The purpose of this study was to determine whether, for first-degree relatives of patients presenting to a mood disorders clinic, family history information on psychiatric conditions collected by a psychiatrist and incorporated into the patient's medical records is as informative as that gathered during an interview specifically designed to collect family history data. The study group consisted of 472 first-degree relatives of 78 randomly selected index cases from a large mood disorders genetic database. Family history of psychiatric disorders recorded in regular psychiatric medical records (“clinician history”), and data obtained by a genetic counsellor administering specific family psychiatric history questionnaires to patients and multiple family informants (“family history”) were compared using a kappa statistic. Good agreement between the two methods on the presence or absence of a psychiatric disorder was found among first-degree relatives of index cases, but poor agreement was found with respect to the presence or absence of a specific mood disorder diagnosis(es) in a relative. The results suggest that a clinician-generated family psychiatric history is sensitive to the presence or absence of a psychiatric disorder when compared to a more structured detailed genetic interview. However, for research purposes, a clinician-generated family psychiatric history of a specific mood disorder diagnosis, without supporting collateral information, may not be reliable for use in supporting a mood disorder diagnosis in a patient and/or his relatives.