There is considerable debate over whether bipolar and related disorders that share common signs and symptoms, but are currently defined as distinct clinical entities in DSM-IV and ICD-10, may be better characterized as falling within a more broadly defined "bipolar spectrum". With a spectrum view in mind, the possibility of broadening the diagnosis of bipolar disorder has been proposed. This paper discusses some of the rationale for an expanded diagnostic scheme from both clinical and research perspectives in light of potential drawbacks. The ultimate goal of broadening the diagnosis of bipolar disorder is to help identify a common etiopathogenesis for these conditions to better guide treatment. To help achieve this goal, bipolar researchers have increasingly expanded their patient populations to identify objective biological or endophenotypic markers that transcend phenomenological observation. Although this approach has and will likely continue to produce beneficial results, the upcoming DSM-IV and ICD-10 revisions will place increasing scrutiny on psychiatry's diagnostic classification systems and pressure to re-evaluate our conceptions of bipolar disorder. However, until research findings can provide consistent and converging evidence as to the validity of a broader diagnostic conception, clinical expansion to a dimensional bipolar spectrum should be considered with caution.