تمایز بین اضطراب جدایی و اضطراب اجتماعی در کودکان و نوجوانان

Separation anxiety and social phobia are intertwined to a considerable degree, and high comorbidity rates have been reported. The present study used latent class analysis (LCA) to investigate if classes of children and adolescents with—simultaneously—high rates of separation anxiety and low rates of social anxiety symptoms, or vice versa, could be identified. Eight- to 18-year-olds from a large general population (n=1000) and referred sample (n=735) were assessed with the Multidimensional Anxiety Scale for Children (MASC). With LCA, a separate class of referred 8–11-year-old children with high separation anxiety scores, and simultaneously lower social anxiety scores was identified, next to a class of children with high scores on separation anxiety and social anxiety. In the other groups (referred 12–18-year-olds and children and adolescents from the general population), a class with individuals who specifically scored high on separation anxiety could not be revealed. The results indicated that separation anxiety represents a different construct than social anxiety in referred children (but not in referred adolescents or in the general population). It can be concluded that, in referred children, research regarding etiology and treatment outcome of anxiety symptoms should be aimed specifically at separation anxiety and social anxiety, instead of just investigating a broader anxiety dimension.

Given the high prevalence rates of separation anxiety disorder and social phobia in children and adolescents (Costello et al., 1996; Essau, Conradt, & Petermann, 2000; Verhulst, van der Ende, Ferdinand, & Kasius, 1997)—lifetime prevalence rates are often estimated at 4% and 13%, respectively (e.g. Compton, Nelson, & March, 2000)—their association with impaired functioning (Canino et al., 2004; McGee & Stanton, 1990; Verhulst et al., 1997), and the tendency of childhood anxiety to persist into adulthood (Ferdinand & Verhulst, 1995; Ferdinand, Verhulst, & Wiznitzer, 1995; Pine, Cohen, Gurley, Brook, & Ma, 1998), refinement of the taxonomy of these disorders is important. With the introduction of DSM-III (American Psychiatric Association, 1980), separation anxiety disorder and social phobia were defined as distinct disorders, and the delineation between the two has survived DSM-III-R, DSM-IV, and DSM-IV-TR (American Psychiatric Association (1987), American Psychiatric (1994) and American Psychiatric Association (2000)). This is remarkable, given the fact that separation anxiety and social phobia seem to be intertwined to a considerable degree. High comorbidity rates have been reported by many authors (Essau et al., 2000; Masi, Mucci, Favilla, Romano, & Poli, 1999; Newman et al., 1996; Verduin & Kendall, 2003). Furthermore, whereas one of the aims of assigning diagnoses is to define optimal treatment possibilities, it is well known that it does not matter much in clinical practice if a child or adolescent receives a diagnosis of social phobia or separation anxiety disorder. For both disorders cognitive behavior therapy is the treatment of first choice, followed by SSRI's (Dadds & Barrett, 2001; Ferdinand, Barrett, & Dadds, 2004; Research Unit on Pediatric Psychopharmacology Anxiety Study Group, 2001; Walkup et al., 2002). If separation anxiety and social anxiety would erroneously be considered as two separate problem dimensions, this would hamper research regarding etiology, treatment outcome, and prevention. For instance, trials specifically aimed at studying treatment outcome of social phobia might not be needed and instead, individuals with separation anxiety disorder could also be included. Given the problems with recruitment of a large enough number of patients for clinical trials, this would be important. Previous studies have found evidence for the presence of separate diagnostic constructs of separation anxiety and social phobia. By performing factor-analyses, several authors found that symptoms of separation anxiety represented a dimension that could be separated from a social anxiety dimension (Chorpita, Yim, Moffitt, Umemoto, & Francis, 2000; Muris, Schmidt, Engelbrecht, & Perold, 2002; Spence, 1997). However, this does still not imply that these two dimensions represent completely distinct groups of children. Factor analysis yields information about which symptoms co-occur frequently, but does not indicate which homogeneous groups of children with similar types of anxiety symptoms can be identified. The categorical nature of our main taxonomic system, DSM-IV (American Psychiatric Association, 1994), often seems to hamper efforts to classify children according to their symptom patterns. For instance, one child may receive a diagnosis of Social Phobia, and another child of social phobia (SoPH) plus separation anxiety disorder (SAD), whereas the two children may still be very much alike. The first child may have two, and the second child may have three symptoms of separation anxiety disorder, which yields a diagnosis for the second, but not for the first child. Hence, these 2 children may be very similar, and will probably even receive the same treatment, but still may receive different diagnoses. If children who have SoPH and SAD symptoms are classified as SoPH, but not as SAD, because the SAD symptoms do not exceed the threshold for a diagnosis, comorbidity between different types of anxiety may be deflated artificially. This example shows that it is important to study the boundaries between possibly different clinical phenomena without using an a priori categorical approach to diagnostic data, which may result in a loss of important statistical information. To investigate which homogeneous groups of children with similar anxiety symptoms can be identified, latent class analysis (LCA) can be conducted. 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, for instance, LCA might yield a class of children who have high probabilities of SAD symptoms, but low probabilities of SoPh symptoms, indicating a group of children with pure SAD. However, it is also possible that LCA only yields a class of children with high probabilities of SAD and SoPh symptoms, and a class of children with low probabilities of SAD and SoPh, indicating high comorbidity. In this case, the concept of ‘pure’ SAD would not be useful to distinguish children with different types of anxiety symptoms. The aim of the present study was to investigate which classes of children with separation anxiety or social anxiety can be found in a general population sample and a clinical sample. Because studying a general population sample has the disadvantage that individuals with high symptom levels are scarce, whereas the disadvantage of a clinical sample is that comorbidity rates may be inflated by referral biases, we studied both a general population sample and a clinical sample. Given the high comorbidity rates between SAD and SoPh, we hypothesized that classes of children with a high probability to have symptoms of separation anxiety disorder, and simultaneously low probabilities to have symptoms of social phobia, or vice versa, cannot be identified. Instead, we expected to find classes of children that differed in the number of anxiety symptoms rated present, but that did not differ in the type of anxiety. Latent class analyses were performed on symptom ratings of separation anxiety and social anxiety that were obtained with a self-report questionnaire from 8–18-year-olds from a general population sample and a clinical sample. It was assessed if different types of classes were present in the clinical versus the general population sample. Because age may influence comorbidity rates (Compton et al., 2000), age effects were assessed as well.