In this video, Mohamed Soliman, MD, of the Dauten Family Center for Bipolar Treatment Innovation at Massachusetts General Hospital and Harvard Medical School, distills the central message of his paper: bipolar disorder often hides inside familiar symptoms. Depression, anxiety, irritability, poor concentration, impulsivity, and disturbed sleep appear across many conditions, so the diagnostic challenge is rarely recognizing the symptom, it is reading the pattern around it. Four questions bring that pattern into focus: When did the symptom begin? How long did it last? What triggered it? And did it represent a clear change from the person’s baseline? Poor sleep with anxiety leaves someone exhausted, while in hypomania a person may sleep only a few hours and feel energized; anxious racing thoughts loop around threat, while in bipolar disorder they may turn expansive and tied to new plans. ADHD symptoms usually persist across time, whereas bipolar symptoms cluster into episodes.
On screening, Dr. Soliman’s point is that no single tool is best; each answers a slightly different question, so the right choice depends on setting and population. The Rapid Mood Screener (RMS) is a practical first step in busy primary care, six items in about two minutes with strong performance for bipolar I. The Mood Disorder Questionnaire (MDQ) performs well in psychiatric settings but loses sensitivity in community populations and can miss bipolar II. The Hypomania Checklist-32 helps with subtler presentations but yields more false positives, and the Bipolar Spectrum Diagnostic Scale captures softer or atypical cases. Crucially, screening does not establish a diagnosis; it tells you when a story deserves a closer look.
The clinical implication runs deeper than any one tool. Many patients first present during depression, and in primary care populations with depression, bipolar spectrum disorders may occur in roughly one in six. That argues for asking about lifetime changes in energy, sleep, and impulsivity, family history, and any prior activation after antidepressants, not just the current depressive symptoms. More broadly, psychiatric diagnosis works best when clinicians move from isolated symptoms to patterns across time. Future work will likely pair more sensitive screening with collateral history, repeated measurement, and digital tracking of sleep and activity, but technology is unlikely to replace the core clinical task of understanding how a symptom moves and how it changes a person’s life.
