The bipolarity index: how bipolar are you?
This exchange indicates just how far the bipolar diagnostic paradigm has shifted—not toward replacing DSM, but toward thinking in spectrum as well as categorical terms. Dr Sachs is referring to a new instrument that is currently in use, among other tools, at the bipolar disorders clinic at MGH: the Bipolarity Index. This diagnostic tool, which is still being tested, awards 100 total points for 5 dimensions of bipolarity:
- Episode characteristics (hypomania or mania).
- Age at onset.
- Illness course.
- Response to medication.
- Family history.
At present, each dimension is worth 20 points. As the tool is studied, these will likely acquire different weightings. For now, the index simply illustrates an interest in systematically gathering data regarding signs and symptoms other than hypomania or mania. It shows that Dr Sachs and colleagues are thinking in spectrum terms (as well as categorically). The index and the current scoring system are available for download, along with other tools used in the MGH clinic (via www.manicdepression.org), although Dr Sachs would likely emphasize how preliminary the work with this index remains at present.
Paper-and-pencil bipolar testing: MDQ or Bipolar Spectrum Diagnostic Scale (BSDS)?
A more widely studied aid for bipolar diagnosis is the MDQ. As most readers are likely to know, the MDQ presents a series of yes-or-no questions for the patient to answer, which involve the DSM criteria for hypomania/mania. The MDQ has become the de facto standard for bipolar screening, even though in some settings the sensitivity and specificity are not very reassuring, leading some authors to conclude that the test is not sufficiently accurate for use as a screening tool.9 However, it appears that a powerful determinant of its accuracy is the “hunch” of the user before the test—the “pretest probability.”10 Surprisingly, this has at least as much impact as the sensitivity and specificity on the predictive value of the test result. Thus, results from this test are only of value if the pretest probability can be estimated; otherwise, the risk of false positives, in particular, is dangerously high. Nevertheless, the test can still be used to educate people about bipolar disorder, and it can be a handy means of showing the patient his or her bipolarity from another angle. Unfortunately, the MDQ is being widely promulgated,11 often without these warnings about its accuracy.
Moreover, note that the MDQ—like the DSM—is a categorical instrument: it has been studied and designed for yielding a yes-or-no answer as to whether bipolar disorder is present. At high pretest probabilities (ie, when a skilled practitioner has a high index of suspicion), a positive MDQ result can raise post-test probabilities to about 90%.10
However, the MDQ, used in this way, reinforces the idea that bipolar disorder is either present or absent (a DSM-based categorical perspective). A more nuanced view is presented in another, less widely known questionnaire: the BSDS, originally developed by Ron Pies, MD.12 Probability of bipolar disorder is reported in 4 ranges, from “highly likely” to “very unlikely.” The test thus sends a message to patients that is consistent with the spectrum concept.
Unfortunately, neither test has been validated using BSDS-style probabilistic scoring. Determining a test’s sensitivity and specificity requires using a yesor- no scoring system, for comparison with a gold standard. When evaluated in this way, using a cutoff scoring approach, the BSDS performs rather similarly to the MDQ.10 Unfortunately, evaluating it in this way obviates one of its primary advantages, namely the very spectrum perspective its name emphasizes.
For now, in my opinion, these tests are best used in 1 of 2 ways. First, they can be used as educational tools, to make a means of testing oneself easily accessible to the general public, and thereby raising awareness of and questions about bipolar disorder. The BSDS performs this role substantially better than the MDQ. Second, primary care providers (PCPs) can use them to save time.
Realistically, PCPs do not have adequate time for diagnostic interviews sufficiently thorough to “rule out hypomania,” as the DSM requires in order to establish the diagnosis of major depression. Recently, however, the FDA suggested that every patient who is to receive antidepressant medication be screened for bipolar disorder,13 a recommendation that now appears on every antidepressant package insert (not in the black box section at the top but in the warnings section about halfway through).
The FDA’s recommendation is very nearly a requirement for action from a liability standpoint, so PCP screening for bipolar disorder has effectively been mandated. Since they are not being awarded any additional time or funding, and since most of them—in the Western United States, at least—find it very difficult to refer patients to psychiatrists because so few are taking new patients, the MDQ may be a very necessary tool in their practice. We psychiatrists might regard a cursory, yes-or-no analysis of bipolarity primitive, but unless we can help by taking the PCPs’ patients, we may need to help them use the MDQ as effectively as possible.10
For skilled mental health practitioners, however, the MDQ and the BSDS have value primarily as a shorthand approach to accelerate interviewing. In that role, the HCL-32 may be superior.
The HCL-32: a laundry list of hypomanic symptoms
Did you ever wish you had a 1-page list of hypomanic symptoms and behaviors you could hand your patient, to review and discuss—one that would capture many of the variations you’ve seen in your practice? The HCL-32 was developed as an instrument for self-assessment, like the MDQ and BSDS, but it is much more thorough, with 32 different aspects of hypomania queried.14 As with any such test, it must be used by a skilled practitioner to derive a (at least intuitive, if not formally quantitative) post-test probability of bipolar disorder. However, as raw material for discussion, when degree of bipolarity remains in doubt, or as a list of symptoms that should arouse concern in scanning for future episodes of illness (eg, when an antidepressant is being introduced and when monitoring for hypomania is warranted), the HCL-32 is very handy. A 1-page symptom list version can be downloaded from the following site: http://www.psycheducation.org/depression/HCL-32.htm.
Mixed states as waves
Though it has not yet been subjected to validation testing, Mackinnon and Pies15 have put forth a striking new model of bipolar cycling. They suggest that mixed states can be understood, at least for some patients, as a product of cyclic changes in mood, energy, and speed of thought/creativity that are asynchronous, as shown in the upper portion of the Figure (see Psychiatric Times, August 2006, p. 87). In the lower portion, one can see that if a patient were to experience this kind of continuous fluctuation, his or her mood states would vary almost continually without clear demarcation of “episodes.” The patient’s predominant experience would be a “mixed” mood state, yet full of variations—enough to leave him very uncertain of how he might feel and function on any given day, as we commonly hear from our patients. When I have shown this graph to patients, it seems to quickly shift their understanding of bipolarity from the classic mania/depression/ well-interval model to a continuousvariation model that often better matches their experience. This model deserves attention and study but is already a useful teaching instrument for conveying the spectrum of bipolar experience that can occur within a single patient.
Conclusion
Might using these tools (or the spectrum perspective in general) lead to overdiagnosis of bipolar disorder? Certainly the intent of this and my previous article has been to expand the diagnostic range in which bipolar disorder might be considered. As noted last month, the principle value of the spectrum concept is to add bipolarity to the differential of the apparently “unipolar” disorder even when a thorough search for hypomania has yielded little. This does indeed raise the risk of overdiagnosis.
One way to address the concern about swinging this pendulum too far is to set the bar high for treatment with any medication, emphasizing the use of exercise, psychotherapy, and other such nonpharmaceutical modalities before turning to therapies with greater risks. This will guard against overprescription, if not overdiagnosis. (As discussed last month, at least some of the worry about overdiagnosis may represent concern about the risks of mood stabilizers relative to antidepressants.)
“Overdiagnosis” presumes we have some way of knowing what “correct” diagnosis really means. Until we have biologic underpinnings or much better long-term outcome data for the conditions we treat, correct diagnosis will remain uncertain, and thus overdiagnosis will be impossible to assess. It may be that we are currently overdiagnosing bipolar disorder; it could also be that for decades it has been so dramatically underrecognized that a greater shift in diagnostic practices is indicated. We shall see. In the meantime, however, another paradigm shift is underway. The process of medical care is shifting from physician control of information and decision making toward a new model in which patients and their families actively seek information outside the doctor-patient relationship and play a more active role in diagnosis and treatment.16 For such clients, we may have to adjust our role toward guiding their education, including which diagnostic models and tools they might best use. I hope these articles help you prepare for that shift.
Dr Phelps has been practicing psychiatry for more than 15 years in Corvallis, Oregon and specializes in treating bipolar disorder. He recently published Why Am I Still Depressed? Recognizing and Managing the Ups and Downs of Bipolar II and Soft Bipolar Disorder (McGraw-Hill), a book version of his Web site (www.PsychEducation.org).
Dr Phelps reports that he has received grants and honoraria from GlaxoSmithKline, Astra- Zeneca, and Abbott Laboratories.
References
1. Saenger E. The bipolarity index as a tool for assessment and creating rapport: an expert interview with Gary Sachs, MD. Medscape Psychiatry & Mental Health. 2005;10(1). Available at: http://www.medscape. com/viewarticle/503893. Accessed March 28, 2006.
2. Mitchell AJ. High medication discontinuation rates in psychiatry: how often is it understandable? J Clin Psychopharmacol. 2006;26:109-112.
3. Judd LL, Akiskal HS, Schettler PJ, et al. The longterm natural history of the weekly symptomatic status of bipolar I disorder. Arch Gen Psychiatry. 2002;59: 530-537.
4. Judd LL, Akiskal HS, Schettler PJ, et al. A prospective investigation of the natural history of the longterm weekly symptomatic status of bipolar II disorder. Arch Gen Psychiatry. 2003;60:261-269.
5. Madden M. Reports: Internet evolution. Internet penetration and impact. Pew Internet and American Life Project. April 2006. Available at: http://www.pewinternet.org/PPF/r/182/report_display.asp. Accessed May 4, 2006.
6. Bipolar disorder (main page). National Institute of Mental Health Web site. Available at: http://www. nimh.nih.gov/Publicat/bipolar.cfm#bp1. Accessed June 29, 2006.
7. Ghaemi SN, Ko JY, Goodwin FK. “Cade’s disease” and beyond: misdiagnosis, antidepressant use, and a proposed definition for bipolar spectrum disorder. Can J Psychiatry. 2002;47:125-134.
8. Phelps JR. Why am I still depressed? Recognizing and managing the ups and downs of bipolar II and soft bipolar disorders. New York: McGraw-Hill; 2006.
9. Zimmerman M, Posternak MA, Chelminski I, Solomon DA. Using questionnaires to screen for psychiatric disorders: a comment on a study of screening for bipolar disorder in the community. J Clin Psychiatry. 2004;65:605-610.
10. Phelps JR, Ghaemi SN. Improving the diagnosis of bipolar disorder: predictive value of screening tests. J Affect Disord. 2006;92:141-148.
11. AstraZeneca Pharmaceuticals. Is it really depression? Available at: http://www.IsItReallyDepression. com/mini_c/isitreallydepression/. Accessed April 30, 2006.
12. Ghaemi SN, Miller CJ, Berv DA, et al. Sensitivity and specificity of a new bipolar spectrum diagnostic scale. J Affect Disord. 2005;84:273-277.
13. Class suicidality labeling language for antidepressants. Food and Drug Administration antidepressant warnings template. 2005. Available at: http://www.fda.gov/cder/drug/antidepressants/PI_template.pdf. Accessed May 12, 2006.
14. Angst J, Adolfsson R, Benazzi F, et al. The HCL- 32: towards a self-assessment tool for hypomanic symptoms in outpatients. J Affect Disord. 2005; 88:217-233.
15. Mackinnon DF, Pies R. Affective instability as rapid cycling: theoretical and clinical implications for borderline personality and bipolar spectrum disorders. Bipolar Disord. 2006;8:1-14.
16. Rainie L. Blogs and health care. Presented to the National Association of Children’s Hospitals and Related Institutions (NACHRI). Pew Internet and American Life Project. March 2006. Available at: http://www.pewinternet.org/PPF/r/62/presentation_ display.asp. Accessed May 4, 2006.
