The recent follow-up studies of Geller et al

Transcription

1 The Poor Prognosis of Childhood-Onset Bipolar Disorder GABRIELE S. LEVERICH, MSW, LCSW-C, BCD, ROBERT M. POST, MD, PAUL E. KECK JR, MD, LORI L. ALTSHULER, MD, MARK A. FRYE, MD, RALPH W. KUPKA, MD, PHD, WILLEM A. NOLEN, MD, PHD, TRISHA SUPPES, MD, PHD, SUSAN L. MCELROY, MD, HEINZ GRUNZE, MD, KIRK DENICOFF, MD, MARIA K.M. MORAVEC, BS, AND DAVID LUCKENBAUGH, MA Objective We examined age of onset of bipolar disorder as a potential course-of-illness modifier with the hypothesis that early onset will engender more severe illness. Study design A total of 480 carefully diagnosed adult outpatients with bipolar disorder (mean age, years) were retrospectively rated for age of illness onset, time to first pharmacotherapy, and course of illness. Clinicians prospectively rated daily mood fluctuations over 1 year. Results Of the 480 patients, 14% experienced onset in childhood (12 years or younger); 36% in adolescence (13 to 18 years); 32% in early adulthood (19 to 29 years); and 19% in late adulthood (after 30 years). Childhood-onset bipolar illness was associated with long delays to first treatment, averaging more than 16 years. The patients with childhood or adolescent onset reported more episodes, more comorbidities, and rapid cycling retrospectively; prospectively, they demonstrated more severe mania, depression, and fewer days well. Conclusions This study demonstrates that childhood onset of bipolar disorder is common and is associated with long delays to first treatment. Physicians and clinicians should be alert to a possible bipolar diagnosis in children in hopes of shortening the time to initiating treatment and perhaps ameliorating the otherwise adverse course of illness. See editorial, p 459 (J Pediatr 2007;150:485-90) The recent follow-up studies of Geller et al 1,2 suggest a relatively difficult course of illness in children with bipolar disorder treated naturalistically in the community. Although 65% of the patients studied recovered for 2 weeks by the end of 2 years, 55% relapsed within this 2-year period, and 64% relapsed during the 4-year follow-up. Birmaher et al 3 reported extremely long delays to achieving acute remission, along with relatively high subsequent relapse rates, in children with bipolar I, bipolar II, and especially bipolar NOS (not otherwise specified) disorders. Controlled clinical trial findings in long-term prophylaxis of childhood bipolar illness have reported high dropout rates due to inefficacy of either lithium or valproate monotherapy. 4 There remains much controversy about the quality of symptoms and severity/ duration thresholds sufficient for diagnosing childhood-onset bipolar illness. 5-8 Given this ongoing controversy, many children with bipolar illness are not being identified until after many years of psychopathology. In some instances, these children are being treated with stimulants or antidepressants for their comorbid attention deficit hyperactivity disorder (ADHD) without the recommended coverage of a mood stabilizer or atypical antipsychotic. 6 Much of the diagnostic controversy derives from the lack of characteristic discrete episodes so often seen in adult-onset bipolar disorder. 9 In contrast, many of the youngest children with otherwise profound manic and depressive mood and behavioral swings associated with considerable social or educational dysfunction show brief bursts and rapid changes in mood, sometimes several times within a 24-hour period, a phenomenon called ultradian cycling. 1,2,10,11 A second reason underlying the controversy is that the extremes ADHD Attention deficit hyperactivity disorder NIMH-LCM National Institutes of Mental Health Life Chart Method SCID Structured Clinical Interview for DSM-IV Axis I Disorders From the Department of Health and Human Services, National Institutes of Health, National Institute of Mental Health, Biological Psychiatry Branch, Bethesda, MD; Psychopharmacology Research Program, Department of Psychiatry, University of Cincinnati College of Medicine, Cincinnati, OH; Mental Health Care Line and General Clinical Research Center, Cincinnati VA Medical Center, Cincinnati, OH; Department of Psychiatry and Biobehavioral Science, David Geffen School of Medicine, University of California Los Angeles and West Los Angeles VA Medical Center, Los Angeles, CA; Altrecht Institute for Mental Health Care, Utrecht, The Netherlands; Department of Psychiatry, University Hospital, Groningen, The Netherlands; University of Texas- Southwestern Medical Center, Dallas; and Psychiatrische Klinik der LMU, Munich, Germany. Supported by the National Institutes of Mental Health and the Stanley Medical Research Institute. Submitted for publication Mar 3, 2006; last revision received Aug 3, 2006; accepted Oct 27, Reprint requests: Gabriele S. Leverich, MSW, LCSW-C, BCD, NIMH, NIH, DHHS, Biological Psychiatry Branch, Bldg 10, Room 3S239, 10 Center Drive MSC-1272, Bethesda, MD levericg@ mail.nih.gov /$ - see front matter Copyright 2007 Mosby Inc. All rights reserved /j.jpeds

2 of activation, impulsivity, irritability, and anger that these children exhibit can occur in other childhood disorders, such as ADHD, oppositional defiant disorder, or major depressive disorder. 12 In this article, we examine the proportion of adults who had childhood and adolescent onset of their illness and assess the illness outcomes related to these different ages of onset. One retrospective study by Perlis et al 13 based on the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) cohort of 983 patients found that 27.7% had on onset of bipolar illness in childhood (age 12) while another 37.6% had their onset in adolescence (from ages 13 to 18). They found that both of these groups had a much more difficult course of illness compared with those with adult onset, including faster cycling, more days depressed, greater number lifetime of manic and depressive episodes, an increased risk of substance abuse and other comorbidities, and a greater lifetime risk of suicide attempts. We also revisit this issue in another adult outpatient cohort with several new methodological approaches. We obtained daily prospective clinician ratings for 1 year during naturalistic treatment of adult bipolar illness by experts to supplement the retrospective self-report findings. We also examined data on the lag between the onset of first symptoms likely meeting DSM-IV diagnostic criteria and the onset of first drug treatment for either mania or depression. METHODS The methods of study of this outpatient cohort have been presented in detail elsewhere Outpatients were recruited between 1995 and 2002 largely from local advertisements and outpatient settings located near 4 research clinic sites in the United States (Los Angeles, Dallas, Cincinnati, and Bethesda) and 3 sites in Europe (Utrecht, Freiburg, and Munich). All patients provided informed written consent for participation in naturalistic follow-up. The population was a broad and representative sample, because patients were not excluded for the presence of comorbidities; the only exclusions were for active ongoing substance abuse that required separate treatment in another facility or other serious medical illnesses that might preclude participation in subsequent treatment protocols if this became necessary. Patients were diagnosed using the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID) 19 and completed questionnaires about the extent of their previous course of illness ,20 The self-report questions included (1) age of onset of first depressive symptoms associated with dysfunction, (2) age of onset of first hypomanic or manic symptoms, and (3) age of first treatment for either mania or depression. Age of onset was also obtained from the SCID 19 interviews by trained clinicians based on an assessment of when symptoms similar to those experienced in adulthood first occurred. Ages of onset as assessed by these 2 methods were highly correlated (r.80; P.001). In this study, we used patients self-reported age, because it was derived from the same patient assessment instrument that also provided retrospective data on course of illness and time to first treatment. The 480 patients were assessed on various cross-sectional measures 18,20 during each visit, as well as on the daily prospective National Institutes of Mental Health Life Chart Method (NIMH-LCM). 21 Because the severity of mania and depression ratings on the NIMH-LCM are based on the degree of mood-related functional impairment in a patient s usual social, educational, or occupational roles, 21 they are readily remembered by the patient and rated by a clinician during the clinical interview. Such prospective NIMH-LCM clinician ratings are highly reliable over a period of several weeks to 1 month between visits and have been validated against more conventional cross-sectional rating scales, as described by Denicoff et al. 22,23 Sequentially admitted patients who also had a full year of prospective daily clinician ratings were included in the analysis. Because preliminary analysis suggested potential nonlinear relationships between age of onset and various outcome measures, age of onset was divided into 4 groups as was done by Perlis et al. 13 Onset at age 0 to 12 years was considered childhood onset; at 13 to 18 years, adolescent onset; at 19 to 29 years, early-adult onset; and after 30 years, late-adult onset. Average severity of days ill (with euthymic days included) for the prospective year was calculated using the mood ratings for mania and depression on the NIMH-LCM on a scale of 0 to 4, with 0 euthymic (not ill); 1 mild severity, with little or no role dysfunction; 2 low moderate (some dysfunction); 3 high moderate (much dysfunction); and 4 severe (essentially incapacitated). 21 Days with ultradian cycling (ie, switching between mania and depression within a 24-hour period) were recorded separately and not counted in the assessment of number of episodes. The normality of continuous measures was examined using the Kolmogorov-Smirnov test, and homogeneity of variance was examined using Levene s test. Continuous demographic and outcome measures were examined using 1-way analysis of variance with the 4 onset groups. Bonferroni-corrected pairwise comparisons were used post hoc. A 2 test was used with categorical measures, with omnibus significance followed up with a test to determine the location of differences. All P values were evaluated for significance at.05 (2-tailed). Results were Bonferroni-corrected for the 15 measures compared across onset groups. RESULTS The mean age at network entry ( years) was not significantly different among the patients with childhood, adolescent, and early-adult onset; however, the late-adult onset group was significantly older at network entry than the other groups (Table I). Of the 480 individuals with at least 1 year of complete prospective daily ratings, 65 (14%) reported onset at age 12 years or younger, 171 (36%) reported onset during adolescence, 154 (32%) reported early-adult onset, and 90 (19%) reported late-adult onset. 486 Leverich et al The Journal of Pediatrics May 2007

3 Table I. Patient demographics and retrospective illness variables as a function of age of onset Variable Childhood (0 to 12 years) Adolescent (13 to 18 years) Early adulthood (19 to 29 years) Late adulthood (30 years) 2 P n (percent) 65 (14%) 171 (36%) 154 (32%) 90 (19%) Mean age at first symptom, years SD Mean age at network entry, F years SD Female 42 (65) 98 (57) 78 (51) 45 (50) Bipolar type I 54 (83) 132 (80) 125 (83) 63 (73) II 11 (17) 33 (20) 26 (17) 23 (27) Married/cohabitating 28 (46) 82 (50) 78 (30) 49 (50) Parental history of bipolar 23 (47) 52 (37) 27 (21) 12 (17) disorder Parental history of depression 25 (52) 60 (43) 39 (29) 18 (25) Physical abuse as a child 22 (36) 36 (22) 21 (15) 10 (12) Sexual abuse as a child 19 (31) 33 (20) 15 (11) 8 (10) Dysphoric mania 48 (79) 94 (59) 68 (46) 46 (52) Rapid cycling (lifetime) 45 (70) 83 (51) 66 (45) 35 (40) Number of mood episodes 50 (83) 105 (66) 63 (44) 30 (38) ( 20 lifetime) Comorbidities: Lifetime anxiety disorder 38 (60) 63 (37) 56 (37) 21 (24) Lifetime alcohol abuse or 27 (42) 60 (35) 46 (30) 22 (25) dependence Lifetime drug abuse (excluding alcohol) 16 (25) 42 (25) 30 (20) 7 (8) Includes patients with 1 full year (365 days) of prospective life chart data. Figure. Childhood- and adolescent-onset bipolar disorders are associated with extremely long delays from first symptoms to first treatment for either manic or depressive symptoms. Total n 420. Plotted is the average delay per onset group 1 standard error of the mean. The duration from onset of illness to first pharmacologic treatment for depression or mania, available in 420 patients, was strongly inversely related to age of onset (Figure). The lag averaged years in those with childhood onset and years in those with adolescent onset. In contrast, those with onsets in early and late adulthood received their first treatment after average delays of only years and years, respectively. As shown in Table I, patients with childhood-onset bipolar disorder had the highest incidence of parental history of bipolar illness and unipolar depression, and these percentages declined progressively in patients with later age of onset. Environmental adversity, also a potential vulnerability factor, was higher in those with early onsets of bipolar disorder, as revealed by the incidence of physical or sexual abuse in childhood. A higher percentage of patients with childhood and adolescent onset reported having more than 20 affective episodes before network entry compared with those with adult onset. The incidence of dysphoric (as opposed to euphoric) mania was also substantially higher in patients with childhood onset (ie, mania was unpleasant, with more anxious and depressive symptoms). Among the comorbidities, patients with childhood onset had a higher prevalence of lifetime anxiety disorders and of drug abuse, with a nonsignificant trend in the same direction for alcohol abuse. As shown in Table II, in the first year of daily prospective ratings during naturalistic treatment, patients with childhood onset remained substantially more ill than those with adult onset. The childhood-onset group had greater severity of both mania and depression, as well as more days depressed and days with ultradian cycling. They also had more total The Poor Prognosis of Childhood-Onset Bipolar Disorder 487

4 Table II. Outcome in childhood- and adolescent-onset bipolar disorder: One-year prospective follow-up on the NIMH-LCM (n 480) Variable Childhood (0 to 12 years) Adolescent (13 to 18 years) Early adulthood (19 to 29 years) Late adulthood (30 years) F P Total days (hypo-) manic Total days depressed Days cycling Days euthymic Mean severity of depression* Mean severity of mania* Mean number of episodes (DSM IV) *1, mild; 2, low moderate; 3, high moderate; 4, severe on the NIMH-LCM. affective episodes during the prospective year and fewer days well (euthymic). DISCUSSION Our findings reveal that approximately 50% of adult outpatients with bipolar disorder had an onset of illness in either childhood or adolescence (ie, before age 19), and these individuals had long lags before initial treatment for mania or depression. These findings in a clinically identified cohort are convergent with recent epidemiologic data reported by Kessler et al 27 demonstrating a substantial incidence of early onset and long lag to treatment. However, Kessler et al could not determine whether or not the lag to first treatment reflected the presence of mild symptomatology and had little functional impact. Our combined retrospective and prospective data indicate that early onset of bipolar disorder is associated with an adverse lifetime course of illness that extends into 1 year of naturalistic treatment in adulthood. When we analyzed data by site, we found that the sites in the United States had a significantly greater rate of childhood- and adolescent-onset bipolar disorder than those in Europe, as discussed in detail elsewhere. 28 When considering only the US sites, our findings (62% onset in childhood or adolescence) are even more similar to those of Perlis et al, 13 who found this onset pattern in 65% of their US participants. Our data are also consistent with findings of Perlis et al 13 indicating that earlier age of onset was associated with a more difficult and complicated course of illness, including faster cycling, more days depressed, greater lifetime manic and depressive episodes, increased risk of substance abuse and other comorbidities, and greater lifetime risk of suicide attempts. In our study, those with childhood onset had more mood episodes, as well as more anxiety disorder and substance abuse comorbidity (Table I). Furthermore, they also experienced more dysphoric mania and more ultradian cycling than the other groups. These 2 retrospective accounts and the report by Lish et al 29 of a more severe course of bipolar disorder in patients with earlier illness onset are now validated by clinician-rated daily prospective follow-up as adults. During naturalistic treatment (i.e., treatment as usual), those patients with early onset had more time depressed and more severe mania and depression, as well as less time well. The fact that these patients experienced a greater number of episodes in this 1 year of prospective follow-up indicates that the greater proportion of early-onset patients with high numbers of episodes reported retrospectively is not just an artifact of a greater number of years of potential exposure (ie, duration of illness). Kessler et al, 30,31 in their epidemiologic data from 983 community survey participants in the National Comorbidity Study Replication, found that 13% of bipolar patients had an onset of illness by age 10 and that time lags between illness onset and first treatment correlated inversely with age of onset. The data from our cohort and that of Perlis et al 13 indicate that 14% to 28% of patients with bipolar disorder experience onset of illness before age 13 and that 50% to 67% experience onset before age 19. Lish et al, 29 in a study of bipolar patients surveyed from an affective illness advocacy group, also found that early onset of illness and long delay to appropriate diagnosis and treatment were very common. All of these findings taken together with the longest lags in onset of first treatment in those with the earliest onsets suggest that very early-onset bipolar disorder was quite common even a generation ago, but was relatively underrecognized and undertreated at that time. Lange and McGinnis 32 reviewed evidence likely indicating both cohort (year of birth) and anticipation (generational) effects in childhood-onset bipolar illness, possibly yielding even more children with early onset today than were found several decades ago when the adults in our study were at risk. Consequently, providers who assess and treat children and adolescents in various health care settings, including primary and family practice, pediatrics, and neurology, who see many children with ADHD and other externalizing disorders, should be particularly alert to the possibility of bipolar disorder in the differential diagnosis. Such vigilance may begin to shorten what were the extraordinary long delays to first treatment some 20 years ago. Keep in mind, however, that even in adults, in whom there is much less diagnostic controversy, only 9.8% of those in a primary care clinic who screened positive for bipolar symptomatology were recognized as potentially having this disorder. 33 Similarly, the underdi- 488 Leverich et al The Journal of Pediatrics May 2007

5 agnosis and treatment of bipolar illness in adults is apparent in epidemiologic samples ,34,35 Most strikingly, even when bipolar illness is diagnosed, adults in the community are often given antidepressants without concomitant mood stabilizers, 36 contrary to the recommendation in most treatment guidelines Instead of being a diagnosis of exclusion and last resort in children, as has often been the case, bipolar disorder should be actively considered and ruled in or out. This is particularly important in children at high risk due to having 1 or both parents with a diagnosis of bipolar illness (as seen in this study and in the studies of Lapalme et al 41 and Pavuluri et al 12 )or the presence of environmental stressors, such as the occurrence of traumatic events in childhood Most children with bipolar disorder also have a comorbid diagnosis of ADHD, whereas the vast majority of children with ADHD do not have bipolar disorder. 6,12,42,43 Nonetheless, in patients with apparent ADHD and additional indicators of extreme mood lability and behavioral dyscontrol (especially in the presence of periods of euphoria, decreased need for sleep, increased sexual interest or acts, delusions, hallucinations, or suicidal behaviors), providers should carefully consider the possibility of bipolar disorder before initiating treatment with stimulants and antidepressants alone, because first treatment with mood stabilizers or atypical antipsychotics is the recommended approach in consensus guidelines. 6 Inadequate recognition, diagnosis, and treatment of adult-onset bipolar illness in clinical samples and in the community now appear to be paralleled by equal problems in recognizing and treating childhood-onset bipolar illness, particularly in the United States. Given the likely adverse consequences of long periods of untreated or inadequately treated mood-related symptomatology, as seen in this and other studies, 44,45 efforts to ensure earlier diagnosis and institution of treatment are important. Whether such earlier intervention will in fact ameliorate the long-term difficulties associated with childhood-onset bipolar disorder remains to be ascertained. Additional statistical support was provided by Sun Hwang at the Department of Biostatistics, School of Public Health, UCLA, Los Angeles, CA. We thank Chris Gavin for his editorial support. REFERENCES 1. Geller B, Tillman R, Craney GL, Bolhofner K. Four-year prospective outcome and natural history of mania in children with a prepubertal and early adolescent bipolar disorder phenotype. Arch Gen Psychiatry 2004;61: Geller B, Craney GL, Bolhofner K, Nickelsburg MJ, Williams M, Zimmerman B. Two-year prospective follow-up of children with a prepubertal adolescent bipolar disorder phenotype. Am J Psychiatry 2002;159: Birmaher B, Axelson D, Strober M, Gill MK, Valeri S, Chiapetta L, et al. Clinical course of children and adolescents with bipolar spectrum disorders. Arch Gen Psychiatry 2006;63: Findling RL, McNamara NK, Youngstrom EA, Stansbrey R, Gracious BL, Reed MD, et al. Double-blind 18-month trial of lithium versus divalproex maintenance treatment in pediatric bipolar disorder. J Am Acad Child Adolesc Psychiatry 2005;44: National Institute of Mental Health research roundtable on prepubertal bipolar disorder. J Am Acad Child Adolesc Psychiatry 2001;40: Kowatch RA, Fristad M, Birmaher B, Wagner KD, Findling RL, Hellander M. The Child Psychiatric Workgroup on Bipolar Disorder: treatment guidelines for children and adolescents with bipolar disorder. J Am Acad Child Adolesc Psychiatry 2005;44: Carlson G. Early-onset bipolar disorder: clinical and research considerations. J Clin Child Adolesc Psychology 2005;34: Biederman J. Resolved. Mania is mistaken for ADHD in prepubertal children: affirmative. J Am Acad Child Adolesc Psychiatry 1998;37: American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. Washington, DC: American Psychiatric Association; Kramlinger KG, Post RM. Ultra-rapid and ultradian cycling in bipolar affective illness. Br J Psychiatry 1996;168: Findling RL, Gracious BL, McNamara NK, Youngstrom EA, Demeter CA, Branicky LA, et al. Rapid, continuous cycling and psychiatric co-morbidity in pediatric bipolar I disorder. Bipolar Disord 2001;3: Pavuluri MN, Birmaher B, Naylor MW. Pediatric bipolar disorder: a review of the past 10 years. J Am Acad Child Adolesc Psychiatry 2005;44: Perlis RH, Miyahara S, Marangell LB, Wisniewski SR, Ostacher M, DeBello MP, et al. Long-term implications of early onset in bipolar disorder: data from the first 1000 participants in the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD). Biol Psychiatry 2004;55: Leverich GS, Nolen WA, Rush AJ, McElroy SL, Keck PE, Denicoff KD, et al. The Stanley Foundation Bipolar Treatment Outcome Network: longitudinal methodology. J Affect Disord 2001;67: Post RM, Nolen WA, Kupka RW, Denicoff KD, Leverich GS, Keck PE Jr, et al. The Stanley Foundation Bipolar Network, 1: rationale and methods. Br J Psychiatry 2001;178:S Suppes T, Leverich GS, Keck PE, Nolen WA, Denicoff KD, Altshuler LL, et al. The Stanley Foundation Bipolar Treatment Outcome Network, II: demographics and illness characteristics of the first 261 patients. J Affect Disord 2001;67: Kupka RW, Nolen WA, Altshuler LL, Denicoff KD, Leverich GS, Keck PE Jr, et al. The Stanley Foundation Bipolar Network, 2: preliminary summary of demographics, course of illness and response to novel treatments. Br J Psychiatry Suppl 2001;42:s Nolen WA, Luckenbaugh DA, Altshuler LL, Suppes T, McElroy SL, Frye MA, et al. Correlates of 1-year prospective outcome in bipolar disorder: results from the Stanley Foundation Bipolar Network. Am J Psychiatry 2004;161: First MB, Spitzer RL, Gibbon M, Williams JB: Structured clinical interview for DSM-IV axis I disorders, patient edition (research version). New York: Biometrics Research Department, New York State Psychiatric Institute; Kupka RW, Luckenbaugh DA, Post RM, Suppes T, Altshuler LL, Keck PE Jr, et al. A comparative study of rapid and non-rapid cycling bipolar disorder using daily mood ratings in 539 outpatients. Am J Pychiatry 2005;162: Leverich GS, Post RM: Life charting of affective disorders. CNS Spectrum 1998;3: Denicoff KD, Leverich GS, Nolen WA, Rush AJ, McElroy SL, Keck PE Jr, et al. Validation of the prospective NIMH-Life Chart Method (NIMH-LCMp) for longitudinal assessment of bipolar illness. Psychol Med 2000;30: Denicoff KD, Smith-Jackson EE, Disney ER, Suddath RL, Leverich GS, Post RM. Preliminary evidence of the reliability and validity of the prospective life-chart methodology (LCMp). J Psychiatry Res 1997;3: Leverich GS, McElroy SL, Suppes T, Keck PE Jr, Denicoff KD, Nolen WA, et al. Early physical or sexual abuse and the course of bipolar illness. Biol Psychiatry 2002;51: Leverich GS, Post RM. Course of bipolar illness after history of childhood trauma. Lancet 2006;367: Garno JL, Goldberg JF, Ramirez PM, Ritzler BA. Impact of childhood abuse on the clinical course of bipolar disorders. Br J Psychiatry 2005;186: Kessler RC, Demler O, Frank RG, Olfson M, Pincus HA, Walters EE, et al. Prevalence and treatment of mental disorders, N Engl J Med 2005; Post RM, Luckenbaugh DA, Leverich GS, Altshuler LL, Frye MA, Suppes T, et al. Increased rate of childhood onset bipolar illness in the US compared with two European countries. Presented at the International Conference on Bipolar Illness, Barcelona, Spain, October 6, Lish JD, Dime-Meenan S, Whybrow PC, Price RA, Hirschfeld RMA. The National Depressive and Manic-Depressive Association (DMDA) survey of bipolar members. J Affect Disord 1994;31: Kessler RC, Chiu WT, Demler O, Walters EE. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry 2005;62: Kessler RC, Berglund P, Demler O, Jin R, Walters EE. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatry 2005;62: Lange KJ, McInnis MG. Studies of anticipation in bipolar affective disorder. CNS Spectrum 2002;7: The Poor Prognosis of Childhood-Onset Bipolar Disorder 489

6 33. Das AK, Olfson M, Gameroff MJ, Pilowsky DJ, Blanco C, Feder A, et al. Screening for bipolar disorder in a primary care practice. JAMA 2005;293: Hirschfeld RM, Calabrese JR, Weissman MM, Reed M, Davies MA, Frye MA, et al. Screening for bipolar disorder in the community. J Clin Psychiatry 2003;64: Kessler RC, McGonagle KA, Zhao S, Nelson CB, Hughes M, Eshleman S, et al. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States: results from the National Comorbidity Survey. Arch Gen Psychiatry 1994;51: American Psychiatric Association. Practice guideline for the treatment of patients with bipolar disorder (revision). Am J Psychiatry 2002;159(4 Suppl): Grunze H, Kasper S, Goodwin G, Bowden C, Baldwin D, Licht R, et al. World Federation of Societies of Biological Psychiatry Task Force on Treatment Guidelines for Bipolar Disorder, part I: treatment of bipolar depression. World J Biol Psychiatry 2002;3: Grunze H, Kasper S, Goodwin G, Bowden C, Baldwin D, Licht R, et al. World Federation of Societies of Biological Psychiatry Task Force on Treatment Guidelines for Bipolar Disorders, part II: treatment of mania. World J Biol Psychiatry 2003;4: Goodwin GM, Young AH. The British Association for Psychopharmacology guidelines for treatment of bipolar disorder: a summary. J Psychopharmacol 2003;17(4 Suppl): Keck PE Jr, Perlis RH, Otto MW, Carpenter D, Ross R, Docherty JP. The expert consensus guideline series: treatment of bipolar disorders. Postgrad Med 2004;Dec: Lapalme M, Hodgins S, LaRoche C. Children of parents with bipolar disorder: a meta-analysis of risk for mental disorders. Can J Psychiatry 1997;42: Post RM, Chang KD, Findling RL, Geller B, Kowatch RA, Kutcher SP, et al. Prepubertal bipolar I disorder and bipolar disorder NOS are separable from ADHD. J Clin Psychiatry 2004;65: Biederman J, Faraone SV, Wozniak J, Mick E, Kwon A, Cayton GA, et al. Clinical correlates of bipolar disorder in a large referred sample of children and adolescents. J Psychiatric Res 2005;39: Wilens TE, Biederman J, Millstein RB, Wozniak J, Hahesy AL, Spencer TJ. Risk for substance use disorders in youths with child- and adolescent-onset bipolar disorder. J Am Acad Child Adolesc Psychiatry 1999;38: Wilens TE, Biederman J, Kwon A, Ditterline J, Forkner P, Moore H, et al. Risk of substance use disorders in adolescents with bipolar disorder. J Am Acad Child Adolesc Psychiatry 2004;43: Leverich et al The Journal of Pediatrics May 2007