یک پیشنهاد جدید برای انواع هراس اجتماعی در یک نمونه از نوجوانان اسپانیایی

The present study aimed to analyze validity of a new proposal of delimitation and operationalization of the social phobia subtypes in the adolescence. The sample consisted of 971 Spanish adolescents ranged between 14 and 18 years, attending 59 high schools in rural and urban areas of the south-east of Spain. Principal component and cluster analyses were conducted. Results indicated that the principal components analysis revealed two factors, which could be categorized as interaction and performance anxiety, whereas the cluster analysis delimited four subtypes: specific social phobia, mild generalized social phobia, moderate generalized social phobia and severe generalized social phobia. Subsequently, comparisons in psychopathological and demographic variables showed statistically significant differences among the four subtypes. The emerging multidimensional structure was consistent with the differential psychopathological and demographic features of each group. The current study establishes preliminary support for a new proposal of subdivision and operationalization of the construct social phobia in adolescence.

Although Social Phobia (SP) was recognized as a clinically significant mental disorder by the American Psychiatric Association in 1980, the subtypes were not added to the Diagnostic and Statistical Manual of Mental Disorders until 1987 (DSM-III-R; APA, 1987). At present there is a relative consensus regarding the existence of at least two different subtypes of SP: Generalized Social Phobia (GSP) and Specific or circumscribed Social Phobia (SSP) ( Liebowitz, Ninan, Schneier, & Blanco, 2005). The body of research on identifying subtypes of SP has followed two different perspectives: quantitative versus qualitative. First, some researchers (e.g., Hofmann et al., 1999, McNeil, 2001 and Vriends et al., 2007) interpret the distinction between GSP and SSP as quantitative, arguing that the heterogeneity within social phobia has to be seen as a continuum of severity of social phobia. From this point of view, GSP represents the most severe form of social phobia and SSP the less severe/less interfering type. The last three editions of APA, 1987, APA, 1994 and APA, 2000 have also adopted a classification congruent with this framework. According to this classification, GSP should be diagnosed if an individual fears most social situations, that is, it is characterized by a pervasive pattern of social fears and it is thought to be the more incapacitating form of social phobia. But not even the fourth edition (DSM-IV-TR, APA, 2000) has offered a clearer operationalization for the diagnosis of the GSP subtype, given that the number of social situations required for the diagnosis of GSP is not defined. Regarding the SSP, although it is not recognized as a subtype in the DSM-IV-TR (APA, 2000) at the moment, it has frequently been referred to as one in scientific literature (Beidel & Turner, 1999; Heimberg, Holt, Schneier, Spitzer, & Liebowitz, 1993). Its delimitation has in fact been presented as a subtype, characterized by the fear of a limited number of social situations, also here without specifying a more detailed operationalization. Opposed to the quantitative perspective, the qualitative explanation focuses on the type of feared social situations (Stemberger, Turner, Beidel, & Calhoun, 1995). Individuals with SSP are thought to fear only performance-based social situations (e.g., writing in public). In comparison, people with GSP may fear performance-oriented situations but are distinguished by their fear of social situations involving interactions with others (e.g., asking someone on a date). This implies that GSP and SSP are two different behavior disorders, although they share some phenotypical similarities (McNeil, 2001). However, both of the attempts to classify the SP subtypes have been repeatedly questioned because, besides being proposed from a reductional point of view, those attempts have not permitted to generate a theoretically coherent alternative to the APA's proposal (1987) (Hofmann, Heinrichs, & Moscovitch, 2004; Hook & Valentiner, 2002). They also fail in giving rise to a more explicit operationalization of the APA proposal oriented towards the search of criteria that could provide access to a specific diagnosis and election of a specific treatment for different groups or subgroups which could be established (Ruipérez, García-Palacios, & Botella, 2002). In order to deal with these limitations, some authors have tried to examine heterogeneity of SP using factor analytic techniques or cluster analysis. In addition to SSP and GSP, at least two other subtypes have appeared. For example, in eastern cultures (e.g., Japan), “offending fear” or “Taijin Kyofusho” (Kleinknect, Dinnel, Kleinknect, Hiruma, & Harada, 1997) and “relationship fear” (Sakurai et al., 2005) appeared to be culture-specific situational domains of social phobia. Another example of SP subtype is the non-generalized social phobia, defined as a fear of social situations with at least one social situation domain not affected (Holt, Heimberg, & Hope, 1992). In Spain, Olivares and Caballo (2003) have described an explanatory proposal about the genesis, development and maintenance of social phobia, which considers the issue of social phobia subtypes from another point of view. The model postulates the existence of an association among a negative social context and a group of particular variables of the individual that facilitate the genesis of the disorder either in an acute or an insidious way. In the negative social context three main groups of variables are distinguished: family, school, and other contexts. Among the characteristic variables of the individual, they identify evolutionary and not evolutionary aspects, escape responses, personality variables and learning deficits. The association is supposed to be established from the processes of classic and instrumental conditioning, modelling and instruction. The model attributes a key role to the “uncertainty or doubt perceived by the subject” and to the related escape and avoidance responses in the onset, development and maintenance of the disorder. Specifically, they take into account the history of learning of the subject with SP and the level of control/doubt that the subject thinks he has about his behavior and about the behavior of the others (the person he/she was talking to and the audience) in the situation that he has to face/expose himself to. From this point of view, the situation is just one of the independent variables which determine the intensity, frequency and duration of the anxiety response provoked by the fear of negative evaluation. Other variables would be chronicity of the disorder operationalized as the number of situations where the subject has observed himself without being able to control his own response and the intensity of that observation of decontrol; the existence of avoidance behavior in the subject's repertoire of behaviors associated with that kind of social situations; the history of learning, etc. As a consequence, what seems to be relevant is the “degree of doubt perceived by the subject,” regarding the probability of controlling his behavior in a social situation to which he will be exposed and the probability of controlling the possibility of obtaining punishment by the interlocutor (the person he is talking to) or by the audience (e.g., inadequate comments, sarcasms, ironies, unpleasant comparisons, etc.). However, to debate, this model exceeds our immediate purpose. Readers interested in more details about this model are referred to Olivares and Caballo (2003), paper published in a special issue of the journal “Behavioral Psychology” (Psicología Conductual; published in Spain), on the topic of “Social Phobia: Advances in the description, assessment, and treatment”. The interest of this model for present purposes is that permits us to explain a new integrative proposal of delimitation of social phobia subtypes. These authors point out that the subtypes included in the DSM-IV-TR (APA, 2000) can be defined on the basis of a combination of qualitative and quantitative criteria. For that, they take into account the differentiation between performance and interactional social phobia (in the same sense proposed by Turner, Beidel, and Townsley, 1992) and, later, between specific social phobia and generalized social phobia (APA, 1994 and APA, 2000). On the one hand, since this model, performance and relationship are not considered as qualitatively different, but on the other hand, they involve distinct intensities of doubt, significant different interference with activities of daily life and diverse intensities of distress. This is due to that this proposal assumes that the “degree of doubt perceived by the subject” is partially determined by the kind of social situation to which he will be exposed. That is, the “degree of doubt” means the level of uncertainty that the subject has about the control of his behavior and of the other's behaviors (the interlocutor or the audience), in relation to the possibility of obtaining punishment as a consequence of a negative evaluation about his responses in performance and relation situations. In consequence, for this model, the fear is not of negative evaluation by others (parents, peers, members of opposite sex, and people in authority), but fear of negative evaluation is the consequence due to the inadequate response to the situation or to the “doubt/uncertainty” about the adjustment of the reaction. Thus, degree of distress and interference generated by relation-based social situations will be greater than that produced by those other performance-based social situations. In the same way, they also defend that the distress and the interference will be as much greater as the number of social situations involving relationships with other people. This permits us to formulate a hypothesis about the existence of a quantitative continuum between SSP and GSP, but taking also into account the nature of the social situation in it (Olivares & Caballo, 2003). Nevertheless, the specific operationalization of the SP subtypes will be better described further down on the description of the empirical identification process of subgroups. Another relevant aspect in the issue of subtypes is the fact that most of the studies investigating the subtypes of SP have been conducted on samples of adults. Few studies have treated the delimitation of the subtypes in adolescent samples (Hofmann et al., 1999, Lieb et al., 2000 and Marmorstein, 2006; Wittchen, Stein, & Kessler, 1999) and in all cases the formulations used in the adulthood have been adapted and used in the adolescence. Thus, in the study made by Hofmann et al. (1999) a mixed approach (quantitative and qualitative) was used. While Marmorstein (2006) administered the Diagnostic Interview Schedule for Children (DISC; version 2.3) assessing DSM-III-R symptoms), Lieb et al. (2000) and Wittchen et al. (1999), followed the quantitative criterion according to the DSM-IV's guidelines (APA, 1994). Therefore, the general objective of this study is to determine validity of a new proposal to empirically specify a descriptive profile of the magnitude of the interference/severity of the SP in a community sample of Spanish adolescents. The specific aim is to establish a typology that could give useful information for improving the prognosis of the disorder as well as for determining the design and selection of specific treatment. This will be conducted following Olivares and Caballo's proposal (2003), regarding the subdivision and the operationalization of the SP subtypes, and the nomenclature style of the American Psychiatric Association (APA, 2000, p. 2) to specify the magnitude and the severity of the disorders (light, moderate, severe). For that, it is expected that: (i) the data analysis will reveal the existence of two independent types of social anxiety: performance and interactional anxiety; (ii) the analysis, based on the performance and interactional anxiety scores, will permit us to group the adolescents in clusters compatible with the operationalization proposed by Olivares and Caballo (2003); (iii) the comparison between the obtained clusters will present statistically significant differences in their demographical and psychopathological characteristics.