Background: Most research on the familial aggregation of mental disorders has been based on probands selected from clinics. Sparse research examines the clustering of psychiatric illnesses among families ascertained from the community. Such ascertainment bias limits the generalizability and may compromise the validity of study findings. The objective of this study was to illustrate differences between familial aggregation on the basis of the source of proband ascertainment in a family study of anxiety disorders.
Methods: Forty anxiety probands were recruited from clinics; 36 anxiety probands and 60 control subjects were recruited from the community. All probands and their relatives were assessed using standardized psychiatric assessments.
Results: Selection of probands with panic and generalized anxiety disorders from clinical settings compared with the same local community settings was associated with greater risk to relatives. In contrast, the familial aggregation of social anxiety was greater in the community sample, and there were no differences in the risk of specific phobia to relatives between probands from clinics versus community.
Conclusions: Clinic versus community sampling affects familial clustering of anxiety disorders. In general, the effect of clinic sampling is likely attributable to the greater severity of clinic cases, which may also be associated with increased familial morbidity. Differences observed between the anxiety disorders may be artifactual because of the diagnostic criteria of the anxiety disorders or low power. This study illustrates the importance of establishing standards for case-control selection, especially as complex disorder genetics moves increasingly toward population-based sampling.