale over varying time spans — spanning ages 6–21 and 8–21 across two cohorts, as well as ages 6–15, respectively. (Evans et al., 2020; Mallidi et al., 2023); however, Ezpeleta et al. (2020) found evidence for only partial metric and scalar invariance across age groups (7–11 years) when using teacher-report on the ARI. Thus far, youth-reported irritability has displayed weak invariance for individuals aged 18–25 using the ARI, Youth Self Report (YSR), and Brief Child and Family Phone Interview (BCFPI; ages 18–25) (Evans et al., 2020; Leadbeater and Homel, 2015; Mallidi et al., 2023). It is worth noting that differences between these studies, such as clinical versus community samples and varying age ranges, may impact the level of measurement invariance achieved. Some studies included samples as young as age 6 or 8 and continued into young adulthood, while others focused on narrower developmental ranges. Additionally, the work on measurement invariance has generally been separate from research examining trajectories of irritability, where establishing measurement invariance should be a prerequisite for accurately assessing developmental changes (Putnick and Bornstein, 2016).

The current study aimed to 1) assess the measurement invariance of irritability using youth self-report over a 6-year period (ages 12–18), 2) identify the stability and trajectory of irritability across adolescence, and 3) examine associations between adolescent irritability trajectories and psychopathology and functioning at age 18. Drawing on the Stony Brook Temperament Study, (Klein and Finsaas, 2017), a longitudinal study of the development of psychopathology in a community sample, we used structural equation modeling to test the measurement invariance of the ARI, one of the most widely used measures of irritability. As there is limited research on measurement invariance of self-reported irritability in this age range, specific predictions about the level of invariance were not formed (Evans et al., 2020; Ezpeleta et al., 2020; Mallidi et al., 2023). Assuming sufficient invariance, latent growth curve modeling was used to explore the trajectory of irritability from age 12–18, hypothesizing relative stability with significant variability in initial levels and change over time (Caprara et al., 2007; Leadbeater and Homel, 2015). In addition, we hypothesized that higher irritability at age 12 and greater increases over time would be associated with higher rates of depression, ADHD, and disruptive behavior disorders, as well as greater functional impairment and mental health service utilization at age 18 (Ezpeleta et al., 2016; Orri et al., 2019; Pagliaccio et al., 2018; Riglin et al., 2019). Predictions about anxiety and specific types of suicidal behaviors were unclear due to mixed and limited research on these outcomes (Caprara et al., 2007; Ezpeleta et al., 2016; Riglin et al., 2019).