General Hospital Psychiatry 36 (2014) 347–351
Contents lists available at ScienceDirect
General Hospital Psychiatry journal homepage: http://www.ghpjournal.com
The use of 15-point hypomanic checklist in differentiating bipolar I and bipolar II disorder from major depressive disorder☆ Hongbo He, M.D., Ph.D. a, Guiyun Xu, M.D. a, b,⁎, Bin Sun, M.S. a, Huiyi Ouyang, R.N. a, Yamei Dang, M.D. b, Yangbo Guo, M.D., Ph.D. a, b, Guodong Miao, M.D., Ph.D. b, Catherine Rios, Ph.D. c, Hagop S. Akiskal, M.D. c, Kangguang Lin, M.D. b, d, e,⁎ a
Psychiatric Neuroscience Research Institute, Guangzhou Brain Hospital, Affiliated hospital of Guangzhou Medical University Department of Affective Disorder, Guangzhou Brain Hospital, Affiliated Hospital of Guangzhou Medical University c International Mood Center, University of California, San Diego, La Jolla, CA, USA d Laboratory of Neuropsychology, The University of Hong Kong, Hong Kong e Laboratory of Cognitive Affective Neuroscience, The University of Hong Kong, Hong Kong b
a r t i c l e
i n f o
Article history: Received 12 October 2013 Revised 11 December 2013 Accepted 13 December 2013 Keywords: Bipolar disorder Major depressive disorder Psychometrics 15-Point hypomanic checklist
a b s t r a c t Objectives: Individuals with bipolar disorder (BP) are often misdiagnosed with major depressive disorder (MDD). In this study, we developed a Chinese version of 15-point hypomania scale (HCL-15) in order to determine its sensitivity and specificity in the diagnosis of BP and BP-II in particular. Methods: A total of 623 individuals suffering a major depressive episode (MDE) were systematically interviewed with both Structured Clinical Interview for DSM-IV-TR Axis I Disorders, Patient Edition, and HCL15. A cutoff score of 8 or more in HCL-15 was suggested for BP. Results: Of the 623 depressed patients, 115 (18.5%) actually required a diagnosis of BP-I, and another 159 (25.5%) could be more appropriately diagnosed with BP-II, according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, criteria. The sensitivity of 15-HCL in detection of BP-II was 0.78 and 0.46 for BP-I; the specificity was 0.9 and 0.69, respectively. The specificity of HCL-15 for BP versus MDD was as high as 0.93. Approximately 60%–80% of all questions in the HCL-15 questionnaire revealed positive responses from patients, while items 11 and 12, measuring the consumption of alcohol, coffee and cigarettes, demonstrated a low positive response rate. Conclusions: The HCL-15 assessment scale was fairly sensitive and highly specific for a BP-II diagnosis but not for a BP-I diagnosis. Some items in the HCL-15 symptom list need to be further modified to better fit Chinese culture and customs. The HCL-15 scale could be a useful tool in clinical practice for screening individuals with BP-II in order to avoid a misdiagnosis of MDD. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Bipolar disorder (BP) is a mood disorder characterized by the presence of manic (in BP-I) or hypomanic (in BP-II) episodes in addition to episodes of major depression. Hypomania, in contrast to mania, does not cause marked impairment of social functioning in individuals with BP. For some patients, the change in behavior caused by a hypomanic episode might actually result in great productivity, improved efficiency or unexpected creativity [1,2]. Consequently, most patients with hypomania lack insight into the active manifestation and consequences of their hypomanic symptoms [2]. In fact, ☆ Conflict of interest: no conflict of interest in this study. ⁎ Corresponding authors. Guiyun Xu is to be contacted at Department of Affective disorder, Guangzhou Brain Hospital, Affiliated Hospital of Guangzhou Medical University, Guangzhou, Guangdong Province 510370, China. Tel.: +86 18922165291; fax: + 86 020 81891391. Kangguang Lin, Laboratory of Neuropsychology, The University of Hong Kong, Hong Kong. Tel.: +852 59897456. E-mail addresses: [email protected] (G. Xu), [email protected] (K. Lin). 0163-8343/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.genhosppsych.2013.12.008
most patients with BP, especially those with BP-II, seek clinical assistance for depressive symptoms, not hypomanic symptoms [3]. Patients with current depressive symptoms, therefore, should be properly assessed for possible BP in order to avoid a misdiagnosis of major depressive disorder (MDD). An accurate tool for differentiating these disorders is crucial in this clinical assessment. The current diagnostic gold standard, the Structured Clinical Interview for DSM-IV (SCID), is typically not sufficient for detecting hypomanic symptoms [4]. Due to this, several supplementary selfreport screening questionnaires have been developed to improve the detection of BP [5–7]. Included among these is the 32-item hypomania symptom checklist (HCL-32), one of the most widely used scales that are specifically sensitive in detecting hypomania [6,8–14]. The Chinese version of HCL-32 has also proven useful in detecting BP from MDD [10,13,14]. In most Chinese outpatient clinics, however, the average clinician sees over 20 patients in a 3-h session without a nurse helper. The HCL32 usually takes at least 6 to 10 min for a patient to complete, and
348
H. He et al. / General Hospital Psychiatry 36 (2014) 347–351
some items become hard to recall, thus hindering its clinical application. This has led to many misdiagnoses in Chinese outpatient clinics. A shortened and simplified questionnaire is necessary for a more proper diagnosis. Smith et al. once modified the HCL-32 into a 15-item hypomania checklist (HCL-15) that proves valuable in detecting BP from MDD [5]. However, the sensitivity and specificity of HCL-15 have not been tested in western countries or in China. The most recent multicenter trial in China revealed that 1487 patients diagnosed with and being treated for MDD in 13 mental health centers across China could actually be more properly diagnosed with BP according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), criteria [6]. An improved diagnostic tool for detecting BP is crucial. The purpose of this study was to examine the generalizability and practicability of the HCL-15 assessment scale in screening BP in patients currently suffering a major depressive episode (MDE).
consensus of the clinical impression, the SCID interview and a review of medical records. In phase II, the forgoing psychiatrists interviewed and evaluated the patients with systematic clinical scales. As an additional quality control, a group of three senior psychiatrists assigned by the Guangzhou Brain Hospital conducted random inspections. The study was approved by the ethics committee of Guangzhou Brain Hospital and registered at China clinical trial (ChiCTR-TNRC-10001112, http://www.chictr.org/). All the participants gave written informed consent. The present analysis comprised 115 patients with BP-I, 159 patients with BP-II and 349 patients with MDD. According to the exclusion criteria of this study, the patients were free of the following conditions: pregnancy, serious general medical illness, history of seizure disorder, DSM-IV-TR-defined organic mental disorders, dementia, schizophrenia, delusional disorder, schizoaffective disorder, active substance use disorder and history of mental retardation. 2.2. Assessments
2. Methods 2.1. Descriptions of the study and the sample As a part of the Clinical and Biological Characteristics and Optimizing Treatment in Bipolar Depressive Disorder project (June 2007 to November 2010), this study aimed to improve the detection of BP in the context of major depression in terms of clinical features and biological markers of BP, as well as to optimize treatments for BPs largely in terms of functional outcome [8]. This project was a prospective, 6-week, open-label trial on patients with MDD or BP during an MDE. The trial consisted of two phases: a 1- to 7-day screening period and a 6-week seminaturalistic treatment period. In phase I, one full-time research psychiatrist for the study (who had been practicing for 10 years) conducted the clinical interviews with inpatients or outpatients receiving psychiatric services in the two tertiary hospitals (e.g., university teaching hospital) — the Guangzhou Brain Hospital and the first affiliated hospital of Jinan University — referred by their first contact psychiatrists when they were diagnosed with MDE. The research psychiatrist then applied the Chinese version of the Structured Clinical Interview for DSM-IV-TR Axis I Disorders, Patient Edition (SCID-I/P), to confirm the diagnoses [15]. Another senior psychiatrist conducted an independent clinical interview. We diagnosed the patients based on the combination of the
HCL-15 was developed based on the work by Angst et al. [16] and Hantouche et al. [12] that led up to the development of the HCL-32, with a cutoff of 8 or more symptoms. The English version of the HCL15 was first translated into simplified Chinese by the author of this study, and the back translation was performed by a bilingual psychiatrist who was unaware of the content of the original HCL-15. After SCID-I/P evaluation, all enrolled patients were screened with HCL-15. Data collected also included age, gender, education, drug, alcohol, coffee and cigarettes usage history. 2.3. Statistical analyses χ 2 tests or analysis of variance was used to compare differences between the three groups (MDD, BP-I and BP-II) on potentially confounding sociodemographic characteristics. The internal consistency of HCL-15 was measured by Cronbach's alpha coefficient. Factor loading was calculated using exploratory factor analysis with oblique rotation; the cutoff coefficient was set at 0.4. The receiver operating characteristic (ROC) curve was used to evaluate the differential diagnostic effect of HCL-15 and to determine the sensitivity and specificity. All analyses were performed using SPSS13.0. Statistical significance was evaluated at the .05 level for all the tests.
Table 1 Comparison of social demographic characteristics between BP-I, BP-II and MDD patient group (N=623) Characteristic
Unipolar
BP-I
BP-II
Total
Sample size Age Gender (male) Educational years Marital status Married Cohabitation Divorced or widowed Single Alcohol usage None Occasional Addiction or worse Unknown Cigarette usage None Occasional Addiction or worse Unknown Physical illness (yes) Family history (yes)
349 (56.0%) 35.6±13.2 150 (43.0%) 10.9±4.1
115 (18.5%) 32.4±11.6 69 (60.0%) 11.9±4.0
159 (25.5%) 31.9±12.5 73 (45.9%) 12.2±4.2
623 (100%) 34.1±12.9 292 (46.9%) 11.4±4.1
193 (55.3%) 7 (2.0%) 14 (4.0%) 135 (38.7%)
53 (46.1%) 3 (2.6%) 6 (5.2%) 53 (46.1%)
62 (39.0%) 5 (3.1%) 11 (6.9%) 81 (50.9%)
308 (49.4%) 15 (2.4%) 31 (5.0%) 269 (43.2%)
272 (77.9%) 70 (20.1%) 7 (2%)
85 (73.9%) 28 (24.3%) 2 (1.7%)
96 (60.8%) 53 (33.5%) 9 (5.7%)
461 (72.6%) 155 (24.4%) 18 (2.8%) 1 (0.2%)
280 (80.7%) 38 (11.0%) 29 (8.3%)
87 (75.7%) 12 (10.4%) 16 (13.9%)
116 (73.9%) 20 (12.7%) 21 (13.4%)
97 (27.8%) 126 (36.2%)
20 (17.4%) 47 (41.2%)
40 (25.3%) 71 (44.7%)
483 (77.5%) 70 (11.2%) 66 (10.6%) 4 (0.7%) 157 (25.2%) 244 (39.2%)
Statistics
P values
F=5.905 χ2=10.141 F=5.889 χ2=12.759
.003** .006** .003** .047*
χ2=18.347
.001**
χ2=5.098
.277
χ2=4.961 χ2=3.484
.084 .175
H. He et al. / General Hospital Psychiatry 36 (2014) 347–351
349
3. Results 3.1. Sample description The sample included 623 subjects: 349 MDD (56.0%), 115 BP-I (18.5%) and 159 BP-II (25.5%). Sociodemographic characteristics were significantly different across the three groups (Table 1). Of the 623 participants, 292 (46.9%) were male and 331 (53.1%) were female, with a mean age and education of 34.1 years (S.D.=12.9) and 11.4 years (S.D.=4.1) respectively. There were 72.6% (461/623) participants who did not drink alcohol and 77.5% (483/623) participants who did not smoke cigarettes.
3.2. Factor structure
Fig. 1. Frequency of positive response in each item.
The exploratory factor analysis revealed a two-factor solution explaining 60.9% of the total variance. As shown in Table 2, factor 1 included items 1–9 and items 13–15, most of which were associated with “energetic/euphoric” features; factor 2 comprised three items (items 10–12), with item 10 related to “irritability” and the other two items associated with increasing consumption of coffee and cigarette, as well as alcohol. High internal consistency reliability (Cronbach's alpha) was found for factor 1 (alpha=0.946), with a less satisfactory value for factor 2 (alpha=0.389). Combining all the items together, the Cronbach's alpha coefficient was as high as 0.932.
3.3. Frequency of positive responses There were significant differences in most items among the three groups in terms of the frequency of positive responses (Fig. 1). The main exceptions were items 11 and 12, where all three groups gave much less positive responses. Ten items (items 2, 3, 4, 5, 6, 7, 8, 9, 13 and15, respectively) exhibited good capacity in differentiating the three Table 2 The Chinese version of the HCL-15 Signs and symptoms
Factor 1
Factor 2
1. Less sleep. 睡眠减少。 2. More drive and energy. 欲望增多和精力旺盛。 3. More self-confidence. 过分自信。 4. Increased social activity and work motivation. 社会活动增多和工作积极性增加。 5. Increased physical activity. 体力/体育活动增多。 6. More plans and ideas. 计划和设想较多。 7. Less shy, less inhibited. 很少害羞、很少拘束。 8. More talkative than usual. 比平时健谈 (话多)。 9. More puns and jokes, faster thinking, laughing more. 双关语和笑话增多、思维增快、常大笑。 10. More irritable, impatient. 较容易发怒和急躁。 11. Increased consumption of coffee, cigarettes. 喝咖啡和吸烟增多。 12. Increased consumption of alcohol. 饮酒增多。 13. Extremely happy mood, overeuphoric. 过分快乐和欣喜若狂。 14.Increased sex drive, interest in sex. 更喜欢参加与性有关的活动、性欲增加。 15. Over-activity (e.g., shopping, business, telephone calls, travelling, driving, visiting people). 活动过多 (如,购物、公干、打电话、旅游、驾车、访友等)
0.587
0.343
0.865
0.280
0.865
0.269
0.878
0.220
0.875
0.215
0.826
0.349
0.842
0.244
0.855
0.248
0.749
0.205
0.367
0.573
0.188
0.753
0.184
0.713
0.781
0.205
0.494
0.341
0.827
0.259
Factor loading N0.4 is shown in bold.
groups, with 60%–80% positive response in the BP-II group, 40%–50% positive response in the BP-I group and 5%–20% in the MDD group. 3.4. ROC curve analysis In the ROC curve analysis, the area under the curve was 0.69 [95% confidence interval (CI): 0.63–0.75] for MDD vs. BP-I and 0.90 (95% CI: 0.86–0.93) for MDD vs. BP-II (Fig. 2). The HCL-15 screening score of 7 was the optimal cutoff, with the sensitivity of 0.46 for BP-I and 0.79 for BP-II. The specificity reached 0.93, indicating that 93% of MDD patient who scored less than 7 in HCL-15 test truly did not have BP. 4. Discussion In our study, the mean age of MDD patients was significantly higher than that of BP patients, while the educational level of MDD patients was significantly lower than that of BP patients, and the percentage of female patients was comparatively higher in the MDD group. The result was consistent with the concept that BP patients tend to be younger. Better educational level in BP patients was consistent with our previous finding that cognitive impairment in patients with major depression during clinical remission stage was more severe than that in BP patients [8]. In previous factor analytic studies of hypomania, assessed with HCL-32, a dual-factor structure was consistently found, which characterized with “active/elated” and “risk-taking/irritable” items [9–13]. In our study, factor analysis of HCL-15 also revealed two factors: an “energetic/euphoric” factor and a mixed factor of irritability and substance use (cigarette, coffee and alcohol). The two factors virtually included all the items in the list, explaining more than 60% of the total variance. The internal consistent reliability of HCL-15 was as high as 0.93. It suggested that this simplified selfassessment tool has a good validity, although all the samples including MDD, BP-I and BP-II were combined in the analyses. The frequencies of positive responses of 15 items were generally high in BP-II patients, but items 11 (increased consumption of coffee, cigarettes) and 12 (increased consumption of alcohol) received comparatively low positive response in all participants. This was consistent with the demographic features of the study sample, which showed that 72.6% (461/623) of participants did not drink alcohol and 77.5% (483/623) of participants did not smoke cigarettes. Among the patients who did drink alcohol, 28.6% (18/63) BP-II patients recalled more consumption of alcohol, which was significantly higher than in BP-I patients (13.3%, 4/30) and MDD patients (13%, 10/77). Given the relatively low percentage of BP patients who used coffee, cigarettes and alcohol possibly due to Chinese culture and customs, we suggest to discard the two relevant items from the HCL-15. In light of the feature of affective instability overpresenting in BP versus MDD [17] — the core characteristic of cyclothymic temperament and whose
350
H. He et al. / General Hospital Psychiatry 36 (2014) 347–351
Fig. 2. ROC curve analyses. (A) HCL-15 exhibited good sensitivity and specificity in diagnosing BP-II (area under the curve=0.90) from MDD. (B) HCL-15 exhibited comparatively fair sensitivity and specificity (area under the curve=0.69) in diagnosing BP-I from MDD.
dysregulation may be the substrate for BP [18] — we suggest adding a domain with two additional items reflecting mood lability in the HCL15 for the future study. In this study, the cutoff score of 7 yielded a sensitivity of 0.79 between BP-II and MDD and a sensitivity of 0.46 between BP-I and MDD. The specificity reached 0.93. A substantial body of studies showed that, in different nations, including mainland China and Taiwan, HCL-32 displayed good sensitivity and specificity between BP and MDD, ranging from 0.68 to 0.85 and 0.46 to 0.91, respectively [13,14,19,20]. HCL-16, a shortened version of HCL-32, was reported to have a sensitivity of 0.83 and specificity of 0.71 between BP and MDD [21]. Since differentiation of BP-I and MDD is not a major concern in clinical practice, these results indicated that HCL-15 could be a very practical aid in differentiating BP-II and MDD. After 2 min of self-assessment, the clinician should be alerted by HCL scores of 7 or above that the currently depressed patient might need further evaluation in order to differentiate BP, especially BP-II, from MDD. If scores are less than 7, by 93% chance, this patient will probably be an MDD patient. Although HCL-32 has been widely tested in different nations with generally good evaluations, some clinicians complained that part of the scale was redundant and some items were confusing and hard to recall correctly. It was suggested by previous studies that a fundamental feature of hypomania was energized activity instead of the hypomanic mood [22]. Some studies have shown by factor analysis that, among multiple questions regarding hypomanic mood, only irritability was incorporated in the factor structure of hypomania, and it was suggested that, for the detection of BP-II, the focus of clinical questions on past hypomania should be on hypomanic behavior rather than hypomanic mood [11,23]. HCL-15 clearly focused more on hypomanic behavior, with only one question regarding the euphoria mood. It might be the reason that, with much less questions, HCL-15 exhibited even slightly better sensitivity and specificity in detecting BP-II from MDD, and all the items in the HCL15 were actually incorporated into the hypomanic factor structure. 4.1. Limitations Several limitations of this study deserve comments. Firstly, a limitation is that the main study tool, HCL-15, is a self-assessment instrument and requires patients to recall the past hypomanic experience. Secondly, all patients were in depressive episodes with or without psychotic symptoms while being assessed. The severity of symptoms, insight and some other clinical index may affect the results of recall, and we did not include these factors into our analyses. Thirdly, currently available HCL-15 study results were very limited, and the test–retest reliability was not known yet. These limitations
are counterbalanced by the fact that this paper is on an adult cohort of patients and does not have the limitation of having studied hypomania in adolescents where hypomania may be transient [24]. 4.2. Conclusion Despite the limitations mentioned above, this is the first study to test the validity of HCL-15. It yielded good sensitivity and specificity between BP, especially BP-II, and MDD. For some busy clinics, it can potentially be a practical choice for clinicians to use as a screening tool for each first-contact patient with current depressive episode. More studies from different areas are needed to further verify the effectiveness of HCL-15. Acknowledgments This study was supported by grants from (a) the Tackling Key Problems of Guangzhou Scientific and Technological Project (2007Z3E0611), Guangzhou Scientific and Technological Bureau of China; (2) Guangdong Scientific and Technological Project; (c) the Project of National Clinical Key Construction Specialty of China. References [1] American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed., Text Revision (DSM-IV-TR). Washington, DC: American Psychiatric Association; 2000. p. 365–8. [2] Simpson SG, McMahon FJ, McInnis MG, et al. Diagnostic reliability of bipolar II disorder. Arch Gen Psychiatry 2002;59:736–40. [3] ten Have M, Vollebergh W, Bijl R, Nolen WA. Bipolar disorder in the general population in the Netherlands (prevalence, consequences and care utilisation): results from the Netherlands Mental Health Survey and Incidence Study (NEMESIS). J Affect Disord 2002;68:203–13. [4] Benazzi F, Akiskal HS. Refining the evaluation of bipolar II: beyond the strict SCIDCV guidelines for hypomania. J Affect Disord 2003;73:33–8. [5] Smith DJ, Harrison N, Muir W, Blackwood DH. The high prevalence of bipolar spectrum disorders in young adults with recurrent depression: toward an innovative diagnostic framework. J Affect Disord 2005;84:167–78. [6] Yang HC, Xiang YT, Liu TB, et al. Hypomanic symptoms assessed by the HCL-32 in patients with major depressive disorder: a multicenter trial across China. J Affect Disord 2012;143:203–7. [7] Yang HC, Yuan CM, Liu TB, et al. Validity of the Chinese version Mood Disorder Questionnaire (MDQ) and the optimal cutoff screening bipolar disorders. Psychiatry Res 2011;189:446–50. [8] Xu G, Lin K, Rao D, et al. Neuropsychological performance in bipolar I, bipolar II and unipolar depression patients: a longitudinal, naturalistic study. J Affect Disord 2012;136:328–39. [9] Meyer TD, Hammelstein P, Nilsson LG, Skeppar P, Adolfsson R, Angst J. The Hypomania Checklist (HCL-32): its factorial structure and association to indices of impairment in German and Swedish nonclinical samples. Compr Psychiatry 2007;48:79–87. [10] Wu YS, Angst J, Ou CS, Chen HC, Lu RB. Validation of the Chinese version of the hypomania checklist (HCL-32) as an instrument for detecting hypo(mania) in patients with mood disorders. J Affect Disord 2008;106:133–43.
H. He et al. / General Hospital Psychiatry 36 (2014) 347–351 [11] Benazzi F, Akiskal HS. The dual factor structure of self-rated MDQ hypomania: energized-activity versus irritable-thought racing. J Affect Disord 2003;73:59–64. [12] Hantouche EG, Angst J, Akiskal HS. Factor structure of hypomania: interrelationships with cyclothymia and the soft bipolar spectrum. J Affect Disord 2003;73: 39–47. [13] Yang HC, Yuan CM, Liu TB, et al. Validity of the 32-item Hypomania Checklist (HCL-32) in a clinical sample with mood disorders in China. BMC Psychiatry 2011;11:84. [14] Forty L, Smith D, Jones L, et al. Identifying hypomanic features in major depressive disorder using the hypomania checklist (HCL-32). J Affect Disord 2009;114:68–73. [15] First, MB, Spitzer, RL, Gibbon, M, Williams, JBW. Structured clinical interview for DSM-IV-TR axis I disorders—patient edition (SCID-I/P, 11/2002 revision). Biometrics Research Department, New York State Psychiatric Institute, New York, 2002. [16] 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;88:217–33. [17] Singh V, Bowden CL, Gonzalez JM, et al. Discriminating primary clinical states in bipolar disorder with a comprehensive symptom scale. Acta Psychiatr Scand 2013;127:145–52.
351
[18] Perugi G, Akiskal HS. The soft bipolar spectrum redefined: focus on the cyclothymic, anxious–sensitive, impulse–dyscontrol, and binge-eating connection in bipolar II and related conditions. Psychiatr Clin North Am 2002;25:713–37. [19] Chou CC, Lee IH, Yeh TL, et al. Comparison of the validity of the Chinese versions of the Hypomania Symptom Checklist-32 (HCL-32) and Mood Disorder Questionnaire (MDQ) for the detection of bipolar disorder in medicated patients with major depressive disorder. Int J Psychiatry Clin Pract 2012;16:132–7. [20] Haghighi M, Bajoghli H, Angst J, Holsboer-Trachsler E, Brand S. The Farsi version of the Hypomania Check-List 32 (HCL-32): applicability and indication of a fourfactorial solution. BMC Psychiatry 2011;11:14. [21] Forty L, Kelly M, Jones L, et al. Reducing the Hypomania Checklist (HCL-32) to a 16item version. J Affect Disord 2010;124:351–6. [22] Perugi G, Akiskal HS, Micheli C, et al. Clinical subtypes of bipolar mixed states: validating a broader European definition in 143 cases. J Affect Disord 1997;43:169–80. [23] Benazzi F, Akiskal HS, Micheli C, et al. Sensitivity and specificity of clinical markers for the diagnosis of bipolar II disorder. Compr Psychiatry 2001;42:461–5. [24] Paaren A, von Knorring A-L, Olsson G, et al. Hypomania spectrum disorders from adolescence to adulthood: a 15-year follow-up of a community sample. J Affect Disord 2013;145:190–9.
Contents lists available at ScienceDirect
General Hospital Psychiatry journal homepage: http://www.ghpjournal.com
The use of 15-point hypomanic checklist in differentiating bipolar I and bipolar II disorder from major depressive disorder☆ Hongbo He, M.D., Ph.D. a, Guiyun Xu, M.D. a, b,⁎, Bin Sun, M.S. a, Huiyi Ouyang, R.N. a, Yamei Dang, M.D. b, Yangbo Guo, M.D., Ph.D. a, b, Guodong Miao, M.D., Ph.D. b, Catherine Rios, Ph.D. c, Hagop S. Akiskal, M.D. c, Kangguang Lin, M.D. b, d, e,⁎ a
Psychiatric Neuroscience Research Institute, Guangzhou Brain Hospital, Affiliated hospital of Guangzhou Medical University Department of Affective Disorder, Guangzhou Brain Hospital, Affiliated Hospital of Guangzhou Medical University c International Mood Center, University of California, San Diego, La Jolla, CA, USA d Laboratory of Neuropsychology, The University of Hong Kong, Hong Kong e Laboratory of Cognitive Affective Neuroscience, The University of Hong Kong, Hong Kong b
a r t i c l e
i n f o
Article history: Received 12 October 2013 Revised 11 December 2013 Accepted 13 December 2013 Keywords: Bipolar disorder Major depressive disorder Psychometrics 15-Point hypomanic checklist
a b s t r a c t Objectives: Individuals with bipolar disorder (BP) are often misdiagnosed with major depressive disorder (MDD). In this study, we developed a Chinese version of 15-point hypomania scale (HCL-15) in order to determine its sensitivity and specificity in the diagnosis of BP and BP-II in particular. Methods: A total of 623 individuals suffering a major depressive episode (MDE) were systematically interviewed with both Structured Clinical Interview for DSM-IV-TR Axis I Disorders, Patient Edition, and HCL15. A cutoff score of 8 or more in HCL-15 was suggested for BP. Results: Of the 623 depressed patients, 115 (18.5%) actually required a diagnosis of BP-I, and another 159 (25.5%) could be more appropriately diagnosed with BP-II, according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, criteria. The sensitivity of 15-HCL in detection of BP-II was 0.78 and 0.46 for BP-I; the specificity was 0.9 and 0.69, respectively. The specificity of HCL-15 for BP versus MDD was as high as 0.93. Approximately 60%–80% of all questions in the HCL-15 questionnaire revealed positive responses from patients, while items 11 and 12, measuring the consumption of alcohol, coffee and cigarettes, demonstrated a low positive response rate. Conclusions: The HCL-15 assessment scale was fairly sensitive and highly specific for a BP-II diagnosis but not for a BP-I diagnosis. Some items in the HCL-15 symptom list need to be further modified to better fit Chinese culture and customs. The HCL-15 scale could be a useful tool in clinical practice for screening individuals with BP-II in order to avoid a misdiagnosis of MDD. © 2014 Elsevier Inc. All rights reserved.
1. Introduction Bipolar disorder (BP) is a mood disorder characterized by the presence of manic (in BP-I) or hypomanic (in BP-II) episodes in addition to episodes of major depression. Hypomania, in contrast to mania, does not cause marked impairment of social functioning in individuals with BP. For some patients, the change in behavior caused by a hypomanic episode might actually result in great productivity, improved efficiency or unexpected creativity [1,2]. Consequently, most patients with hypomania lack insight into the active manifestation and consequences of their hypomanic symptoms [2]. In fact, ☆ Conflict of interest: no conflict of interest in this study. ⁎ Corresponding authors. Guiyun Xu is to be contacted at Department of Affective disorder, Guangzhou Brain Hospital, Affiliated Hospital of Guangzhou Medical University, Guangzhou, Guangdong Province 510370, China. Tel.: +86 18922165291; fax: + 86 020 81891391. Kangguang Lin, Laboratory of Neuropsychology, The University of Hong Kong, Hong Kong. Tel.: +852 59897456. E-mail addresses: [email protected] (G. Xu), [email protected] (K. Lin). 0163-8343/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.genhosppsych.2013.12.008
most patients with BP, especially those with BP-II, seek clinical assistance for depressive symptoms, not hypomanic symptoms [3]. Patients with current depressive symptoms, therefore, should be properly assessed for possible BP in order to avoid a misdiagnosis of major depressive disorder (MDD). An accurate tool for differentiating these disorders is crucial in this clinical assessment. The current diagnostic gold standard, the Structured Clinical Interview for DSM-IV (SCID), is typically not sufficient for detecting hypomanic symptoms [4]. Due to this, several supplementary selfreport screening questionnaires have been developed to improve the detection of BP [5–7]. Included among these is the 32-item hypomania symptom checklist (HCL-32), one of the most widely used scales that are specifically sensitive in detecting hypomania [6,8–14]. The Chinese version of HCL-32 has also proven useful in detecting BP from MDD [10,13,14]. In most Chinese outpatient clinics, however, the average clinician sees over 20 patients in a 3-h session without a nurse helper. The HCL32 usually takes at least 6 to 10 min for a patient to complete, and
348
H. He et al. / General Hospital Psychiatry 36 (2014) 347–351
some items become hard to recall, thus hindering its clinical application. This has led to many misdiagnoses in Chinese outpatient clinics. A shortened and simplified questionnaire is necessary for a more proper diagnosis. Smith et al. once modified the HCL-32 into a 15-item hypomania checklist (HCL-15) that proves valuable in detecting BP from MDD [5]. However, the sensitivity and specificity of HCL-15 have not been tested in western countries or in China. The most recent multicenter trial in China revealed that 1487 patients diagnosed with and being treated for MDD in 13 mental health centers across China could actually be more properly diagnosed with BP according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), criteria [6]. An improved diagnostic tool for detecting BP is crucial. The purpose of this study was to examine the generalizability and practicability of the HCL-15 assessment scale in screening BP in patients currently suffering a major depressive episode (MDE).
consensus of the clinical impression, the SCID interview and a review of medical records. In phase II, the forgoing psychiatrists interviewed and evaluated the patients with systematic clinical scales. As an additional quality control, a group of three senior psychiatrists assigned by the Guangzhou Brain Hospital conducted random inspections. The study was approved by the ethics committee of Guangzhou Brain Hospital and registered at China clinical trial (ChiCTR-TNRC-10001112, http://www.chictr.org/). All the participants gave written informed consent. The present analysis comprised 115 patients with BP-I, 159 patients with BP-II and 349 patients with MDD. According to the exclusion criteria of this study, the patients were free of the following conditions: pregnancy, serious general medical illness, history of seizure disorder, DSM-IV-TR-defined organic mental disorders, dementia, schizophrenia, delusional disorder, schizoaffective disorder, active substance use disorder and history of mental retardation. 2.2. Assessments
2. Methods 2.1. Descriptions of the study and the sample As a part of the Clinical and Biological Characteristics and Optimizing Treatment in Bipolar Depressive Disorder project (June 2007 to November 2010), this study aimed to improve the detection of BP in the context of major depression in terms of clinical features and biological markers of BP, as well as to optimize treatments for BPs largely in terms of functional outcome [8]. This project was a prospective, 6-week, open-label trial on patients with MDD or BP during an MDE. The trial consisted of two phases: a 1- to 7-day screening period and a 6-week seminaturalistic treatment period. In phase I, one full-time research psychiatrist for the study (who had been practicing for 10 years) conducted the clinical interviews with inpatients or outpatients receiving psychiatric services in the two tertiary hospitals (e.g., university teaching hospital) — the Guangzhou Brain Hospital and the first affiliated hospital of Jinan University — referred by their first contact psychiatrists when they were diagnosed with MDE. The research psychiatrist then applied the Chinese version of the Structured Clinical Interview for DSM-IV-TR Axis I Disorders, Patient Edition (SCID-I/P), to confirm the diagnoses [15]. Another senior psychiatrist conducted an independent clinical interview. We diagnosed the patients based on the combination of the
HCL-15 was developed based on the work by Angst et al. [16] and Hantouche et al. [12] that led up to the development of the HCL-32, with a cutoff of 8 or more symptoms. The English version of the HCL15 was first translated into simplified Chinese by the author of this study, and the back translation was performed by a bilingual psychiatrist who was unaware of the content of the original HCL-15. After SCID-I/P evaluation, all enrolled patients were screened with HCL-15. Data collected also included age, gender, education, drug, alcohol, coffee and cigarettes usage history. 2.3. Statistical analyses χ 2 tests or analysis of variance was used to compare differences between the three groups (MDD, BP-I and BP-II) on potentially confounding sociodemographic characteristics. The internal consistency of HCL-15 was measured by Cronbach's alpha coefficient. Factor loading was calculated using exploratory factor analysis with oblique rotation; the cutoff coefficient was set at 0.4. The receiver operating characteristic (ROC) curve was used to evaluate the differential diagnostic effect of HCL-15 and to determine the sensitivity and specificity. All analyses were performed using SPSS13.0. Statistical significance was evaluated at the .05 level for all the tests.
Table 1 Comparison of social demographic characteristics between BP-I, BP-II and MDD patient group (N=623) Characteristic
Unipolar
BP-I
BP-II
Total
Sample size Age Gender (male) Educational years Marital status Married Cohabitation Divorced or widowed Single Alcohol usage None Occasional Addiction or worse Unknown Cigarette usage None Occasional Addiction or worse Unknown Physical illness (yes) Family history (yes)
349 (56.0%) 35.6±13.2 150 (43.0%) 10.9±4.1
115 (18.5%) 32.4±11.6 69 (60.0%) 11.9±4.0
159 (25.5%) 31.9±12.5 73 (45.9%) 12.2±4.2
623 (100%) 34.1±12.9 292 (46.9%) 11.4±4.1
193 (55.3%) 7 (2.0%) 14 (4.0%) 135 (38.7%)
53 (46.1%) 3 (2.6%) 6 (5.2%) 53 (46.1%)
62 (39.0%) 5 (3.1%) 11 (6.9%) 81 (50.9%)
308 (49.4%) 15 (2.4%) 31 (5.0%) 269 (43.2%)
272 (77.9%) 70 (20.1%) 7 (2%)
85 (73.9%) 28 (24.3%) 2 (1.7%)
96 (60.8%) 53 (33.5%) 9 (5.7%)
461 (72.6%) 155 (24.4%) 18 (2.8%) 1 (0.2%)
280 (80.7%) 38 (11.0%) 29 (8.3%)
87 (75.7%) 12 (10.4%) 16 (13.9%)
116 (73.9%) 20 (12.7%) 21 (13.4%)
97 (27.8%) 126 (36.2%)
20 (17.4%) 47 (41.2%)
40 (25.3%) 71 (44.7%)
483 (77.5%) 70 (11.2%) 66 (10.6%) 4 (0.7%) 157 (25.2%) 244 (39.2%)
Statistics
P values
F=5.905 χ2=10.141 F=5.889 χ2=12.759
.003** .006** .003** .047*
χ2=18.347
.001**
χ2=5.098
.277
χ2=4.961 χ2=3.484
.084 .175
H. He et al. / General Hospital Psychiatry 36 (2014) 347–351
349
3. Results 3.1. Sample description The sample included 623 subjects: 349 MDD (56.0%), 115 BP-I (18.5%) and 159 BP-II (25.5%). Sociodemographic characteristics were significantly different across the three groups (Table 1). Of the 623 participants, 292 (46.9%) were male and 331 (53.1%) were female, with a mean age and education of 34.1 years (S.D.=12.9) and 11.4 years (S.D.=4.1) respectively. There were 72.6% (461/623) participants who did not drink alcohol and 77.5% (483/623) participants who did not smoke cigarettes.
3.2. Factor structure
Fig. 1. Frequency of positive response in each item.
The exploratory factor analysis revealed a two-factor solution explaining 60.9% of the total variance. As shown in Table 2, factor 1 included items 1–9 and items 13–15, most of which were associated with “energetic/euphoric” features; factor 2 comprised three items (items 10–12), with item 10 related to “irritability” and the other two items associated with increasing consumption of coffee and cigarette, as well as alcohol. High internal consistency reliability (Cronbach's alpha) was found for factor 1 (alpha=0.946), with a less satisfactory value for factor 2 (alpha=0.389). Combining all the items together, the Cronbach's alpha coefficient was as high as 0.932.
3.3. Frequency of positive responses There were significant differences in most items among the three groups in terms of the frequency of positive responses (Fig. 1). The main exceptions were items 11 and 12, where all three groups gave much less positive responses. Ten items (items 2, 3, 4, 5, 6, 7, 8, 9, 13 and15, respectively) exhibited good capacity in differentiating the three Table 2 The Chinese version of the HCL-15 Signs and symptoms
Factor 1
Factor 2
1. Less sleep. 睡眠减少。 2. More drive and energy. 欲望增多和精力旺盛。 3. More self-confidence. 过分自信。 4. Increased social activity and work motivation. 社会活动增多和工作积极性增加。 5. Increased physical activity. 体力/体育活动增多。 6. More plans and ideas. 计划和设想较多。 7. Less shy, less inhibited. 很少害羞、很少拘束。 8. More talkative than usual. 比平时健谈 (话多)。 9. More puns and jokes, faster thinking, laughing more. 双关语和笑话增多、思维增快、常大笑。 10. More irritable, impatient. 较容易发怒和急躁。 11. Increased consumption of coffee, cigarettes. 喝咖啡和吸烟增多。 12. Increased consumption of alcohol. 饮酒增多。 13. Extremely happy mood, overeuphoric. 过分快乐和欣喜若狂。 14.Increased sex drive, interest in sex. 更喜欢参加与性有关的活动、性欲增加。 15. Over-activity (e.g., shopping, business, telephone calls, travelling, driving, visiting people). 活动过多 (如,购物、公干、打电话、旅游、驾车、访友等)
0.587
0.343
0.865
0.280
0.865
0.269
0.878
0.220
0.875
0.215
0.826
0.349
0.842
0.244
0.855
0.248
0.749
0.205
0.367
0.573
0.188
0.753
0.184
0.713
0.781
0.205
0.494
0.341
0.827
0.259
Factor loading N0.4 is shown in bold.
groups, with 60%–80% positive response in the BP-II group, 40%–50% positive response in the BP-I group and 5%–20% in the MDD group. 3.4. ROC curve analysis In the ROC curve analysis, the area under the curve was 0.69 [95% confidence interval (CI): 0.63–0.75] for MDD vs. BP-I and 0.90 (95% CI: 0.86–0.93) for MDD vs. BP-II (Fig. 2). The HCL-15 screening score of 7 was the optimal cutoff, with the sensitivity of 0.46 for BP-I and 0.79 for BP-II. The specificity reached 0.93, indicating that 93% of MDD patient who scored less than 7 in HCL-15 test truly did not have BP. 4. Discussion In our study, the mean age of MDD patients was significantly higher than that of BP patients, while the educational level of MDD patients was significantly lower than that of BP patients, and the percentage of female patients was comparatively higher in the MDD group. The result was consistent with the concept that BP patients tend to be younger. Better educational level in BP patients was consistent with our previous finding that cognitive impairment in patients with major depression during clinical remission stage was more severe than that in BP patients [8]. In previous factor analytic studies of hypomania, assessed with HCL-32, a dual-factor structure was consistently found, which characterized with “active/elated” and “risk-taking/irritable” items [9–13]. In our study, factor analysis of HCL-15 also revealed two factors: an “energetic/euphoric” factor and a mixed factor of irritability and substance use (cigarette, coffee and alcohol). The two factors virtually included all the items in the list, explaining more than 60% of the total variance. The internal consistent reliability of HCL-15 was as high as 0.93. It suggested that this simplified selfassessment tool has a good validity, although all the samples including MDD, BP-I and BP-II were combined in the analyses. The frequencies of positive responses of 15 items were generally high in BP-II patients, but items 11 (increased consumption of coffee, cigarettes) and 12 (increased consumption of alcohol) received comparatively low positive response in all participants. This was consistent with the demographic features of the study sample, which showed that 72.6% (461/623) of participants did not drink alcohol and 77.5% (483/623) of participants did not smoke cigarettes. Among the patients who did drink alcohol, 28.6% (18/63) BP-II patients recalled more consumption of alcohol, which was significantly higher than in BP-I patients (13.3%, 4/30) and MDD patients (13%, 10/77). Given the relatively low percentage of BP patients who used coffee, cigarettes and alcohol possibly due to Chinese culture and customs, we suggest to discard the two relevant items from the HCL-15. In light of the feature of affective instability overpresenting in BP versus MDD [17] — the core characteristic of cyclothymic temperament and whose
350
H. He et al. / General Hospital Psychiatry 36 (2014) 347–351
Fig. 2. ROC curve analyses. (A) HCL-15 exhibited good sensitivity and specificity in diagnosing BP-II (area under the curve=0.90) from MDD. (B) HCL-15 exhibited comparatively fair sensitivity and specificity (area under the curve=0.69) in diagnosing BP-I from MDD.
dysregulation may be the substrate for BP [18] — we suggest adding a domain with two additional items reflecting mood lability in the HCL15 for the future study. In this study, the cutoff score of 7 yielded a sensitivity of 0.79 between BP-II and MDD and a sensitivity of 0.46 between BP-I and MDD. The specificity reached 0.93. A substantial body of studies showed that, in different nations, including mainland China and Taiwan, HCL-32 displayed good sensitivity and specificity between BP and MDD, ranging from 0.68 to 0.85 and 0.46 to 0.91, respectively [13,14,19,20]. HCL-16, a shortened version of HCL-32, was reported to have a sensitivity of 0.83 and specificity of 0.71 between BP and MDD [21]. Since differentiation of BP-I and MDD is not a major concern in clinical practice, these results indicated that HCL-15 could be a very practical aid in differentiating BP-II and MDD. After 2 min of self-assessment, the clinician should be alerted by HCL scores of 7 or above that the currently depressed patient might need further evaluation in order to differentiate BP, especially BP-II, from MDD. If scores are less than 7, by 93% chance, this patient will probably be an MDD patient. Although HCL-32 has been widely tested in different nations with generally good evaluations, some clinicians complained that part of the scale was redundant and some items were confusing and hard to recall correctly. It was suggested by previous studies that a fundamental feature of hypomania was energized activity instead of the hypomanic mood [22]. Some studies have shown by factor analysis that, among multiple questions regarding hypomanic mood, only irritability was incorporated in the factor structure of hypomania, and it was suggested that, for the detection of BP-II, the focus of clinical questions on past hypomania should be on hypomanic behavior rather than hypomanic mood [11,23]. HCL-15 clearly focused more on hypomanic behavior, with only one question regarding the euphoria mood. It might be the reason that, with much less questions, HCL-15 exhibited even slightly better sensitivity and specificity in detecting BP-II from MDD, and all the items in the HCL15 were actually incorporated into the hypomanic factor structure. 4.1. Limitations Several limitations of this study deserve comments. Firstly, a limitation is that the main study tool, HCL-15, is a self-assessment instrument and requires patients to recall the past hypomanic experience. Secondly, all patients were in depressive episodes with or without psychotic symptoms while being assessed. The severity of symptoms, insight and some other clinical index may affect the results of recall, and we did not include these factors into our analyses. Thirdly, currently available HCL-15 study results were very limited, and the test–retest reliability was not known yet. These limitations
are counterbalanced by the fact that this paper is on an adult cohort of patients and does not have the limitation of having studied hypomania in adolescents where hypomania may be transient [24]. 4.2. Conclusion Despite the limitations mentioned above, this is the first study to test the validity of HCL-15. It yielded good sensitivity and specificity between BP, especially BP-II, and MDD. For some busy clinics, it can potentially be a practical choice for clinicians to use as a screening tool for each first-contact patient with current depressive episode. More studies from different areas are needed to further verify the effectiveness of HCL-15. Acknowledgments This study was supported by grants from (a) the Tackling Key Problems of Guangzhou Scientific and Technological Project (2007Z3E0611), Guangzhou Scientific and Technological Bureau of China; (2) Guangdong Scientific and Technological Project; (c) the Project of National Clinical Key Construction Specialty of China. References [1] American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. 4th ed., Text Revision (DSM-IV-TR). Washington, DC: American Psychiatric Association; 2000. p. 365–8. [2] Simpson SG, McMahon FJ, McInnis MG, et al. Diagnostic reliability of bipolar II disorder. Arch Gen Psychiatry 2002;59:736–40. [3] ten Have M, Vollebergh W, Bijl R, Nolen WA. Bipolar disorder in the general population in the Netherlands (prevalence, consequences and care utilisation): results from the Netherlands Mental Health Survey and Incidence Study (NEMESIS). J Affect Disord 2002;68:203–13. [4] Benazzi F, Akiskal HS. Refining the evaluation of bipolar II: beyond the strict SCIDCV guidelines for hypomania. J Affect Disord 2003;73:33–8. [5] Smith DJ, Harrison N, Muir W, Blackwood DH. The high prevalence of bipolar spectrum disorders in young adults with recurrent depression: toward an innovative diagnostic framework. J Affect Disord 2005;84:167–78. [6] Yang HC, Xiang YT, Liu TB, et al. Hypomanic symptoms assessed by the HCL-32 in patients with major depressive disorder: a multicenter trial across China. J Affect Disord 2012;143:203–7. [7] Yang HC, Yuan CM, Liu TB, et al. Validity of the Chinese version Mood Disorder Questionnaire (MDQ) and the optimal cutoff screening bipolar disorders. Psychiatry Res 2011;189:446–50. [8] Xu G, Lin K, Rao D, et al. Neuropsychological performance in bipolar I, bipolar II and unipolar depression patients: a longitudinal, naturalistic study. J Affect Disord 2012;136:328–39. [9] Meyer TD, Hammelstein P, Nilsson LG, Skeppar P, Adolfsson R, Angst J. The Hypomania Checklist (HCL-32): its factorial structure and association to indices of impairment in German and Swedish nonclinical samples. Compr Psychiatry 2007;48:79–87. [10] Wu YS, Angst J, Ou CS, Chen HC, Lu RB. Validation of the Chinese version of the hypomania checklist (HCL-32) as an instrument for detecting hypo(mania) in patients with mood disorders. J Affect Disord 2008;106:133–43.
H. He et al. / General Hospital Psychiatry 36 (2014) 347–351 [11] Benazzi F, Akiskal HS. The dual factor structure of self-rated MDQ hypomania: energized-activity versus irritable-thought racing. J Affect Disord 2003;73:59–64. [12] Hantouche EG, Angst J, Akiskal HS. Factor structure of hypomania: interrelationships with cyclothymia and the soft bipolar spectrum. J Affect Disord 2003;73: 39–47. [13] Yang HC, Yuan CM, Liu TB, et al. Validity of the 32-item Hypomania Checklist (HCL-32) in a clinical sample with mood disorders in China. BMC Psychiatry 2011;11:84. [14] Forty L, Smith D, Jones L, et al. Identifying hypomanic features in major depressive disorder using the hypomania checklist (HCL-32). J Affect Disord 2009;114:68–73. [15] First, MB, Spitzer, RL, Gibbon, M, Williams, JBW. Structured clinical interview for DSM-IV-TR axis I disorders—patient edition (SCID-I/P, 11/2002 revision). Biometrics Research Department, New York State Psychiatric Institute, New York, 2002. [16] 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;88:217–33. [17] Singh V, Bowden CL, Gonzalez JM, et al. Discriminating primary clinical states in bipolar disorder with a comprehensive symptom scale. Acta Psychiatr Scand 2013;127:145–52.
351
[18] Perugi G, Akiskal HS. The soft bipolar spectrum redefined: focus on the cyclothymic, anxious–sensitive, impulse–dyscontrol, and binge-eating connection in bipolar II and related conditions. Psychiatr Clin North Am 2002;25:713–37. [19] Chou CC, Lee IH, Yeh TL, et al. Comparison of the validity of the Chinese versions of the Hypomania Symptom Checklist-32 (HCL-32) and Mood Disorder Questionnaire (MDQ) for the detection of bipolar disorder in medicated patients with major depressive disorder. Int J Psychiatry Clin Pract 2012;16:132–7. [20] Haghighi M, Bajoghli H, Angst J, Holsboer-Trachsler E, Brand S. The Farsi version of the Hypomania Check-List 32 (HCL-32): applicability and indication of a fourfactorial solution. BMC Psychiatry 2011;11:14. [21] Forty L, Kelly M, Jones L, et al. Reducing the Hypomania Checklist (HCL-32) to a 16item version. J Affect Disord 2010;124:351–6. [22] Perugi G, Akiskal HS, Micheli C, et al. Clinical subtypes of bipolar mixed states: validating a broader European definition in 143 cases. J Affect Disord 1997;43:169–80. [23] Benazzi F, Akiskal HS, Micheli C, et al. Sensitivity and specificity of clinical markers for the diagnosis of bipolar II disorder. Compr Psychiatry 2001;42:461–5. [24] Paaren A, von Knorring A-L, Olsson G, et al. Hypomania spectrum disorders from adolescence to adulthood: a 15-year follow-up of a community sample. J Affect Disord 2013;145:190–9.
