[Skip to Content]
[Skip to Content Landing]
February 1984

Information From RelativesDiagnosis of Affective Disorders

Author Affiliations

From the Section on Psychogenetics, Biological Psychiatry Branch, National Institute of Mental Health, Bethesda, Md.

Arch Gen Psychiatry. 1984;41(2):173-180. doi:10.1001/archpsyc.1984.01790130069010

• We compared diagnoses made blindly by experienced clinicians from interview records with diagnoses obtained by computer pooling and scoring of relatives' information on the same persons. For major affective disorder diagnosed by interview, the relatives' information agreed on presence of affective illness for 96% of 159 probands and 48% of 195 relatives of affectively ill and control probands. In 1,093 relatives of affectively ill and control probands, the k values for diagnostic agreement were as follows: any major affective disorder,.51; bipolar I disorder,.61; and unipolar disorder,.42. Schizoaffective and bipolar II diagnoses did not show significant agreement. Only 15% of interview-diagnosed relatives were identified as having a major affective disorder by one informant alone, going up to 64% agreement with four or more informants. Final diagnostic estimate from all available information, including medical records, generally followed the interview diagnosis rather than the relatives' information.