Trends in the prevalence of social phobia in the United States: a synthetic cohort analysis of changes over four decades

Abstract

Previous analysis of data from the U.S. National Comorbidity Survey (NCS) [24] suggested that the lifetime prevalence of social phobia in the community has increased significantly in recent cohorts. Furthermore, a latent class analysis of NCS data [21] revealed two primary classes of persons with social phobia: those with exclusive speaking fears and those with one or more other social-evaluative fears. Social phobia in the other social fear group is more persistent, more impairing, and more highly co-morbid with other DSM-III-R disorders. The current report presents data on whether the cohort effect is a general aspect of social phobia or is specific to one of the NCS social phobia subtypes, and whether the cohort effect varies as a function of socio-demographic characteristics. Data were drawn from the NCS. Social phobia was assessed with a revised version of the Composite International Diagnostic Interview. Retrospective age of onset reports were used to estimate Kaplan-Meier survival curves for first onset of social phobia in each cohort represented in the survey. Comparison of these curves allowed us to make synthetic estimates based on retrospective reports of intercohort trends in lifetime prevalence. The lifetime prevalence of social phobia appears to have increased in recent cohorts. However, this increase does not exist among social phobics with exclusive fears of speaking. The increase is most pronounced among white, educated, and married persons, and it is not explained by increased co-morbidity with other mental disorders. The fact that the cohort effect is more pronounced for social phobia with one or more non-speaking fears is important in that this is generally a more severe form of the disorder with an earlier age of onset than social phobia with pure speaking fears. The fact that the cohort effect is most pronounced among people with social and economic advantage (i.e., white, married, well-educated) is intriguing and raises questions about the etiologic process that warrant further study in future research.