Latent class analysis of anxiety and depressive symptoms of the Youth Self-Report in a general population sample of young adolescents

Abstract

This study examined whether distinct groups of young adolescents with mainly anxiety or mainly depression could be identified in a general population sample. Latent class analysis was used on self-report ratings of DSM-IV symptoms of anxiety and depressive disorders, because it was hypothesized that these ratings provide a bigger chance to identify distinct groups than parent ratings of symptoms that are poorly associated with DSM-IV. Results from exploratory and confirmatory latent class analysis showed that only very small numbers of young adolescents had mainly anxiety or mainly depressive symptoms. Instead, a five-group model fitted the data best. These five groups contained young adolescents who either had a high, intermediate, or low probability to have comorbid symptoms of anxiety and depression. It was concluded that symptoms of DSM-IV anxiety and depressive disorders co-occur in young adolescents, and that latent class analysis on items that capture also severe symptoms like suicidal thoughts are needed to derive groups with specific comorbidity patterns in a general population sample.

Introduction

This study investigated the association between anxiety and depression in early adolescence. Effective treatment methods for anxiety and depression for this age groups mainly consist of cognitive behavior therapy programs (e.g., Cartwright-Hatton, Roberts, Chitsabesan, Fothergill, & Harrington, 2004; Curry, 2001). However, most programs are aimed at anxiety or depression, which reflects the assumption that anxiety and depressive disorders represent two different diagnostic entities. Given the need for developing more effective treatment methods for anxiety and depression in children, it is important to investigate the validity of the basic inclusion criteria for such treatments. Therefore, it is important to know which comorbidity patterns of anxiety and depressive disorders exist. If it would really be the case that depression can occur without an anxiety disorder being present, or vice versa, this would underscore the need for treatments that are specific for each disorder. However, if depression and anxiety occur simultaneously in most cases, a differential treatment approach would not be necessary. This would enhance the pace of treatment development and research regarding efficacy of interventions.

Most studies showed that anxiety disorders and depressive disorders co-occur in children and adolescents (Cohen et al., 1993; Costello, Mustillo, Erkanli, Keeler, & Angold, 2003; Lewinsohn, Zinbarg, Seeley, Lewinsohn, & Sack, 1997; Verhulst, van der Ende, Ferdinand, & Kasius, 1997). The meta-analysis by Angold, Costello and Erkanli (1999) for instance, indicated that anxiety disorders and depressive disorders in children and adolescents are more strongly related to each other than to other psychiatric disorders. It has even been suggested that anxiety disorders and depressive disorders are two different manifestations of the same underlying liability (Clark & Watson, 1991). This might explain the fact that anxiety disorders in childhood are a risk factor for depressive disorders (Wittchen, Beesdo, Bittner, & Goodwin, 2003).

The studies that assessed comorbidity between anxiety and depression (e.g., Costello, Mustillo, Erkanli, Keeler, & Angold, 2003; Lewinsohn, Zinbarg, Seeley, Lewinsohn, & Sack, 1997) usually applied a categorical approach, based on the criteria of the Diagnostic and Statistical Manual of mental disorders (DSM-IV; American Psychiatric Association, 2000). In these studies, anxiety and depressive disorders were judged absent or present, by applying the clinical cutoff for a DSM-IV diagnosis. Although comorbidity levels were found in these studies, considerable numbers of individuals also fulfilled criteria of a single diagnosis of anxiety disorder or depressive disorder. However, by applying a categorical approach, comorbidity rates between anxiety and depressive disorders may have been underestimated, because individuals who almost, but not entirely, fulfilled criteria for a diagnosis did not receive this diagnosis. In other words, a child with a depressive disorder and with subthreshold symptoms of an anxiety disorder was considered as a non-comorbid case, which inflated rates of single, non-comorbid, diagnoses.

To investigate whether there are homogeneous groups of children with anxiety symptoms, with depressive symptoms, or with symptoms of both anxiety and depression, latent class analysis (LCA) can be used. LCA yields the possibility to define these groups, or classes, by the probability that specific symptoms are present. Furthermore, LCA indicates which groups of children should be formed to make optimal distinctions between children with different types or frequencies of symptoms. Thus, LCA might yield a class of children in the general population who have a high probability to be endorsed positive on depressive symptoms, but negative on anxiety symptoms. However, it is also possible that LCA does not yield classes of children with a high likelihood of having symptoms of depression but not of anxiety. This would indicate that the concept of ‘pure’ depression is not useful to distinguish children with different types of symptoms.

Wadsworth, Hudziak, Heath, and Achenbach (2001) performed LCA on parent-reported symptoms of children's anxiety and depression. They found in a general population sample and in a referred sample three different classes of individuals aged 4–18 years with respectively, high, moderate, and low levels of symptoms of both anxiety and depression. Classes with high depression levels and low anxiety levels or vice versa were not identified. This might question the validity of the concept of anxiety and depressive disorders as separate diagnostic constructs. However, this study relied on parent reports, whereas it is known that agreement between parent- and self-report ratings of anxiety and depression is generally poor (Cantwell, Lewinsohn, Rohde, & Seeley, 1997; Gould, Bird, & Jaramillo, 1993; Verhulst & van der Ende, 1991). In addition, self-report ratings yield higher levels of anxiety and depressive symptoms than parent ratings (Essau & Petermann, 2001; Stanger & Lewis, 1993). Hence, use of self-reports may yield different results.

The present study identified groups of young adolescents with specific patterns of anxiety and depressive symptoms. Exploratory and confirmatory LCA were performed on self-report data on DSM-IV symptoms of anxiety and depressive disorders that were obtained from 10- to 12-year-olds from a general population.