This paper presents the first international data on the prevalence of DSM-IV bipolar disorder and the broader bipolar spectrum using common diagnostic procedures and methodology. In a combined sample of 61,392 adults from 11 countries, the total lifetime prevalence of BP-I was 0.6%, BP-II was 0.4%, and subthreshold BP was 1.4%, yielding a total prevalence estimate BPS of 2.4% worldwide. Comparable 12-month prevalence rates were 0.4% for BP-I, 0.3% for BP-II, and 0.8% for subthreshold BP, with a total 12-month prevalence BPS of 1.5%. These rates are somewhat lower than those from earlier reviews of European studies 12
and international studies 9
, which yielded aggregate estimates of 1.5% and 0.8% for BP-I and BP-II, respectively, but with a far wider range of estimates than those reported here. In fact, variation in prevalence rates in these studies was attributed primarily to differences in the diagnostic interviews and definitions that were used to characterize bipolar disorder 6
. The use of common diagnostic definitions in the WMH reduces the methodologic diversity that has hindered international prevalence estimates and that has prevented accurate descriptions of the personal and economic impact of this disorder12
These data also provide the first international evidence that supports the validity of the spectrum concept of bipolarity. There was a direct association between increasingly restrictive definitions of bipolar disorder and indicators of clinical severity including symptom severity, role impairment, comorbidity, suicidality and treatment. For example, the proportion of mood episodes rated as clinically severe increased from 42.5% for subthreshold BP to 68.8% for BP-II to 74.5% for BP-I, and the proportion of cases reporting severe role impairment ranged from 46.3% for subthreshold BP to 57.1% for BP-I. The finding that more than half of those with bipolar disorder in adulthood date their onset to adolescence highlights the importance of early detection and intervention, and possibly prevention of subsequent comorbid disorders. Since the average age at onset of bipolar disorder occurs at one of the most critical periods of educational, occupational, and social development, its consequences often lead to lifelong disability.
These findings also confirm those of previous epidemiologic surveys that have highlighted pervasive comorbidity between bipolar disorder and other mental disorders 11, 31
. Despite the differences in prevalence rates of comorbid disorders across countries, nearly three-quarters of those with BP-I or BP-II, and more than half of those with bipolar spectrum disorders have a history of three or more disorders. With respect to specific types of conditions, the association of bipolar disorder with anxiety disorders, particularly panic attacks, was notable; 62.9% of those in the bipolar spectrum have an anxiety disorder, with nearly half reporting panic attacks, and about one-third meeting criteria for a phobic disorder. Confirmation of a strong link between bipolar disorder and anxiety is particularly interesting in light of results from prospective studies of adolescents32
and follow-up studies of children of bipolar parents suggesting that anxiety disorders may constitute an early form of expression of the developmental pathway of bipolar disorders 33, 34
There was striking similarity in patterns of comorbidity of bipolar spectrum disorders with substance use disorders despite large differences in cross-national prevalence rates of substance use and abuse. The strong association of bipolar disorder with substance use disorders has also been widely described in both community and clinical samples35
. Recent findings from a 20-year prospective cohort study revealed a dramatic increase in risk of alcohol dependence associated with symptoms of mania and bipolar disorder in early adulthood 36
. This suggests that bipolar disorder can be considered a risk factor for the development of substance use disorders, which has important implications for prevention efforts. These findings also support the need for careful probing of a history of bipolarity among those with substance use disorders.
Finally, the large proportion of those with severe symptoms and role impairment highlights the serious nature of bipolar disorder. Three-quarters of those with BPS reported severe levels of depressive symptoms and a comparable magnitude of severe role impairment. Most striking, one in every 4–5 person with BPD had made suicide attempts. When taken together with the early age at onset and strong association with other mental disorders, these results provide further documentation of the individual and societal disability associated with this disorder 16, 37, 38
. In light of the disability associated with bipolar disorder, the lack of mental health treatment among those with bipolar disorder, particularly in low income countries, is alarming. Only one-quarter of those with bipolar disorder in low income countries, and only half of those in high income countries, had contacted mental health services. Previous findings of service use in the WMH surveys described the large gap between the burden of mental disorders and mental health care in the studies participating in the WMH survey 21
. However, it is notable that the majority of those with BP disorder received treatment in the mental health sector, even in low income countries.
Interpretations of our findings must take account of the following conceptual and methodological issues. First, the surveys are cross-sectional so the findings are based on retrospective recall of symptoms and their correlates. Second, despite the use of common interview and diagnostic methods, there was still substantial cross-national variation in the rates of BPS. Although it is possible that these differences reflect real variation in prevalence perhaps due to higher false negatives in countries with greater stigma associated with mental illness, further inspection of these differences suggested that there was also variation in the translation, implementation and quality control in some countries that may have led to reduced prevalence rates. Although we found evidence for the validity of the diagnosis of bipolar spectrum based on the CIDI compared with clinical reappraisal interviews (k
, comparable clinical validation studies were not carried out in all of the participating WMH countries. Third, comparisons across countries may be limited because of variability in the availability of mental health treatment.
There are also several features of the WMH Survey Initiative that represent advances over prior cross-national studies in psychiatric epidemiology. First, the WMH initiative includes far larger representation of several regions of the world including low income countries. Second, the high degree of coordination across studies enhanced the validity of the cross-national comparisons. Third, the inclusion of standardized methods for assessing severity and role impairment facilitates estimation that can provide a context for the public health significance of the prevalence estimates.
These findings demonstrate the important growth of international collaborations that permit investigation of cultural and regional differences in prevalence and risk factors for mental disorders. Recent efforts such as the proposal of a common global nomenclature to define the course and outcome in bipolar disorders as proposed by a task force under the auspices of the International Society for Bipolar Disorders39
should facilitate outcome studies across geographies. Contemporary issues concerning bipolar disorder that warrant further study include: further evaluation of the thresholds and boundaries of bipolar disorder; better integration of adult and child epidemiology of bipolar disorder and its evolution in light of its onset in adolescence, and further investigation of explanations for the patterns of comorbidity between bipolar disorder with other disorders. In summary, this paper reports the first data on the prevalence and correlates of the full spectrum of bipolar disorder in a series of nationally representative surveys using common methodology. As such, they document the magnitude and major impact of bipolar disorder worldwide and underscore the urgent need for increased recognition and treatment facilitation.