Bipolar disorder is widely acknowledged to be underdiagnosed, resulting in a huge unnecessary global impact of the disease. Two schools of thought explain this poor recognition: the first considers that bipolar is simply missed; the second proposes that the underdiagnosis is due to unnoticed conversion from unipolar depression.

We examined how the natural history of bipolar disorder should affect our understanding of its underdiagnosis.

We endeavoured to find the points in the patient's history where they first experienced depressive and manic (or hypomanic) symptoms and to follow the progress of their diagnoses through primary and secondary care.

We analysed patient records in a database held by a community mental health trust, finding 128 patients with confirmed bipolar disorder. We then conducted a retrospective audit on the patients’ notes to find the relevant data.

Only 29.5% of the patients had been given a diagnosis of bipolar disorder in primary care, while only 59.3% were diagnosed as such initially by secondary psychiatric services, the majority of the remainders having been diagnosed with major depressive disorder. In general, manic or hypomanic symptoms followed depressive symptoms with a delay of several years (μ=7.3, σ=7.9), but even after manic symptoms were manifested, there was still a considerable delay before a diagnosis of bipolar was made (μ=7.6, σ=8.3).

Our study demonstrates that there is some evidence for both theories of the underdiagnosis of bipolar. Psychiatrists must be aware both of missing manic symptoms and of conversion from unipolar depression.