Clinical methods in psychiatric genetics. II. The high risk approach

The study of individuals at "high" risk for developing psychiatric disorders is useful in confirming that a biological trait marker identified in patient populations is also present in genetically susceptible individuals who have never been ill, and predicts the future onset of illness. We outline a systematic method for deciding which variables to choose and how many individuals are required in order for a study to have sufficient power. We demonstrate how these decisions depend on the assumptions that can be made with regard to the mode of inheritance of the biological trait, the relationship of the biological trait to illness, and the magnitude of the mean difference observed between patients and controls. We also quantify the increased power of studying offspring of two affected parents rather than offspring of one affected parent.