Improving the validity of FH-RDC diagnosis of major affective disorder in uninterviewed relatives in family studies: a model based approach.

Analysis of family history and family study diagnoses of major affective disorder in 4806 relatives of affectively ill probands from the NIMH Collaborative Study of the Psychobiology of Depression (Clinical) suggests that rates of disorder in uninterviewed relatives are greatly underestimated by family history. The implications of these underestimates for family and genetics analyses using the family history method are discussed, and other estimates are developed that have better statistical properties. Using the Model Based Direct Adjustment method, the rate of major affective disorder in all relatives (interviewed and uninterviewed) is estimated to be 32%, compared to 25% by the consensus (standard) method, which uses the Family History-RDC interview with one or more family members to make diagnoses on all uninterviewed relatives. This difference (over ten standard errors) is due to the much higher rate of illness estimated for the uninterviewed relatives (28% compared to 14%). Analysis of the sources of insensitivity of the FH-RDC is used to explain the difference between observed and imputed diagnosis rates.