Diagnostic certainty of a voluntary bipolar disorder case registry

Background: Strategies for identifying and recruiting persons with bipolar disorder are of importance as interest in studying this relatively uncommon, but highly disabling illness increases. The development and implementation of a bipolar disorder case registry and the assessment of diagnostic certainly of the resulting sample are described. Methods: Eight hundred and four individuals who self-reported a history of bipolar disorder were recruited. Telephone interviewers gathered demographic information and clinical, medical and treatment history information. One hundred randomly-selected registrants completed an in-person structured diagnostic interview. Self-report of diagnosis was compared to the results of the diagnostic interview. Results: Ninety three percent of registrants interviewed met criteria for a lifetime bipolar spectrum diagnosis; of these, 76.3% were diagnosed with bipolar I disorder. Agreement between self-reported and SCID diagnoses was 93%, indicating that self-report of a bipolar diagnosis is highly reliable. Two-thirds had experienced at least one other Lifetime Axis I diagnosis, with substance abuse/dependence (55.9%) and panic disorder (19.4%) the most common comorbidities. Limitations: Since nearly all of the sample have previously been diagnosed as having bipolar disorder by a professional, the sample's representativeness of the population as a whole may be somewhat limited. Conclusions: Persons with bipolar disorder can accurately identify themselves as having the disorder via a telephone interview, indicating that a case registry method is a useful strategy for recruiting very large samples of persons with this disorder. Such large samples will allow for further study of treatment variations among patient subgroups, of pathways to treatment, and of the effectiveness of new treatments. (C) 1999 Elsevier Science B.V. All rights reserved.