The issue of comorbid conditions has been more readily recognized in contemporary studies. For example, McElroy et al. (2001) evaluated 288 outpatients with bipolar I or bipolar II, using structured diagnostic interviews to determine the diagnosis of bipolar, comorbid Axis I diagnoses and demographics. They found that 187 (65%) with bipolar disorder also met DSM-IV criteria for at least one comorbid lifetime Axis I disorder. More had anxiety (42%) and substance misuse (42%) than eating disorder (5%). There were no differences in comorbidity between bipolar I and bipolar II. Both lifetime and current Axis I comorbidity were associated with an earlier age of onset. Current Axis I comorbidity was associated with history of both cycle acceleration and more severe episodes over time.

One overarching problem in the epidemiology of bipolar disorder is the debate surrounding broad or narrow criteria. The above discussion illustrates the case for more specific criteria. However, there has been a move by some toward considerable expansion of the concept. Akiskal et al. (2000) have challenged the narrow definition of what constitutes bipolar disorder. Kraepelin's own broad-brush, inclusive definitions included hypomania; temperamental dispositions of a cyclothymic, irritable, manic type; and a family history of manic depression.

The current classifications are narrower and adopt the unipolar/bipolar approach. Conservative criteria have estimated bipolar disorder to account for 1% of the population and only 10-15% of all mood disorders (Regier et al., 1988; Weissman et al., 1996). These figures have been challenged by Akiskal et al. (2000) on the basis of the broader concept of bipolarity that has evolved over the last 20 years to include manic hypomanic, rapid-cycling, and mixed mania.

The unipolar-bipolar distinction (proposed, among others, by Angst, 1966/1973, and Winokur et al., 1969) has become the accepted nomenclature of most clinicians and researchers. As Akiskal et al. (2000) put it, the dichotomy may have a degree of heuristic value but fails to help in placing and understanding those affective disorders which lie between.

The rise in genetic studies of bipolar pedigrees has uncovered many affected individuals with principally depressive features (Akiskal et al., 1985; Gershon et al., 1982; Tsuang et al., 1985). Akiskal and Mallya (1987) estimated that 4-5% of the general population belongs to a broad bipolar spectrum with chiefly depressive phenomenology coupled with less-than-manic excitements.

The clinical reality of these "less-than-manic" patients has led to various reclassifications. For example, Dunner et al. (1976) described less-than-manic patients as bipolar II on the basis of hospitalization for depression and excited periods that did not require hospitalization. Fieve and Dunner (1975) had reserved bipolar I for those who were admitted for mania. It should be remembered that "less-than-manic" excitements are very controversial concepts, and the debate remains active. The "soft bipolar spectrum" (Akiskal & Mallya, 1987), a more inclusive term for bipolar conditions beyond classical mania, revises previous definitions of bipolar II by incorporating depressions with hypomanic episodes, and cyclothymic and hyperthymic traits, as well as familial bipolarity. The spectrum also includes hypomanic episodes which occur during pharmacotherapy or other somatic treatments.

Other terms used for "less-than-manic" bipolar conditions with depressive presentations include "Dm" (Angst etal., 1980), "unipolar-L"(Kupferetal., 1975), and "pseudo-unipolar" depression (Mendels, 1976).

In a series of more formal bipolar spectrum proposals (Akiskal, 1983, 1996; Akiskal & Akiskal, 1988), bipolarity is categorized into the following types:

type I: mania with or without depression type II: depression with hypomania and/or cyclothymia type III: hypomania associated with antidepressants, as well as depression with hyper-thymic temperament and/or bipolar family history.

Table 10.2 illustrates the results of a consecutive series of cases examined by Akiskal and Mallya (1987) in response to an American Psychiatric Association request for an assessment of whether the (then) new DSM-III provided sufficient coverage for all affective diagnoses. It demonstrates that the bipolar spectrum conditions were as prevalent as their unipolar counterparts.

Angst (1998) demonstrated a high prevalence of brief hypomanic episodes below the 4-day requirement of DSM-IV. His work argues in favour of a broadening of the bipolar

Table 10.2 Primarily affective diagnoses in a community setting (adapted from Akiskal et al., 2000)



5 44



Dysthymia concept to include both the severe—psychotic mania—end of the spectrum and the subthreshold end—brief hypomania.

Subthreshold does not equate with subclinical or clinically insignificant. Indeed, subthreshold episodes have been shown to have significant psychosocial consequences. However, the question of whether this constitutes inclusion criteria in a categorical classification remains unanswered. The real question might be whether categories are appropriate for these criteria or whether they are better placed in the context of continua.

Genetic research in the form of twin studies has provided evidence for, rather than against, the broader concept of bipolarity. Monozygotic twins that were discordant for strictly defined mood disorders were broadly concordant for mood-labile temperaments at the milder, untreated end of the spectrum and mood-incongruent psychoses beyond the boundaries of the classical affective psychoses at the severe end of the spectrum (Bertelsen et al., 1977).

Most epidemiological surveys have excluded less-than-manic forms of bipolar disorder. Rates of bipolar disorder, based mainly on ascertaining a history of mania, have been found to be 1%.

The two US national studies which affected the received wisdom with respect to rates of bipolar are the ECA (Regier et al., 1988) and the NCS (Kessler et al., 1994). Respectively, the rates of bipolar disorder recorded from these surveys are 1.2% and 1.6%. Weissman et al. (1996) conducted a cross-national study of rates and reported a range of 0.3-1.5%.

Newer data demonstrate higher rates owing to the broadening of the concept of bipolar disorder to include mania of less than 4 days, hypomania, brief hypomania, and cyclothymia.

Epidemiological literature from community studies in the USA and several European countries has included soft bipolar within the spectrum of bipolar (Table 10.3). This newer

Table 10.3 Lifetime prevalence rates of bipolar disorder


Regieret al. (1988)/USA Kessler et al. (1994)/USA Lewinsohn et al. (1995)/USA

Weissman et al. (1996)/cross-national 0.3-1.5

Szadoczky et al. (1998)/Hungary Angst (1998)/Switzerland

literature broadens the bipolar spectrum from 1% to at least 5%. This is hardly a surprising change. Broader criteria are bound to increase the rates recorded. By contrast, the data from Narrrow et al. (2002) showed that using clinical significance criteria, rather than simply broadening the concept of bipolar, reduced prevalence rates.

Understanding And Treating Bipolar Disorders

Understanding And Treating Bipolar Disorders

Are You Extremely Happy One Moment and Extremely Sad The Next? Are You On Top Of The World Today And Suddenly Down In The Doldrums Tomorrow? Is Bipolar Disorder Really Making Your Life Miserable? Do You Want To Live Normally Once Again? Finally! Discover Some Highly Effective Tips To Get Rid Of Bipolar Disorder And Stay Happy And Excited Always! Dont Let Bipolar Disorder Ruin Your Life Anymore!

Get My Free Ebook

Post a comment