The “Bipolarity Index”

(updated 11/2014)


This system has recently been described in an article by Harvard’s Director of the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD), Dr. Gary Sachs. It represents a “clinical” diagnostic approach — what matters to doctors and patients when they are trying to understand a particular patient’s mood problems. Unlike a DSM-based diagnostic system which emphasizes separate diagnostic categories, the Bipolarity Index places patients on a spectrum. As you’ll see, it also emphasizes diagnostic features other than hypomania.

Five dimensions (including the DSM as one)

Dr. Sachs’ system was developed with several colleagues at Harvard, including Dr. Nassir Ghaemi, whose name you will find mentioned frequently on this website, as he is a well-known researcher of this “spectrum” way of looking at bipolar diagnosis.

The system considers 5 “dimensions” of bipolarity. Note that the presence of hypomania or mania is only one of the five dimensions. All the others receive equal weight, for now (the system has not been subjected to the usual tests a “diagnostic instrument” would receive; thus we do not yet know how much weight each of these 5 dimensions should carry):

  1. Hypomania or mania
  2. Age of onset of first mood symptoms
  3. Illness course and other features generally only visible over time
  4. Response to medications (antidepressants and mood stabilizers)
  5. Family history of mood and substance use problems

Update 2015: the following table details the authors’ intent to quantify each of these 5 dimensions. To my knowledge there have been no publications showing whether these weightings produce an effective characterization of “bipolarity”. What’s important is to recognize that when they built it, the authors gave 80% of the weight to the non-manic markers, only 20% of the weight to the DSM criteria.

Table: The 100-point system

Dimension 20 points 15 points 10 points 5 points 2 points
Episode Characteristics Manic symptoms with “prominent euphoria, grandiosity or expansiveness”. Manic symptoms with dysphoria, irritability Hypomanic symptoms; or mania following an antidepressant Hypomanic symptoms following an antidepressant; or hypomania below DSM threshold; or major soft signs: atypical or postpartum depression Psychosis, without other signs of mania
Age of Onset 15-19 15 or 20-30 30-45 > 45
Illness Course (and Other Features) Manic episodes separated by periods of full recovery Incomplete recovery between manic episodes;or hypomania with full recovery between episodes Mania, incomplete recovery, but also substance use; or psychosis only during mood episodes; or legal problems associated with mania Repeated episodes of unipolar depression, no hypomania (3 or more); or hypomania with incomplete recovery between episodes; or any of several other features: borderline; anxiety disorder; ADHD as a child; gambling or other risk behaviors without mania per se; or PMS Hyperthymic temperament; >3 marriages, or two jobs in two years; or two advanced degrees (see Akiskal reference on these latter features)
Response to Medications Full recovery within 4 weeks of treatment with mood stabilizers Full recovery within 12 weeks of treatment; or relapse within 12 weeks of stopping mood stabilizers; or switch to mania within 12 weeks of starting antidepressant Worsening dysphoria or mixed state symptoms during antidepressant; or partial response to mood stabilizers; or antidepressant induced rapid cycling or worsening thereof Lack of response to 3 or more antidepressants; or mania/hypomania when antidepressant stopped Immediate response, almost complete, to antidepressant within 1 week or less
Family History 1st degree relative (brother/sister, parent, or child) with clear bipolar disorder 2nd degree relative with bipolar diagnosis; or 1st degree relative with recurring unipolar depression and features suggestive of bipolar disorder 1st degree relative with recurring unipolar depression or schizoaffective disorder; or any relative with clear bipolar diagnosis; or any other relative with unipolar depression and symptoms suggestive of bipolar 1st degree relative has clear problem with drugs or alcohol 1st degree relative has repeated episodes of depression; or has an anxiety disorder, an eating disorder, or ADHD


Dr. Sachs’ article indicates that in their experience so far, most Bipolar I patients score above 60. Note that’s Bipolar I, not Bipolar II or other more subtle variations, which would presumably have lower scores. Notice that a patient can “get points” for many different features of bipolarity, not just hypomania or mania. In fact, she could get 60 points even without any history of hypomania or mania at all, you see? For example, she might have had: her first depression at age 18 (20 points); post-partum depression (5 points); more than 3 episodes of depression (5 points); agitation while taking an antidepressant (10 points); and have a sister with clear bipolar disorder (20 points).

But beyond this — Bipolar I patients score above 60 — we do not yet know what a particular “score” might mean. Therefore this table is not offered to help you score patients (or yourself). It is presented to demonstrate that some of the most respected mood experts in the world are now using a system which approaches bipolar disorders as existing on a “spectrum”, rather than a yes/no, you-have-it-or-you-don’t matter. Dr. Sachs himself explains this in an interview conducted shortly after this article, excerpted here.

If after all this you still want a “fine-tooth comb” to look for hypomanic/manic symptoms — only 1/5th of the points here, but still the major diagnostic feature in the official DSM system — here is a list of such symptoms.