مسیرهای رشد اضطراب دوران کودکی: شناخت تداوم و تغییر در احساسات اضطراب

This paper outlines a way for thinking about continuity and change in childhood anxiety symptoms. Considerations for a model of continuity and change in anxious emotion are discussed first. Then, a perspective which may resolve inconsistencies across studies on the stability of childhood anxiety problems overtime is presented. The perspective views dysregulation of the anxiety response system and distress/negative affect as the core primary and necessary features of maladaptive anxious emotion and views the disorder-specific symptoms (generalized anxiety disorder, social phobia, separation anxiety, panic disorder, and simple phobia) as secondary characteristics of maladaptive anxious emotion. The perspective emphasizes ordered complexity in the developmental expression of anxious emotion, and delineates expectations for continuity and change in the features of anxious emotion by describing major normative trajectories across childhood and positing multiple sub-trajectories.

Anxiety disorders in childhood and adolescence are highly prevalent and can cause intense psychosocial impairment (Silverman & Treffers, 2001). Childhood anxiety disorders, if left untreated, predict increased risk for mental disorders and substance use problems later in life (Kendall et al., 2004 and Pine et al., 1998) and are associated with negative cognitive, neuro-developmental, and hormonal outcomes (Carrión et al., 2007 and De Bellis et al., 1999). The field has advanced considerably with demonstrated efficacy of cognitive-behavioral and pharmacological strategies in the treatment of childhood anxiety and phobic disorders (see Albano and Kendall, 2002, Creswell and Cartwright-Hatton, 2007, In-Albon and Schneider, 2007, Kendall, 1994, Silverman et al., 1999 and Walkup et al., 2002). There have also been advances in knowledge about the processes and mechanisms responsible for childhood anxiety problems as well as the processes that ameliorate these problems (i.e., the how and why). However, the field continues to lack a truly comprehensive theory of the developmental psychopathology of childhood anxiety. An important step in building such a theory would be a description of continuity and change in anxious emotion followed by multidisciplinary research elucidating the factors responsible for continuity and change. The purpose of this paper is to provide an empirically supported concise model that describes continuity and change in anxious emotion across childhood (i.e., school-aged years).1 The field of child and adolescent anxiety problems in recent years has consistently moved toward understanding anxiety problems by focusing on the specific anxiety disorders (e.g., separation anxiety disorder, social anxiety, generalized anxiety disorder, etc.) as delineated in the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV, American Psychiatric Association, 1994). Although the system has produced important advances in knowledge, researchers continue to question the validity of the anxiety disorder diagnoses for the childhood years (see Costello et al., 2004, Curry et al., 2004, Dadds et al., 2004, Saavedra and Silverman, 2001 and Weems and Stickle, 2005). For instance, research suggests that the DSM-defined disorders have incredibly high rates of comorbidity, that the only variables that distinguish them are definitional (e.g., more intense worry in generalized anxiety disorders, more social concerns in social anxiety disorder—making discrimination largely tautological), and that with the possible exception of obsessive compulsive disorder and posttraumatic stress disorder 2 they do not predict differential treatment outcomes (see Berman et al., 2000, Costello et al., 2004, Curry et al., 2004, Dadds et al., 2004, Saavedra and Silverman, 2001 and Weems, 2005). Thus, the diagnoses may not represent ideal outcome variables when trying to understand continuity and change in anxious emotion. Alternatively, another focus has been on continuous measures of broader constructs such as internalizing symptoms (e.g., Child Behavior Checklist, Achenbach, 1991); however, lumping symptoms together using continuous assessment may also be limited with regard to understanding continuity and change in anxious emotion because there may be different trajectories for different types of symptoms. In the following sections, a perspective is presented that describes continuity and change in anxious emotion among school-aged youth (approximately age 6–18 years). I use the phrase “maladaptive anxious emotion” as a way to discuss problematic or interfering anxiety, thus implying that there is a core set of features that underly the DSM anxiety disorders that nonetheless ontologically transcend the DSM definitions of what a specific anxiety disorder is (see Moses and Barlow, 2006 and Watson, 2005 for related discussion). Developmental considerations for a model of continuity and change in anxious emotion Perhaps the biggest limitation of using DSM anxiety disorder diagnoses for developing a developmental theory is that longitudinal research on the stability of childhood anxiety disorders has produced highly inconsistent results (e.g., Last et al., 1996 and Newman et al., 1996). Prospective longitudinal studies of childhood anxiety disorders have reported estimates of stability from 4% to 80% (e.g., Keller et al., 1992, Last et al., 1996 and Newman et al., 1996). These studies may show wide variability for many reasons (e.g., the type of disorder, the informant, the sample, the method of assessment, or the amount of time that has passed between the initial evaluation and the follow-up 3). Interestingly, however, studies have shown similarly wide estimates even for the same anxiety disorder across similar time frames using similar methodology. For example, Last et al. (1996) reported that 13.6% of youth with social phobia retained the diagnosis after 3–4 years, whereas Newman et al. (1996) reported that 79.3% of youth with social phobia retained the diagnosis after 3–10 years. The main difference between these studies was the age of participants (starting as young as in 5 years old in Last et al. versus 11–21 in Newman et al.). An important reason for the inconsistencies or seemingly unstable pattern of DSM anxiety disorder diagnoses may be due to a limited appreciation of age-related developmental differences in the symptom expression of childhood anxiety and phobic disorders. Research and theory on normative emotional development suggests specific age differences in the predominant expression of the symptoms of childhood anxiety and phobic disorders (Warren and Sroufe, 2004 and Westenberg et al., 2001). For example, Westenberg and colleagues have utilized developmental theories of personality development (e.g., Loevinger, 1976) to show how the predominant expression of anxious and phobic symptoms may be tied to normative developmental periods and challenges. Drawing together Westenberg et al. (2001) and Warren and Sroufe’s (2004) theorizing regarding the sequence and timing of symptom expression for school-aged youth (aged 6–17 years) one can predict, for instance, that separation anxiety symptoms and animal fears will be predominant in youth around ages 6–9 years, generalized anxiety symptoms and fears concerning danger and death in youth 10–13 years, and social anxiety symptoms and social/performance related fears in adolescents around age 14–17 years. Epidemiological data on the age of onset of anxiety disorder diagnoses is generally consistent with the developmental predictions of age related differences (see Costello et al., 2004 for a review of age of onset findings). Research in clinical samples also suggests that separation anxiety disorder is more common in children while social phobia is more common in adolescents. For instance, Weems, Hammond-Laurence, Silverman, and Ginsburg (1998) reported differences in the distribution of anxiety disorders with separation anxiety disorder more common in children (aged 6–11 years) while social phobia was more common in older youth (aged 12–17 years) in a sample of 280 youth who met diagnostic criteria for anxiety disorders. Research examining specific anxious symptoms and fears dimensionally across age ranges also support the idea of sequential developmental differences in the expression of specific fears and anxiety symptoms. For example, research by Ollendick and colleagues (e.g., Ollendick et al., 1989 and Ollendick et al., 1985) in normative samples (in the US and internationally) showed higher rates of animal fears in children as compared to adolescents, whereas Poulton, Trainor, Stanton, and McGee (1997) found social fears to be more common in adolescents than children. In terms of a priori tests of the developmental hypothesis, Westenberg, Siebelink, Warmenhoven, and Treffers (1999) have reported that separation anxiety disorder developmentally precedes overanxious disorder. Westenberg, Drewes, Siebelink, and Treffers (2004) found that child-reported fears of physical danger and punishment decrease with age and that fears of social and achievement evaluation increase with age when controlling for overall fears. Weems and Costa (2005) also provided a test of the above developmental theory and results pointed toward specific symptoms predominant at certain ages (i.e., separation anxiety symptoms in youth 6–9 years, death and danger fears in youth 10–13 years, and social anxiety symptoms, as well as failure and criticism fears in youth 14–17 years) when controlling for overall trends in anxiety disorder symptoms. Recently, Westenberg, Gullone, Bokhorst, Heyne, and King (2007) provided a longitudinal and cross sectional test of the normative developmental pattern predicted for social evaluation fears and confirmed that social evaluation fear increased between childhood and adolescence. Individual differences in social evaluation fear displayed modest stability over a 3-year follow-up period and the stability of social evaluation fear was stronger in older youth. Taken together, the findings to date suggest that models of continuity and change in childhood anxious emotion should consider normative differences in the developmental expression of childhood fears and anxiety symptoms. In other words inconsistency in longitudinal findings using DSM anxiety disorder diagnosis may result in part from developmental differences in expression and not a complete lack of continuity (a concept that has been termed “heterotypic continuity” and has been previously applied to conduct problems, see Moffitt, 1993). Drawing from the above considerations, inconsistencies in the literature might actually reflect, at least in part, normative developmental differences. Such findings point to the importance of including age-related differences in a model of continuity and change. For example, above I noted that Last et al. (1996) reported that only around 14% of youth with a primary social anxiety disorder still had that same diagnosis 3–4 years later [range across the disorders 4.3% (separation anxiety disorder) to 30.8% (Specific Phobia) stable overtime]. However, 20% retained one or more of their initial comorbid anxiety disorder diagnoses and 30% developed another DSM disorder, including 16% of the sample who developed a different anxiety disorder. Thus, maladaptive anxious emotion may be more stable than are particular DSM disorders in childhood. In addition, if there are systematic developmental differences in the type of symptoms that are expressed then using global measures of internalizing symptoms or focusing solely on core features of anxious emotion might also be limiting, because lumping, for instance, social, separation, and mortality fears and symptoms together could obscure real developmental change. In one of the most comprehensive longitudinal examinations of anxiety symptoms across childhood through adolescence, Bosquet and Egeland (2006) found only low (e.g., r = .27) stability of anxiety across the childhood assessment and a moderate correlation (r = .43) between the 16-year-old and 17.5-year-old anxiety assessment. However, these researchers used items from the Child Behavior Checklist (CBCL, Achenbach, 1991) and did not distinguish amongst types of anxiety symptoms. If one set of symptoms is on average increasing (e.g., social anxiety) and another set is on average decreasing (e.g., separation anxiety), stability estimates will be affected at different ages. Thus, in describing continuity and change it is also important to identify the separate components or features of anxious emotion that are important to consider.

Viewing the DSM symptoms of anxiety disorders as secondary features of anxiety-disordered emotion shows that there are developmental consistencies among the apparent inconsistencies in anxiety disorder stability in youth. There may be a relatively larger degree of continuity in the core features of anxious emotion and there appears to be consistency in the trends of the core and secondary features overtime. By describing expected continuity and change, this perspective points to the need to identify the specific processes that contribute to continuity and change by focusing on trajectories in the core features of anxiety-disordered emotion and that shape the specific expression and trajectories of the secondary features. The perspective draws attention to developmental differences in stability and symptom expression. For example, maladaptive anxious emotion may be much more stable than are particular DSM disorders in childhood. Similarly, the perspective draws attention to the need to appreciate the specific features of anxious emotion that are being assessed in longitudinal studies of anxiety. Research has identified biological (e.g., genetics, temperament, psychophysiology), behavioral (e.g., operant and respondent learning models), cognitive (e.g., information processing, stimuli/event interpretation), interpersonal (e.g., attachment theory, parent–child relationship), and contextual (e.g., family, home, school, and community) processes important to understanding the origins of childhood anxiety (see Essau and Peterman, 2002, Vasey and Dadds, 2001 and Weems and Stickle, 2005). Research aimed at determining the extent to which these factors shape continuity and change in the core features and shape the expression of secondary features of anxiety-disordered emotion will facilitate the development of a comprehensive theory. It is hoped that the perspective presented here will encourage research to focus on trajectories in anxious emotion overtime. The hope is to turn descriptions of trajectories into explanations and systematic understanding of the developmental pathways to chronic maladaptive anxious emotion, recovered anxious children, etc. However, there is a need to continue to test the adequacy of the model to simply describe continuity and change. Person-centered analyses of anxiety symptoms overtime such as hierarchical linear modeling and cluster analyses may be particularly useful in this regard as will the use of cross-sequential designs.