Introduction

Until the last decade, epidemiological studies have suggested a relatively consistent and low lifetime risk of bipolar disorder in the order of 0.2–0.8% (for manic illness) and some 1% for bipolar spectrum disorders (Goodwin and Jamison, 1990). In the last decade, distinctly higher lifetime community prevalence rates have been reported, as shortly detailed. Any increase is likely to reflect a number of factors, including broadening of disorder boundaries imposed by the ‘bipolar spectrum’ concept, shortening of duration criteria for ‘highs’, as well as greater community and clinician awareness and improved detection of the bipolar disorders.

Many clinicians suggest that, in addition to such factors, there appears to have been an increase in patients presenting with bipolar disorders – and Bipolar II (BP II) in particular. If not an artefact of improved awareness and modified detection and diagnostic processes, any true increase should be identifiable in community studies undertaken over time. The limitation to this argument is that, while there have been a number of relevant community studies over the last quarter-century, few have captured BP II, while the case-finding diagnostic tools have also varied considerably.

Epidemiological studies

Several US community studies illustrate a focus on Bipolar I Disorder (BP I). In the Epidemiologic Catchment Area (ECA) study (Robins et al., 1984), where DSM–III diagnoses were generated by the Diagnostic Interview Schedule (DIS), the lifetime rate of a ‘manic episode’ was 0.92%, and represented 11.8% of the overall affective disorders group (i.e. manic episode, major depressive episode, dysthymia).