Research paper
Temporal dynamics and longitudinal co-occurrence of depression and different anxiety syndromes in youth: Evidence for reciprocal patterns in a 3-year prospective study

https://doi.org/10.1016/j.jad.2018.02.074Get rights and content

Highlights

  • A model describing the development of anxiety-depression co-occurrence is proposed.

  • Physical panic and depression symptoms reciprocally predicted each other over time.

  • Social anxiety and depression symptoms also predicted each other systematically.

  • Anxiety-depression co-occurrence does not develop strictly unidirectionally.

  • Results are clinically useful for predicting risk and tracking symptoms.

Abstract

Background

Depression is highly comorbid with anxiety in youth. It is frequently reported that anxiety precedes depression; however, evidence surrounding the temporal precedence of anxiety over depression is mixed. Many studies of anxiety-depression co-occurrence lump distinct forms of anxiety, obscuring information regarding trajectories of specific anxiety syndromes. This study sought to more accurately describe the development of anxiety and depression over time by moving beyond the question of temporal precedence to investigate a developmentally dynamic model of anxiety-depression co-occurrence.

Methods

A community sample of 665 youth (M= 11.8, SD= 2.4; 55% female) completed repeated self-report measures of depression and anxiety (social, physical, and separation anxiety) over a 3-year longitudinal study. Prospective associations between distinct syndromes of anxiety with depression were analyzed using an autoregressive cross-lagged path model over four time points.

Results

Physical symptoms and depression symptoms reciprocally predicted each other, above and beyond the stability of either domain. Social anxiety and depression symptoms similarly predicted each other in a systematic pattern.

Limitations

Our study is limited in its generalizability to other forms of anxiety, like worry. Additional research is needed to determine whether similar patterns exist in clinical populations, and whether these processes maintain symptoms once they reach diagnostic levels.

Conclusions

The development of syndromes of depression, physical, and social anxiety during childhood and adolescence occurs in a predictable, systematic reciprocal pattern, rather than sequentially and unidirectionally (i.e., anxiety syndromes precede depression). Results are clinically useful for predicting risk for disorder, and demonstrate the necessity of tracking symptom levels across domains.

Introduction

It is well established that depression is highly comorbid with anxiety during childhood and adolescence (Brady and Kendall, 1992). Estimates suggest that half of depressed individuals have a comorbid anxiety disorder (Kessler et al., 2003). Comorbidity is associated with greater symptom and diagnostic severity, and worse social functioning (Queen and Ehrenreich-May, 2014). It has been frequently reported that anxiety precedes depression (e.g., Cole et al., 1998; Keenan et al., 2009; Avenevoli et al., 2001).

However, there are significant, unresolved questions regarding the temporal and longitudinal relations between anxiety and depression. While it has been established that youth who experience symptoms of anxiety are at heightened risk for later experiencing symptoms of depression (Kessler et al., 2005, Merikangas et al., 2010, Rohde et al., 1991, Cole et al., 1998), other research shows that those who experience symptoms of depression are similarly at greater risk for later symptoms of anxiety (Moffitt et al., 2007, Pine et al., 1998, Merikangas et al., 2003). To date, most research has focused on a “which comes first” question regarding diagnostic patterns of depression and anxiety comorbidity. More recently, researchers have suggested that the temporal ordering between anxiety and depression may not be as straightforward as has been commonly presumed (Cummings et al., 2014, Moffitt et al., 2007). The present study sought to advance knowledge on the temporal patterning of different forms of anxiety and depression by moving beyond the question of temporal precedence (i.e., “which comes first”) to investigate a more developmentally dynamic model of anxiety-depression co-occurrence over time.

An important issue that may obscure meaningful findings in the study of anxiety-depression co-occurrence is the lumping together of distinct forms of anxiety. In their “multiple pathways” model of anxiety-depression co-occurrence, Cummings and colleagues highlight the need for empirical work that does not treat anxiety as a single, isomorphic class of symptoms and disorder (Cummings et al., 2014). Anxiety disorders and their core defining features (e.g., generalized anxiety, social anxiety, panic, separation anxiety) are heterogeneous in their age of onset, longitudinal trajectories (Copeland et al., 2014, Kessler et al., 2005), and symptom expression (Weems and Costa, 2005). Moreover, the rate of co-occurrence of these heterogeneous forms of anxiety with depression and other psychopathologies varies (Hankin et al., 2016, Angold et al., 1999). Further, the degree of co-occurrence of different forms of anxiety with depression may vary by age and gender, based on the prevalence of each syndrome across childhood and adolescence for boys and girls (Cummings et al., 2014).

The current study longitudinally examined the reciprocal associations of depressive symptoms with different anxiety syndromes. Our main purpose was to investigate whether particular anxiety syndromes predict, and are predicted by, depression, in a systematic manner in youth, as theoretical work has emphasized the importance of examining interactive processes resulting in anxiety-depression co-occurrence (e.g., Cummings et al., 2014). A secondary goal was to examine whether developmental patterns of anxiety and depression co-occurrence are moderated by age and/or gender. To explore these descriptive questions, we assessed depression and different anxiety syndromes (social anxiety, panic, separation anxiety) using a multi-wave prospective design over a three-year follow-up across different age cohorts with community youth.1 This study aims to contribute to a more nuanced, accurate picture of anxiety-depression co-occurrence over time in childhood and adolescence than the study of temporal precedence alone. By establishing a descriptive pattern of how different forms of anxiety relate to depression among youth, we can better understand processes that may explain why particular anxiety syndromes predict, and are predicted by, depression over time in a systematic manner.

A large body of research has investigated the temporal patterns of different forms of anxiety with depression (Cummings et al., 2014) with the goal of addressing the question of “which comes first.” Most prior work finds that anxiety, when broadly conceptualized and measured, predicts depression (e.g., Kouros et al., 2013; Kessler et al., 2005; Merikangas et al., 2010; Rohde et al., 1991; Cole et al., 1998). Much of this work combines multiple anxiety syndromes into a broad “anxiety symptoms” scale. However, this conclusion that anxiety precedes depression may depend on the specific anxiety syndromes examined as predictors of depression. We focus here on the association of syndromes of social anxiety, panic, and separation anxiety with depression. While much of the prior literature on anxiety-depression co-occurrence focuses on comorbidity at the disorder level, we focus on syndromes, as predictors of disorders (Gerhardt et al., 1999). Existing models further emphasize the importance of studying anxiety-depression co-occurrence at the symptom- or syndrome-level, as youth with subclinical levels of these syndromes experience significant distress and impairment (Epkins and Heckler, 2011).

Social anxiety and depression are related during childhood and adolescence (Epkins and Heckler, 2011, Hamilton et al., 2016, Ranta et al., 2009); however, their temporal association has been debated. Cummings et al. (2014) proposed that social anxiety may predict depression, and vice versa, as impairment in one domain results in an increase in symptoms in the other domain. Many studies show that social anxiety predicts depression across the lifespan (e.g., Aune and Stiles, 2009; Last et al., 1992; Beesdo et al., 2007). Others find that depression predicts social anxiety during adolescence, but that social anxiety did not predict depression (Hamilton et al., 2016). However, most of these studies have not repeatedly assessed social anxiety and depression with more than three time points. With three time points or fewer, it cannot be determined whether prediction from one symptom domain to another happened only once, or whether both temporal sequences exist as part of a larger pattern of mutually reinforcing symptom escalation (i.e., social anxiety predicts, and is predicted by, depression over time). This possibility has yet to be examined empirically.

Studies of adults find that panic is a risk factor for depression (Horn and Wuyek, 2010). However, National Comorbidity Survey data (Kessler et al., 1998) show of those individuals who reported both depressive episodes and panic attacks, 43.4% reported their first panic attack occurred before their first depressive episode, whereas 31.1% reported the reverse sequence. A review similarly found that panic precedes depression in one third of adults, depression precedes panic in another third, and panic and depression develop concurrently in the last third (Johnson and Lydiard, 1998). As panic has a later age of onset than social and separation anxiety, it may be more likely to temporally follow depression than either social or separation anxiety (Lewinsohn et al., 1997). Thus, the precise temporal patterning between panic and depression symptoms is not well understood, especially among youth.

Most research has focused on separation anxiety as a risk factor for depression, and has found separation anxiety precedes depression (e.g., Lewinsohn et al., 1997). The median age of onset of separation anxiety disorder is younger than that of social anxiety and panic disorder across samples (Kessler et al., 2005, Last et al., 1992). Therefore, the temporal patterning of separation anxiety and depression may differ from that of other forms of anxiety with depression, such that separation anxiety may be most likely to precede depression symptoms. The heterogeneous ages of onset for different anxiety syndromes, combined with different temporal patternings of particular anxiety manifestations with depression, support the use of an analytical approach that can parse the independent associations of specific forms of anxiety with depression over time.

Age and gender may moderate the temporal associations of anxiety and depression. Depression and anxiety are more prevalent among girls than boys (Hankin et al., 1998, Costello et al., 1996, Axelson and Birmaher, 2001), and anxiety-depression co-occurrence is more common in girls (Cummings et al., 2014). Research finds greater heterotypic continuity from anxiety to depression, and vice-versa, for girls than for boys (Costello et al., 2003, McGee et al., 1992). However, findings have been equivocal on whether gender moderates the development of anxiety and depression symptoms. Some studies have found stronger associations between anxiety and depression for girls over time (Merikangas et al., 2003), whereas others have not found that gender moderates developmental trajectories of anxiety and depression (McLaughlin and King, 2015, Hamilton et al., 2016). Relatedly, there are well-established age differences in both depression and certain forms of anxiety (Kessler et al., 2005), which have been proposed to lead to different patterns of comorbidity by age (Cummings et al., 2014). Few studies have examined whether age moderates trajectories of anxiety and depression across time.

Much of the existing research has been limited to two- or three-point prospective designs (e.g., Avenevoli et al., 2001; Keenan et al., 2008; Pine et al., 1998), or to retrospective studies relying on participants’ recall of anxiety and depressive symptoms, which underestimate comorbidity rates (Angold et al., 1999, Moffitt et al., 2010). These designs were not structured to test for reciprocal associations in the developmental interplay of anxiety and depression. Important methodological considerations relevant to this issue can be drawn from the cascade model literature (e.g., Masten et al., 2005; Cole and Maxwell, 2003). Cascades refer to the process by which symptoms from one domain spill over to other domains, resulting in elevated symptom levels across domains over time (Masten and Cicchetti, 2010). Consistent with the cascade model literature, accurate delineation of the descriptive patterning of the temporal associations between different anxiety syndromes and depression benefits from a minimum of four time points with adequate spacing between follow-ups. This allows for change to occur in symptoms, and enables reliable prediction of co-occurring internalizing syndromes after controlling for the autoregressive stability of a particular internalizing syndrome (e.g., social anxiety) over time (Cole and Maxwell, 2003, Burt et al., 2008). The current study uses a multi-wave prospective design in which anxiety syndromes and depression were assessed across three years, and the repeated measures data analyzed across four assessment waves.

This study used data from a prospective longitudinal study of 3rd, 6th, and 9th graders recruited from the community to examine and more accurately describe the co-occurring temporal patternings and reciprocal effects of different anxiety syndromes and depression. We move beyond the question of “which came first” and examine the possibility of reciprocal effects in anxiety syndromes and depression to investigate how different anxiety syndromes and depression temporally relate to each other. Especially important to accurately delineating their longitudinal co-occurrence is distinguishing between syndromes of anxiety, given the established heterogeneity in anxiety. We also tested for moderation of the model by age and gender.

Section snippets

Participants

Youth were recruited at the University of Denver and Rutgers University. Informational letters were sent home to families with a child in 3rd, 6th, or 9th grades in the broader Denver and central New Jersey areas. Of the families to whom letters were sent, 1108 parents called the laboratory for more information. Parent report established that the parent and child were fluent in English, the child did not have an autism spectrum disorder, psychotic disorder, or intellectual disability. Of the

Descriptive statistics

The means and standard deviations of each anxiety syndrome and depressive symptoms, separated by gender, are reported in Table 1. Mean levels of physical symptoms, social anxiety, separation anxiety, and depression symptoms were within the range typically observed in community samples of youth of these ages. The pattern of gender differences in our sample was consistent with prior literature on symptom levels for girls and boys across development. Girls reported higher mean levels of depression

Discussion

The temporal patterning of the development of co-occurring syndromes of anxiety and depression during childhood and adolescence has long been a topic of debate. The current study found physical panic symptoms and depression symptoms, as well as social anxiety and depression symptoms, longitudinally and reciprocally predicted each other over the three-year prospective follow-up, above and beyond the stability of either domain from one time point to the next. The results of the current study

Acknowledgements

The authors appreciate the thoughtful feedback provided by Tina Schweizer, Dustin Haraden, Fatema Jivanjee, Julianne Griffith, and Elissa Hamlat on earlier drafts of this manuscript.

Role of the funding source

The research reported in this article was supported by grants from the National Institute of Mental Health to Benjamin L Hankin, R01MH077195, R01MH105501, R21MH102210, MH109662, and to Jami F. Young, MH077178.

References (66)

  • P. Muris et al.

    Three traditional and three new childhood anxiety questionnaires: their reliability and validity in a normal adolescent sample

    Behav. Res. Ther.

    (2002)
  • C.F. Weems et al.

    Developmental differences in the expression of childhood anxiety symptoms and fears

    J. Am. Acad. Child Adolesc. Psychiatry

    (2005)
  • A. Angold et al.

    Comorbidity

    J. Child Psychol. Psychiatry

    (1999)
  • T. Aune et al.

    The effects of depression and stressful life events on the development and maintenance of syndromal social anxiety: sex and age differences

    J. Clin. Child Adolesc. Psychol.

    (2009)
  • S. Avenevoli et al.

    Comorbidity of depression in children and adolescents: models and evidence from a prospective high-risk family study

    Biol. Psychiatry

    (2001)
  • D.A. Axelson et al.

    Relation between anxiety and depressive disorders in childhood and adolescence

    Depress. Anxiety

    (2001)
  • K. Beesdo et al.

    Incidence of social anxiety disorder and the consistent risk for secondary depression in the first three decades of life

    Arch. Gen. Psychiatry

    (2007)
  • E.U. Brady et al.

    Comorbidity of anxiety and depression in children and adolescents

    Psychol. Bull.

    (1992)
  • K.B. Burt et al.

    The interplay of social competence and psychopathology over 20 years: testing transactional and cascade models

    Child Dev.

    (2008)
  • M.S. Caldwell et al.

    Reciprocal influences among relational self-views, social disengagement, and peer stress during early adolescence

    Child Dev.

    (2004)
  • D.A. Cole et al.

    Testing meditational models with longitudinal data: questions and tips in the use of structural equation modeling

    J. Abnorm. Psychol.

    (2003)
  • D.A. Cole et al.

    A longitudinal look at the relation between depression and anxiety in children and adolescents

    J. Consult. Clin. Psychol.

    (1998)
  • C.C. Conway et al.

    Core dimensions of anxiety and depression change independently during adolescence

    J. Abnorm. Psychol.

    (2017)
  • E.J. Costello et al.

    The Great Smoky Mountains Study of youth: goals, designs, methods, and the prevalence of DSM-III-R disorders

    Arch. Gen. Psychiatry

    (1996)
  • E.J. Costello et al.

    Prevalence and development of psychiatric disorders in childhood and adolescence

    Arch. Gen. Psychiatry

    (2003)
  • C.M. Cummings et al.

    Comorbidity of anxiety and depression in children and adolescents: 20 years after

    Psychol. Bull.

    (2014)
  • C.C. Epkins et al.

    Integrating etiological models of social anxiety and depression in youth: evidence for a cumulative interpersonal risk model

    Clin. Child Fam. Psychol. Rev.

    (2011)
  • C.A. Gerhardt et al.

    Association of a mixed anxiety-depression syndrome and symptoms of major depressive disorder during adolescence

    J. Youth Adolesc.

    (1999)
  • J.L. Hamilton et al.

    Cognitive vulnerabilities as predictors of stress generation in early adolescence: pathway to depressive symptoms

    J. Abnorm. Child Psychol.

    (2013)
  • J.L. Hamilton et al.

    The temporal sequence of social anxiety and depressive symptoms following interpersonal stressors during adolescence

    J. Abnorm. Child Psychol.

    (2016)
  • C. Hammen

    Generation of stress in the course of unipolar depression

    J. Abnorm. Psychol.

    (1991)
  • B.L. Hankin et al.

    Development of gender differences in depression: an elaborated cognitive vulnerability-transactional stress theory

    Psychol. Bull.

    (2001)
  • B.L. Hankin et al.

    Development of depression from preadolescence to young adulthood: emerging gender differences in a 10-year longitudinal study

    J. Abnorm. Psychol.

    (1998)
  • Cited by (25)

    • Effects of personalized depression prevention on anxiety through 18-month follow-up: A randomized controlled trial

      2022, Behaviour Research and Therapy
      Citation Excerpt :

      Similarly, interpersonal difficulties with parents and peers have been shown to be risk factors for both youth anxiety and depression (La Greca & Harrison, 2005; Yap & Jorm, 2015). Furthermore, there is evidence that depression and anxiety symptoms prospectively and reciprocally predict each other, suggesting that one may be a risk factor for the other (Long et al., 2018). Interventions targeting shared risk factors may have beneficial effects on both depression and anxiety, reducing the population burden associated with each.

    • Internalized psychopathology dimensions in middle childhood: Cross-sectional and temporal associations

      2020, Journal of Anxiety Disorders
      Citation Excerpt :

      These results are consistent with previous research showing that proneness for anxiety is a prospective predictor of pure depressive disorders, pure anxiety disorders and comorbid depression and anxiety (e.g., Zinbarg et al., 2016). Furthermore, a vast amount of literature has examined the temporal sequencing of depression and anxiety, and it has been widely acknowledged that anxiety precedes depression (Keenan, Feng, Hipwell, & Klostermann, 2009), although recent studies have shown that the temporal ordering between anxiety and depression may be more complicated (Long, Young, & Hankin, 2018). It is worth mentioning that there appeared to be some overlap between the degree of centrality of a node in the cross-sectional networks and to which degree this same node/variable was related to change in other variables temporally using LGCM.

    • Separating within-person from between-person effects in the longitudinal co-occurrence of depression and different anxiety syndromes in youth

      2019, Journal of Research in Personality
      Citation Excerpt :

      Therefore, the common practice of collapsing across these different disorders when examining anxiety-depression comorbidity may obscure meaningful differences in the associations of distinct anxiety disorders with depression over time. In a recent study, we simultaneously modeled the longitudinal, dynamic associations of social anxiety, physical symptoms of anxiety, and separation anxiety with depression at four time points over a 3-year prospective period using a traditional autoregressive cross-lagged panel model to describe the temporal patterns of the co-occurrence of distinct forms of anxiety with depression in a large community sample of children and adolescents (Long, Young, & Hankin, 2018). Results indicated that social anxiety and depression, as well as physical symptoms of anxiety and depression, uniquely predicted each other in a series of bidirectional associations, above and beyond the expected stability of these constructs over time.

    View all citing articles on Scopus
    View full text