- Email: [email protected]
Prevalence, risk factors and multi-group latent class analysis of lifetime anxiety disorders comorbid depressive symptoms
Journal of Affective Disorders 243 (2019) 360–365
Contents lists available at ScienceDirect
Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad
Research paper
Prevalence, risk factors and multi-group latent class analysis of lifetime anxiety disorders comorbid depressive symptoms ⁎
T
⁎⁎
Chen Hongguanga,1, Wang Xiaoa,1, Huang Yueqina, , Li Guohuab, , Liu Zhaoruia, Li Yanxiangb, Geng Hongchunb a Peking University Sixth Hospital (Institute of Mental Health), National Clinical Research Center for Mental Disorders & Key Laboratory of Mental Health, Ministry of Health (Peking University), Beijing 100191, China b Chifeng Anding Hospital, Chifeng, Inner Mongolia, 024000, China
A R T I C LE I N FO
A B S T R A C T
Keywords: Comorbidity Depression Anxiety disorder Epidemiology Latent class analysis
Background: Previous studies about comorbidity have primarily focused on disorders based on diagnostic criteria instead of symptoms. This study aimed to describe the prevalence and risk factors of anxiety comorbid depression based on a population-based sample in Chifeng City Inner Mongolia and explored the gender differences of depressive subtypes in anxiety patients. Methods: This study was a cross-sectional study conducted among 6376 community residents. Logistics analysis and multiple-group latent class analysis was used in exploring the risk factors and subtypes of anxiety comorbid depressive symptoms. Results: A total of 4528 respondents were interviewed in this study. The lifetime prevalence estimates for anxiety in the total sample was 5.70%. Among residents who had ever had anxiety, most of them reported having depressive symptoms while 15.79% of them met the criteria of MDD. Logistics analysis showed childhood adversities were associated with anxiety comorbid depressive symptoms. The results of multiple-group latent class analysis showed that the latent class probabilities were different between males and females. Conclusion: The prevalence rates of comorbidity were similar to the reports of previous regional surveys in China with statistically significant differences of comorbidity occurring between males and females. Precision prevention should therefore be targeted towards different kinds of populations.
1. Introduction
and Anxiety (NESDA) in 2004–2007 showed that of persons with a current anxiety disorder, 63% had a current and 81% had a lifetime depressive disorder (Lamers et al., 2011). In China, there were also some regional epidemiological studies conducted in Beijing (Rui et al., 2013), Shanghai (Shen et al., 2006), and Kunming (Jian et al., 2010) which described the prevalence of comorbidity. Merely using categorical diagnostic constructs can result in loss of valuable information about comorbidity as those who score just below the diagnostic threshold are regarded as non-cases. Currently, little insight is available on the co-occurrence of subthreshold depression and comorbid anxiety. Previous study showed that comorbidity rates increased considerably for lower thresholds of MDD (van Loo et al., 2016) and subthreshold depressive disorder was one of the best established risk factors for the onset of full-syndrome depressive disorders and frequently ran a
Mental disorders are widely recognized as a major contributor (14%) to the global burden of disease (Prince et al., 2007). Depression and anxiety disorders are the most prevalent mental disorders worldwide, jointly making up 50% of the international disease burden attributable to psychiatric and substance use disorders (Whiteford et al., 2013). Compared with patients having only anxiety, patients with comorbid major depressive disorder (MDD) and anxiety display greater psychiatric symptom severity, poorer psychosocial function and elevations in general distress and suicide risk (Cyranowski et al., 2012). While a large body of research on the epidemiology of comorbid depression and anxiety exists, most studies have focused on diagnostic criteria based disorders. Results of the Netherlands Study of Depression
⁎ Corresponding author at: Department of Social Psychiatry and Behavioral Medicine, Institute of Mental Health, Peking University, No. 51 Hua Yuan Bei Road, Beijing 100191, China. ⁎⁎ Corresponding author at: Chifeng Anding Hospital, No. 18 Gonggeer Road, Chifeng, Inner Mongolia, 024000, China. E-mail addresses: [email protected] (H. Chen), [email protected] (Y. Huang), [email protected] (G. Li). 1 The first two authors contributed equally to this work.
https://doi.org/10.1016/j.jad.2018.09.053 Received 22 May 2018; Received in revised form 6 August 2018; Accepted 15 September 2018 Available online 18 September 2018 0165-0327/ © 2018 Elsevier B.V. All rights reserved.
Journal of Affective Disorders 243 (2019) 360–365
H. Chen et al.
including sex, age, employment status, education, income, and marital status was also collected from CIDI. Age was divided into the following four categories: 18–34, 35–49, 50–64 and 65 years old and over. Educational level was grouped into two levels based on years of education. The respondents who received less than or equal to nine years were categorized as having a lower education level, while respondents who completed more than nine years of school were categorized as having a higher education level. Marital status included married, cohabitating, divorced, separated, widowed and never married. Urban and rural areas were classified according to the permanent addresses of the respondents. Depressive symptoms-nine kinds of symptoms could be acquired by CIDI-3.0. Feelings of depression: depressed mood most of the day, nearly every day, as indicated by either subjective report (e.g., feels sad or empty) or observation made by others. Loss of interest: markedly diminished interest or pleasure in all, or almost all, activities most of the day, nearly every day (as indicated by either subjective account or observation made by others). Weight decrease/increase: significant weight loss when not dieting or weight gain (e.g., a change of more than 5% of body weight in a month), or decrease or increase in appetite nearly every day. Insomnia/hypersomnia: insomnia or hypersomnia nearly every day. Psychomotor agitation/retardation: psychomotor agitation or retardation nearly every day (observable by others, not merely subjective feelings of restlessness or being slowed down). Tiredness/lack of energy: fatigue or loss of energy nearly every day. Feelings of guilt or worthlessness: feelings of worthlessness or excessive or inappropriate guilt (which may be delusional) nearly every day (not merely self-reproach or guilt about being sick). Feelings of guilt or worthlessness: diminished ability to think or concentrate, or indecisiveness, nearly every day (either by subjective account or as observed by others). Trouble concentrating: recurrent thoughts of death (not just fear of dying), recurrent suicidal ideation without a specific plan, or a suicide attempt or a specific plan for committing suicide.
chronic course or tends to develop in to an episode of MDD (Klein et al., 2009). Based on the Diagnostic and Statistic Manual of Mental Disorders, 4th edition (DSM-IV) criteria, there are nine kinds of depressive symptoms. Some studies showed that there are specific symptoms of depression in people with different sociodemographic characters. For example, gender differences have been shown as important factors in the study of depression and depressive symptoms in several cases and gender differences in expression of depressive symptoms have also been reported (Marcus et al., 2011). Compared to males, females have a higher prevalence of chronic minor depression, dysthymia and somatic depression (Kessler et al., 1994; Angst and Merikangas, 1997; Silverstein, 1999), and are manifested as more recurrent brief depression episodes. Females reported experiencing significantly more fatigue, hypersomnia, and psychomotor retardation during the most severe major depressive episode, whereas males reported more insomnia and agitation (Khan et al., 2002). Currently there are very few studies examining the comorbidity of mood and anxiety disorders on symptoms level in China. This study was carried out in Chifeng City of Inner Mongolia, and was the first population-based sample epidemiological study on the prevalence of lifetime depression disorders in residents with anxiety disorders in that region. Chifeng City is one of the major cities in Inner Mongolia which is inhabited by a multi-ethnic population with a relatively low socio-economic status. It is a typical region in a period of economic transition with a large number of residents but relatively weak economy and overall low level of school education. More importantly, this study was also designed to explore the gender differences of depressive subtypes in anxiety patients and to provide theoretical evidence for early diagnosis and prevention strategy of mental disorders. 2. Materials and methods 2.1. Sample
2.3. Procedures A three-stage selection scheme was used to select the sample for every eligible individual (adults aged 18 years and over, and living in a family household) in the target population in 2010. “Neighborhood committee” (NC, in urban area) and “Village committee” (VC, in rural area) are local community organizations in China. Since all households belong to such committees, the committees were used as the primary sampling units (PSUs). In the first stage of sampling, 108 PSUs were selected based on the strategy of probabilities proportional to size (PPS). Stratification was made on urban and rural areas, with 57 PSUs from urban areas and 51 PSUs from rural areas. The second stage sampling selected a probability sample of households in each PSU. Based on the difference of estimated response rates, 80 households were selected in each VC, and 50 households were selected in each NC. The third stage sampling selected one random respondent in each sample household. Finally, 6376 respondents were selected in the survey.
62 lay interviewers were trained in a seven day training session at Peking university sixth hospital during November 2010. The detailed training methods can be found elsewhere (Wang et al., 2018). Only those trainees who passed the final test could be selected as interviewers. Face-to-face interviews were conducted by selected interviewers from November 2011 to April 2013. The average interview time was one hour. The study was approved by the ethics review board at Peking University Institute of Mental Health (PKUIMH). Before respondents were interviewed, written informed consent was obtained and declarations of anonymity and confidentiality were signed. 2.4. Statistical analyses Design weight was computed from the different probabilities of selection, and post-stratification weight was generated based on the age and gender distribution of census population. The consolidated weights were applied to the sample data. The weighted prevalence was calculated and the differences of depressive symptoms between male and female were analyzed by chi-square test. Additionally, logistics analysis was applied to explore the related factors of anxiety disorders comorbid 5 and over depressive symptoms. The dependent variable was anxiety disorders comorbid 5 and over depressive symptoms. The independent variables were social-demographic factors (sex, age, marital status, educational attainment, and employment) and childhood adversities. Latent Class Analysis (LCA) is a statistical technique that is used in factor, cluster, and regression techniques; it is a subset of structural equation modeling (SEM). LCA is a technique where constructs are identified and created from unobserved or latent subgroups, which are usually based on individual responses from multivariate categorical data. In this study, LCA on the items of depressive symptoms among the
2.2. Instruments and measures Composite International Diagnostic Interview (CIDI) was applied in the survey. CIDI is a fully structured lay-administered diagnostic interview (Kessler and Ustun, 2004) which has been widely used in more than 30 countries (Kessler and Ustun, 2004). Clinical reappraisal found generally good concordance with the CIDI diagnoses of mood disorder and anxiety disorders (Qin et al., 2010). Over the last twenty years, CIDI has been used as the primary instrument in regional epidemiological mental health surveys in many cities in China; such as Beijing and Shanghai (Shen et al., 2006), Jiangxi (Chen et al., 2004), Liaoning (Pan et al., 2006), Guangzhou (Zhao et al., 2009) and Kunming (Jian et al., 2010). In this study, anxiety disorders and depression disorders were diagnosed using CIDI-3.0, according to the criteria and definition of the DSM-IV. Additionally, social-demographic information 361
Journal of Affective Disorders 243 (2019) 360–365
H. Chen et al.
3.4. Risk factors of anxiety comorbid more than 4 kinds of depressive symptoms
respondents with anxiety was performed with SAS9.4. LCA (Hutchinson and Tang, 1987; Yang and Becker, 1997) is a statistical technique for identifying mutually exclusive classes of respondents, each with its own set of symptom endorsement probabilities. In terms of LCA, a solution of classes is optimal when classes are as homogeneous as possible, while differences between classes are as large as possible. In LCA, classes are added until the model fits the data well. Fifty random sets of starting values were used. Models ranging from 2 to 5 classes were evaluated. Model fit was evaluated with the following commonly-used statistics: Akaike information criterion (AIC), the Bayesian Information Criterion (BIC), sample size adjusted BIC (Vrieze, 2012), and Entropy values. Smaller values of AIC and BIC, and larger values of Entropy, indicate better model fit (Nylund et al., 2007). Gender was included in the multiple-group latent class analysis.
Table 4 also presents the results of logistics analysis. Considering social demographic factors and childhood adversities, childhood adversities were statistically significant risk factors (OR = 2.7). 3.5. Multiple-group latent class analysis LCA revealed that four classes provided the best fit statistics. Fit statistics for the final 4-class model were: AIC = 329.63; BIC = 583.31; Adjusted BIC = 336.23; Entropy = 0.84. Table 5 depicts the observed class membership and endorsement frequencies of the 9 disaggregated depressive symptoms in the best-fitting LCA solution. Class 1 (24.7% of respondents with anxiety) was characterized by very high endorsement rates for most of the 9 disaggregated symptoms of major depression. Class 1 was termed as “severe typical” because of the nearly universal presence of major depression along with classical depressive symptoms (i.e., weight and appetite loss, insomnia, psychomotor retardation, anergia, and poor concentration). Class 2 (31.3% of respondents with anxiety) was characterized by prominent symptoms of low mood and loss of interest in females but nearly no symptoms in males. Thus class 2 was termed as “not depressed”. Like class 1, class 3 (25.5% of respondents with anxiety) was characterized by many depressive symptoms, however, the symptom pattern was more characterized by weight decrease/increase and insomnia/hypersomnia but less characterized by feelings of guilt or worthlessness and thoughts of death/suicide. Class 3 was termed “moderate typical”. Class 4 (18.5% of respondents with anxiety) was characterized by atypical depressive symptoms, being termed as “mild atypical.” From the results of multiple-group latent class analysis, it can be seen that the latent class probabilities were different between males and females. In male respondents with anxiety, the proportions of these four classes were 36.9%, 24.7%, 17.2%, 21.2% respectively while in female respondents with anxiety, the proportions of these four classes were 29.1%, 18.6%, 23.6%,28.7% respectively. The female group in class 2 was characterized by a very high endorsement rate of prominent symptoms while the male group was not. Besides, in class 4, the female group was characterized by thoughts of death while the male group was not. (Table 5)
3. Results 3.1. Sample description A total of 4528 respondents were interviewed in this study. The response rate was 71.2%. More participants are women (53.9%), married or cohabiting (93.1%), of low educational level (87.6%), lived in rural area (67.2%), and reported no steady job (71.1%). Table 1 shows more information of socio-demographic characteristics of the Chifeng sample by gender. 3.2. Prevalence of anxiety disorders and comorbid depression The prevalence estimates for anxiety in the total sample were 5.70% for lifetime and 3.96% for the 12 months before the interview. Among residents who had ever had anxiety, 17.59% of them had depression while 15.79% of them met the criteria of MDD. Prevalence rates of comorbidity between people with different sociodemographic characters were shown in Table 2. There were significant differences between males and females. 3.3. Comorbid depressive symptom Among respondents who met the diagnosis for anxiety disorders, most of them had more than one kind of depressive symptom. There was no significant difference between males and females. 64.31% of those respondents had more than 4 kinds of symptoms but did not meet the criteria of major depression disorder since the duration of symptoms was less than 2 weeks. (Table 3)
4. Discussion 4.1. Discussion Using a representative data of Chifeng City in Inner Mongolia Autonomous Region of China, results confirmed high lifetime comorbid depression disorders rates in anxiety disorders. Of persons with anxiety disorder, 17.59% had a lifetime comorbid depressive disorder in which rates of comorbid anxiety disorder were higher for those with MDD than for those with MND. The results were similar to the reports of some previous regional surveys in China, but were lower than some CIDI surveys in other counties (King-Kallimanis et al., 2009; Johansson et al., 2013; Schuch et al., 2014). Besides, there are statistically significant differences of comorbidity between males and females: as is consistent with previous studies, women reported more comorbid depression when compared to men. In one longitudinal epidemiological study, the prevalence of MDD and of one or two anxiety disorders was about two-fold higher in women than in men (Breslau et al., 1995). Besides, there were 62.08% of the respondents with anxiety who also had more than 4 kinds of symptoms but did not meet the criteria of depression, as the duration of symptoms was less than 2 weeks. These results showed that females with more than 4 kinds of depressive symptoms were more likely to meet the criteria of MDD than male due to the low level of emotion-regulation ability. Besides, although they did not meet criteria for depression, these symptoms still made them
Table 1 Socio-demographic characteristics of the Chifeng sample by gender. Characteristic
Age
Area Marital status
Education Income
Male
18–34 35–49 50–64 65Rural Urban Married/cohabitating Divorced/separated/widowed/ never married Lowera Higherb No Yes
Female
N
%
N
%
303 763 759 260 1445 640 1946 139
30.12 34.94 24.98 9.96 68.54 31.46 86.15 13.85
432 961 771 279 1596 847 2270 173
33.43 36.58 21.39 8.60 66.24 33.76 89.35 10.65
1675 410 479 1606
77.23 22.77 23.93 76.08
2058 385 1048 1395
80.09 19.91 43.13 56.87
a The respondents who received less than or equal to nine years were categorized as having lower education level. b The respondents who completed more than nine years of school were categorized as having a higher education level.
362
Journal of Affective Disorders 243 (2019) 360–365
H. Chen et al.
Table 2 Prevalence of depression disorders in anxiety disorder patients. Characteristic
Gender Age
Area Education Income
a
MNDa
MDD
Male Female 18–34 35–49 50–64 65Rural Urban Lower Higher No Yes
%
Se
P
χ
8.96 19.77 12.92 12.19 18.37 27.90 13.82 23.80 15.70 16.20 18.43 13.90
2.79 3.29 5.43 3.34 3.80 8.98 2.70 5.97 2.70 6.46 3.87 2.84
0.010
6.569
0.234
4.273
0.089
2.887
0.944
0.005
0.293
1.106
2
Depression
%
se
P
χ
0.83 2.98 – 2.80 2.28 5.50 2.00 2.95 2.35 1.44 1.52 2.67
0.84 1.19 – 1.63 1.61 3.76 0.85 2.07 0.92 1.45 1.05 1.20
0.215
1.535
–
–
0.632
0.229
0.644
0.213
0.498
0.459
2
%
Se
P
χ2
9.79 22.13 12.92 13.84 20.66 33.40 15.33 26.75 17.58 17.64 19.94 15.89
2.90 3.34 5.43 3.70 3.96 9.12 2.60 6.68 2.60 7.10 3.84 2.88
0.006
7.478
0.098
6.297
0.071
3.269
0.993
<0.001
0.352
0.868
Minor depression disorder.
programs should target not only the people who met the diagnostic criteria but also the people with subthreshold mental disorder. This paper reported gender differences in anxiety comorbid symptomatic profiles of depression. The pattern of comorbidity is similar with previous study (Sheila M Marcus, 2011). It is the same in class 1 and class 3, indicating that typical depressive symptoms of females and males showed no significant differences. GordonParker's study indicated that female unipolar groups scored only marginally (and nonsignificantly) higher than male groups on the depression severity measures. Gender had minimal if any impact on unipolar depression (Parker et al., 2014). While in class 2, females were more likely to have feelings of depression and loss of interest, and to have more kinds of symptoms than males. One explanation is that compared to males, females with anxiety disorders are more sensitive (Park et al., 2015). Gonadal hormones may be a major factor in this disparity, given that women are more likely to experience mood disturbances during times of hormonal flux, and testosterone may have protective benefits against anxiety and depression (McHenry et al., 2014). Besides, when being stressed, females seek support and express their feelings more than males (Rose and Rudolph, 2006). Therefore, they may receive more relationship provisions, which may later contribute to a positive emotional adjustment. This might be the reason why females had major symptoms that were categorized in “not depressed” class. In class 4, the female group was characterized by thoughts of death but had less symptoms than males. This result was similar with previous study. Ruth's study (Derdikman-Eiron et al., 2012) showed that gender was a moderator variable in the associations between symptoms of anxiety and depression and impairment, meaning that boys' functioning was impaired to a larger extent than girls' functioning. These results suggest that precision prevention should be targeted based on the specific kind of population. Any given prevention strategy may not be equally effective in all individuals because of biological differences in risk and response to the preventive modality. Unlike precision medicine, which has traditionally focused on molecular data to target specific treatments to an individual, precision prevention requires a broader conceptual framework. Consequently, the integration of a broad range of research domains is required, which includes cancer biology, epidemiology, biomedical informatics, bioinformatics, biostatistics, risk estimation, genetics, biomarkers, disparities, health services research, clinical trials, geospatial analysis, and dissemination and implementation research. These domains should be integrated under a community engagement framework to achieve the future precision prevention goals.
Table 3 Anxiety disorders comorbid depressive symptoms by gender. Total
Feelings of depression Loss of interest Weight decrease/increase insomnia/hypersomnia psychomotor agitation/ retardation Tiredness/lack of energy Feelings of guilt or worthlessness Trouble concentrating Thoughts of death/suicide More than 4 kinds of symptoms Depression
Male
P
Female
N
%
N
%
N
%
131 120 105 118 66
68.59 62.83 74.47 83.69 46.81
38 32 31 34 22
62.62 57.93 81.01 84.60 56.77
93 88 74 84 44
71.14 66.83 71.13 84.89 44.34
0.336 0.331 0.296 0.970 0.251
111 64 120 69 116 53
78.72 45.39 85.11 48.94 38.28 22.56
30 20 35 18 32 11
75.47 52.40 86.30 49.14 37.03 9.79
81 44 85 51 84 42
82.23 40.33 87.43 46.18 49.73 22.12
0.426 0.313 0.857 0.775 0.067 0.006
Table 4 Logistic models of anxiety disorders comorbid more than 4 kinds of depressive symptoms. Effect Gender Area Age
Marital status
Employment Income Education Childhood adversities
OR Male Female Rural Urban 18–34 35–49 50–64 65Married/cohabitating Divorced/separated/ widowed/ never married No Yes No Yes Lowera Higherb No Yes
1 1.48 1 0.73 1 2.24 1.27 1.93 1 1.23 1 1.16 1 0.65 1 2.12 1 2.7
(95%CI)
(0.85,2.58) (0.37,1.47) (0.87,5.75) (0.49,3.29) (0.65,5.71) (0.46,3.33)
(0.58,2.33) (0.38,1.11) (0.86,5.24) (1.10,6.60)
a The respondents who received less than or equal to nine years were categorized as having a lower education level. b The respondents who completed more than nine years of school were categorized as having a higher education level.
feel stressed most of the day at that time and influenced their daily life. In light of these findings, not only case-level depression but also symptom-level depressive state should be taken into consideration in future studies on comorbidity. Prevention and early intervention
4.2. Limitations Limitations of this study include the cross-sectional character of this 363
Journal of Affective Disorders 243 (2019) 360–365
H. Chen et al.
Table 5 Multiple-group latent class analysis of depressive symptoms among respondents with anxiety by gender. Male
Feelings of depression Loss of interest Weight decrease/increase Insomnia/hypersomnia Psychomotor agitation/retardation Tiredness/lack of energy Feelings of guilt or worthlessness Trouble concentrating Thoughts of death/suicide Percent of sample by gender Average number of symptoms
Female
Class1 Severe typical
Class2 Not depressed
Class3 Moderate typical
Class4 Mild atypical
Class1 Severe typical
Class2 Not depressed
Class3 Moderate typical
Class4 Mild atypical
0.928 0.994 0.948 0.853 0.726 0.851 0.710 0.998 0.826 0.369 7.92
0.233 0.086 0.044 0.052 0.028 0.519 0.267 0.524 0.036 0.247 0.40
0.992 0.990 0.686 0.996 0.015 0.380 0.192 0.585 0.013 0.172 4.69
0.568 0.116 0.725 0.994 0.979 0.991 0.384 0.995 0.165 0.212 6.33
0.983 0.997 0.824 0.923 0.522 0.939 0.713 0.998 0.969 0.291 7.97
0.895 0.690 0.315 0.475 0.053 0.385 0.133 0.674 0.184 0.186 4.02
0.852 0.959 0.993 0.998 0.567 0.954 0.360 0.783 0.131 0.236 6.58
0.105 0.008 0.525 0.991 0.968 0.988 0.287 0.990 0.967 0.287 0.80
References
study, which does not allow for making causal inferences and the data was based on retrospective self-reports, which is subject to recall bias. So it is important to be cautious in interpreting the results because recall bias almost certainly exists in retrospective study. Furthermore, mental health was possibly underestimated because of the respondents’ stigma of psychological questions.
Angst, J., Merikangas, K., 1997. The depressive spectrum: diagnostic classification and course. J. Affect. Disord. 45 (1-2), 31–39 discussion 39-40. Breslau, N., Schultz, L., et al., 1995. Sex differences in depression: a role for preexisting anxiety. Psychiatry Res. 58 (1), 1–12. Chen, H.L., Bin, H.U., et al., 2004. Epidemiological investigation on mental disorders in Jiangxi Province. Chin. J. Psychiatry 37 (3), 172–175. Cyranowski, J.M., Schott, L.L., et al., 2012. Psychosocial features associated with lifetime comorbidity of major depression and anxiety disorders among a community sample of mid-life women: the SWAN mental health study. Depress Anxiety 29 (12), 1050–1057. Derdikman-Eiron, R., Indredavik, M.S., et al., 2012. Gender differences in psychosocial functioning of adolescents with symptoms of anxiety and depression: longitudinal findings from the Nord-Trøndelag Health Study. Social Psychiatry Psychiatric Epidemiol. 47 (11), 1855–1863. Hutchinson, T.P., Tang, K.Y., 1987. "The value of latent class analysis in medical diagnosis. Stat. Med. 6 (4), 529–531. Jian, Y., Qin, H.Y., et al., 2010. Epidemiological survey of affective disorder in Kunming City. Chin. Mental Health J. 24 (2), 110–115. Johansson, R., Carlbring, P., et al., 2013. Depression, anxiety and their comorbidity in the Swedish general population: point prevalence and the effect on health-related quality of life. Peer J. 1, e98. Kessler, R.C., McGonagle, K.A., et al., 1994. Sex and depression in the national comorbidity survey. II: cohort effects. J. Affect. Disord. 30 (1), 15–26. Kessler, R.C., Ustun, T.B., 2004. The world mental health (WMH) survey initiative version of the world health organization (WHO) composite international diagnostic interview (CIDI). Int. J. Methods Psychiatr. Res. 13 (2), 93–121. Khan, A.A., Gardner, C.O., et al., 2002. Gender differences in the symptoms of major depression in opposite-sex dizygotic twin pairs. Am. J. Psychiatry 159 (8), 1427–1429. King-Kallimanis, B., Gum, A.M., et al., 2009. Comorbidity of depressive and anxiety disorders for older Americans in the national comorbidity survey-replication. Am. J. Geriatr. Psychiatry 17 (9), 782–792. Klein, D.N., Shankman, S.A., et al., 2009. Subthreshold depressive disorder in adolescents: predictors of escalation to full-syndrome depressive disorders. J. Am. Acad. Child Adolesc. Psychiatry 48 (7), 703–710. Lamers, F., Oppen, P.V., et al., 2011. Comorbidity patterns of anxiety and depressive disorders in a large cohort study: the Netherlands study of depression and anxiety (NESDA). J. Clin. Psychiatry 72 (3), 341. Marcus, S.M., Kerber, K.B., et al., 2011. Gender differences in depression symptoms in treatment-seeking adults: STAR*D confirmatory analyses. Comprehensive Psychiatry 49 (3). McHenry, J., Carrier, N., et al., 2014. Sex differences in anxiety and depression: role of testosterone. Front. Neuroendocrinol. 35 (1), 42–57. Nylund, K.L., et al., 2007. Deciding on the number of classes in latent class analysis and growth mixture modeling: a Monte Carlo simulation study. Struct. Eq. Model. Multidisciplin. J. 4 (14), 535–569. Pan, G., Chao, J., et al., 2006. Epidemiological survey of mental disorders in urban and rural areas of Liaoning province. Chin. J. Public Health 73 (2), 13–18. Park, S.C., Lee, M.S., et al., 2015. Gender differences in depressive symptom profiles and patterns of psychotropic drug usage in Asian patients with depression: findings from the research on Asian psychotropic prescription patterns for antidepressants study. Aust. N. Z. J. Psychiatry 49 (9), 833–841. Parker, G., Fletcher, K., et al., 2014. Gender differences in depression severity and symptoms across depressive sub-types. J. Affect. Disord. 167, 351–357. Prince, M., Patel, V., et al., 2007. No health without mental health. Lancet 370 (9590), 859–877. Qin, H.Y., Xie, S.F., et al., 2010. Community-based evaluation of the reliability and validity of Chinese version of composite international diagnostic interview-3.0. Chin. Mental Health J. 24 (1) 21-20. Rose, A.J., Rudolph, K.D., 2006. A review of sex differences in peer relationship processes: potential trade-offs for the emotional and behavioral development of girls and boys.
5. Conclusion This is the first regional representative cross-sectional survey of mental disorders in Chifeng City, the Inner Mongolia Autonomous Region of China. The prevalence rates of comorbidity were similar to the reports of previous regional surveys in China and there are statistically significant differences of comorbidity between males and females. Bearing in mind the importance of precision prevention, our findings may contribute to improved assessment and intervention methods tailored differently to males and females with symptoms of anxiety and depression. Competing of Interests The authors have declared that no competing interests exist. Disclaimers The views expressed in the submitted article are our own and not an official position of the institution or funder. Source of support This research was supported by the Bureau public health of Chifeng, China and The National Key Technology R&D Program of China (No. 2015BAI13B00). Acknowledgments This research was supported by the Bureau public health of Chifeng, China and The National Key Technology R&D Program of China (No. 2015BAI13B00). We thank all of the colleagues in the Chifeng Anding Hospital and Peking University Sixth Hospital for providing data and assisting the data analysis. We also thank Dr. William A Jefferson from Chinese Academy of Science for help with English grammar and wording in our manuscripts. Supplementary materials Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.jad.2018.09.053. 364
Journal of Affective Disorders 243 (2019) 360–365
H. Chen et al.
Commission of Health and Function 2016-present Vice-president, China Disabled Persons’ Federation 2013-present President, Chinese Mental Health Journal 2013-present Honorary Professor in the University of Hong Kong 2003-present Director, Division of Social Psychiatry and Behavior Medicine, Institute of Mental Health, Peking University 2002-present 6. President, Society of Crisis Intervention of Chinese Association for Mental Health
Psychol. Bull. 132 (1), 98–131. Rui, L.Z., Qin, H.Y., et al., 2013. The prevalence of mood disorder, anxiety disorder and substance use disorder in community residents in Beijing: a cross-sectional study. Chin. Mental Health J. 27 (2), 102–110. Schuch, J.J., Roest, A.M., et al., 2014. Gender differences in major depressive disorder: results from the Netherlands study of depression and anxiety. J. Affect. Disord. 156, 156–163. Sheila, MMarcus,K.B.K.A., 2011. Gender differences in depression symptoms in treatment-seeking adults: STAR*D confirmatory analyses. Comprehens. Psychiatry 49 (3). Shen, Y.C., Zhang, M.Y., et al., 2006. Twelve-month prevalence, severity, and unmet need for treatment of mental disorders in metropolitan China. Psychol. Med. 36 (2), 257–267. Silverstein, B., 1999. Gender difference in the prevalence of clinical depression: the role played by depression associated with somatic symptoms. Am. J. Psychiatry 156 (3), 480–482. van Loo, H.M., Schoevers, R.A., et al., 2016. Psychiatric comorbidity does not only depend on diagnostic thresholds: an illustration with major depressive disorder and generalized anxiety disorder. Depression Anxiety 33 (2), 143–152. Vrieze, S.I., 2012. Model selection and psychological theory: a discussion of the differences between the Akaike information criterion (AIC) and the Bayesian information criterion (BIC). Psychol. Methods 17 (2), 228–243. Wang, X., Liu, Z., et al., 2018. Association of comorbidity of mood and anxiety disorders with suicidal behaviors. J. Affect. Disord. 227, 810–816. Whiteford, H.A., Degenhardt, L., et al., 2013. Global burden of disease attributable to mental and substance use disorders: findings from the global burden of disease study 2010. Lancet 382 (9904), 1575–1586. Yang, I., Becker, M.P., 1997. Latent variable modeling of diagnostic accuracy. Biometrics 53 (3), 948–958. Zhao, Z., Huang, Y., et al., 2009. An epidemiological survey of mental disorders in Guangzhou area. Chin. J. Nervous Mental Dis.
2. 3. 4. 5.
Clinical Posts: Psychiatrist, The Sixth Hospital, Peking University 2002-present Psychiatrist, Institute of Mental Health, Beijing Medical University 1987–1996
Current and Recent (Relevant) Research Support 1. Chinese Ministry of Technology. Multilevel analysis on predictors of mental health service use among patients with affective and cognitive disorders. (2015–2017) Key personnel, ¥23,540,000 2. European Research Commission. 10/66 ten years on – monitoring and improving health expectancy by targeting frailty among older people in middle income countries. (2014–2019) Country PI, €137,900 3. Beijing Health Bureau. Community based intervention trail on families with dementia patients, (2012–2015) co-PI, ¥104,000 4. Chinese Ministry of Technology. National epidemiological survey of mental disorders. (2012–2015) PI. ¥4,760,000 5. Chinese Ministry of Health. Study on disease burden of mental disorders and mental health service. (2012–2015) PI.¥20,220,000 6. WHO WPRO. Study on suicide surveillance in Kunming City, Yunnan Province, China. (2011–2014). PI, $25,024 7. Beijing Scientific and Technology Committee. Optimization of treatment of rTMS on depression. (2010–2012). PI, ¥1,600,000 8. Suicide study and prevention: China Collaborative Suicide Research Training Program– (No.D43TW007273-05)(2006.1 −2010.12)
Yueqin Huang:
Education: Bachelor of Medicine (MD equivalent), Beijing Medical University, 1984 Master of Medicine (MPH equivalent), Beijing Medical University, 1987 Ph.D, Peking University, 2004
Publications The editor-in-Chief of six books, and 276 papers including 142 first-author and correspondent-author papers
Academic Posts: 1. Member, Executive Committee of Rehabilitation International and Chair of
365
Contents lists available at ScienceDirect
Journal of Affective Disorders journal homepage: www.elsevier.com/locate/jad
Research paper
Prevalence, risk factors and multi-group latent class analysis of lifetime anxiety disorders comorbid depressive symptoms ⁎
T
⁎⁎
Chen Hongguanga,1, Wang Xiaoa,1, Huang Yueqina, , Li Guohuab, , Liu Zhaoruia, Li Yanxiangb, Geng Hongchunb a Peking University Sixth Hospital (Institute of Mental Health), National Clinical Research Center for Mental Disorders & Key Laboratory of Mental Health, Ministry of Health (Peking University), Beijing 100191, China b Chifeng Anding Hospital, Chifeng, Inner Mongolia, 024000, China
A R T I C LE I N FO
A B S T R A C T
Keywords: Comorbidity Depression Anxiety disorder Epidemiology Latent class analysis
Background: Previous studies about comorbidity have primarily focused on disorders based on diagnostic criteria instead of symptoms. This study aimed to describe the prevalence and risk factors of anxiety comorbid depression based on a population-based sample in Chifeng City Inner Mongolia and explored the gender differences of depressive subtypes in anxiety patients. Methods: This study was a cross-sectional study conducted among 6376 community residents. Logistics analysis and multiple-group latent class analysis was used in exploring the risk factors and subtypes of anxiety comorbid depressive symptoms. Results: A total of 4528 respondents were interviewed in this study. The lifetime prevalence estimates for anxiety in the total sample was 5.70%. Among residents who had ever had anxiety, most of them reported having depressive symptoms while 15.79% of them met the criteria of MDD. Logistics analysis showed childhood adversities were associated with anxiety comorbid depressive symptoms. The results of multiple-group latent class analysis showed that the latent class probabilities were different between males and females. Conclusion: The prevalence rates of comorbidity were similar to the reports of previous regional surveys in China with statistically significant differences of comorbidity occurring between males and females. Precision prevention should therefore be targeted towards different kinds of populations.
1. Introduction
and Anxiety (NESDA) in 2004–2007 showed that of persons with a current anxiety disorder, 63% had a current and 81% had a lifetime depressive disorder (Lamers et al., 2011). In China, there were also some regional epidemiological studies conducted in Beijing (Rui et al., 2013), Shanghai (Shen et al., 2006), and Kunming (Jian et al., 2010) which described the prevalence of comorbidity. Merely using categorical diagnostic constructs can result in loss of valuable information about comorbidity as those who score just below the diagnostic threshold are regarded as non-cases. Currently, little insight is available on the co-occurrence of subthreshold depression and comorbid anxiety. Previous study showed that comorbidity rates increased considerably for lower thresholds of MDD (van Loo et al., 2016) and subthreshold depressive disorder was one of the best established risk factors for the onset of full-syndrome depressive disorders and frequently ran a
Mental disorders are widely recognized as a major contributor (14%) to the global burden of disease (Prince et al., 2007). Depression and anxiety disorders are the most prevalent mental disorders worldwide, jointly making up 50% of the international disease burden attributable to psychiatric and substance use disorders (Whiteford et al., 2013). Compared with patients having only anxiety, patients with comorbid major depressive disorder (MDD) and anxiety display greater psychiatric symptom severity, poorer psychosocial function and elevations in general distress and suicide risk (Cyranowski et al., 2012). While a large body of research on the epidemiology of comorbid depression and anxiety exists, most studies have focused on diagnostic criteria based disorders. Results of the Netherlands Study of Depression
⁎ Corresponding author at: Department of Social Psychiatry and Behavioral Medicine, Institute of Mental Health, Peking University, No. 51 Hua Yuan Bei Road, Beijing 100191, China. ⁎⁎ Corresponding author at: Chifeng Anding Hospital, No. 18 Gonggeer Road, Chifeng, Inner Mongolia, 024000, China. E-mail addresses: [email protected] (H. Chen), [email protected] (Y. Huang), [email protected] (G. Li). 1 The first two authors contributed equally to this work.
https://doi.org/10.1016/j.jad.2018.09.053 Received 22 May 2018; Received in revised form 6 August 2018; Accepted 15 September 2018 Available online 18 September 2018 0165-0327/ © 2018 Elsevier B.V. All rights reserved.
Journal of Affective Disorders 243 (2019) 360–365
H. Chen et al.
including sex, age, employment status, education, income, and marital status was also collected from CIDI. Age was divided into the following four categories: 18–34, 35–49, 50–64 and 65 years old and over. Educational level was grouped into two levels based on years of education. The respondents who received less than or equal to nine years were categorized as having a lower education level, while respondents who completed more than nine years of school were categorized as having a higher education level. Marital status included married, cohabitating, divorced, separated, widowed and never married. Urban and rural areas were classified according to the permanent addresses of the respondents. Depressive symptoms-nine kinds of symptoms could be acquired by CIDI-3.0. Feelings of depression: depressed mood most of the day, nearly every day, as indicated by either subjective report (e.g., feels sad or empty) or observation made by others. Loss of interest: markedly diminished interest or pleasure in all, or almost all, activities most of the day, nearly every day (as indicated by either subjective account or observation made by others). Weight decrease/increase: significant weight loss when not dieting or weight gain (e.g., a change of more than 5% of body weight in a month), or decrease or increase in appetite nearly every day. Insomnia/hypersomnia: insomnia or hypersomnia nearly every day. Psychomotor agitation/retardation: psychomotor agitation or retardation nearly every day (observable by others, not merely subjective feelings of restlessness or being slowed down). Tiredness/lack of energy: fatigue or loss of energy nearly every day. Feelings of guilt or worthlessness: feelings of worthlessness or excessive or inappropriate guilt (which may be delusional) nearly every day (not merely self-reproach or guilt about being sick). Feelings of guilt or worthlessness: diminished ability to think or concentrate, or indecisiveness, nearly every day (either by subjective account or as observed by others). Trouble concentrating: recurrent thoughts of death (not just fear of dying), recurrent suicidal ideation without a specific plan, or a suicide attempt or a specific plan for committing suicide.
chronic course or tends to develop in to an episode of MDD (Klein et al., 2009). Based on the Diagnostic and Statistic Manual of Mental Disorders, 4th edition (DSM-IV) criteria, there are nine kinds of depressive symptoms. Some studies showed that there are specific symptoms of depression in people with different sociodemographic characters. For example, gender differences have been shown as important factors in the study of depression and depressive symptoms in several cases and gender differences in expression of depressive symptoms have also been reported (Marcus et al., 2011). Compared to males, females have a higher prevalence of chronic minor depression, dysthymia and somatic depression (Kessler et al., 1994; Angst and Merikangas, 1997; Silverstein, 1999), and are manifested as more recurrent brief depression episodes. Females reported experiencing significantly more fatigue, hypersomnia, and psychomotor retardation during the most severe major depressive episode, whereas males reported more insomnia and agitation (Khan et al., 2002). Currently there are very few studies examining the comorbidity of mood and anxiety disorders on symptoms level in China. This study was carried out in Chifeng City of Inner Mongolia, and was the first population-based sample epidemiological study on the prevalence of lifetime depression disorders in residents with anxiety disorders in that region. Chifeng City is one of the major cities in Inner Mongolia which is inhabited by a multi-ethnic population with a relatively low socio-economic status. It is a typical region in a period of economic transition with a large number of residents but relatively weak economy and overall low level of school education. More importantly, this study was also designed to explore the gender differences of depressive subtypes in anxiety patients and to provide theoretical evidence for early diagnosis and prevention strategy of mental disorders. 2. Materials and methods 2.1. Sample
2.3. Procedures A three-stage selection scheme was used to select the sample for every eligible individual (adults aged 18 years and over, and living in a family household) in the target population in 2010. “Neighborhood committee” (NC, in urban area) and “Village committee” (VC, in rural area) are local community organizations in China. Since all households belong to such committees, the committees were used as the primary sampling units (PSUs). In the first stage of sampling, 108 PSUs were selected based on the strategy of probabilities proportional to size (PPS). Stratification was made on urban and rural areas, with 57 PSUs from urban areas and 51 PSUs from rural areas. The second stage sampling selected a probability sample of households in each PSU. Based on the difference of estimated response rates, 80 households were selected in each VC, and 50 households were selected in each NC. The third stage sampling selected one random respondent in each sample household. Finally, 6376 respondents were selected in the survey.
62 lay interviewers were trained in a seven day training session at Peking university sixth hospital during November 2010. The detailed training methods can be found elsewhere (Wang et al., 2018). Only those trainees who passed the final test could be selected as interviewers. Face-to-face interviews were conducted by selected interviewers from November 2011 to April 2013. The average interview time was one hour. The study was approved by the ethics review board at Peking University Institute of Mental Health (PKUIMH). Before respondents were interviewed, written informed consent was obtained and declarations of anonymity and confidentiality were signed. 2.4. Statistical analyses Design weight was computed from the different probabilities of selection, and post-stratification weight was generated based on the age and gender distribution of census population. The consolidated weights were applied to the sample data. The weighted prevalence was calculated and the differences of depressive symptoms between male and female were analyzed by chi-square test. Additionally, logistics analysis was applied to explore the related factors of anxiety disorders comorbid 5 and over depressive symptoms. The dependent variable was anxiety disorders comorbid 5 and over depressive symptoms. The independent variables were social-demographic factors (sex, age, marital status, educational attainment, and employment) and childhood adversities. Latent Class Analysis (LCA) is a statistical technique that is used in factor, cluster, and regression techniques; it is a subset of structural equation modeling (SEM). LCA is a technique where constructs are identified and created from unobserved or latent subgroups, which are usually based on individual responses from multivariate categorical data. In this study, LCA on the items of depressive symptoms among the
2.2. Instruments and measures Composite International Diagnostic Interview (CIDI) was applied in the survey. CIDI is a fully structured lay-administered diagnostic interview (Kessler and Ustun, 2004) which has been widely used in more than 30 countries (Kessler and Ustun, 2004). Clinical reappraisal found generally good concordance with the CIDI diagnoses of mood disorder and anxiety disorders (Qin et al., 2010). Over the last twenty years, CIDI has been used as the primary instrument in regional epidemiological mental health surveys in many cities in China; such as Beijing and Shanghai (Shen et al., 2006), Jiangxi (Chen et al., 2004), Liaoning (Pan et al., 2006), Guangzhou (Zhao et al., 2009) and Kunming (Jian et al., 2010). In this study, anxiety disorders and depression disorders were diagnosed using CIDI-3.0, according to the criteria and definition of the DSM-IV. Additionally, social-demographic information 361
Journal of Affective Disorders 243 (2019) 360–365
H. Chen et al.
3.4. Risk factors of anxiety comorbid more than 4 kinds of depressive symptoms
respondents with anxiety was performed with SAS9.4. LCA (Hutchinson and Tang, 1987; Yang and Becker, 1997) is a statistical technique for identifying mutually exclusive classes of respondents, each with its own set of symptom endorsement probabilities. In terms of LCA, a solution of classes is optimal when classes are as homogeneous as possible, while differences between classes are as large as possible. In LCA, classes are added until the model fits the data well. Fifty random sets of starting values were used. Models ranging from 2 to 5 classes were evaluated. Model fit was evaluated with the following commonly-used statistics: Akaike information criterion (AIC), the Bayesian Information Criterion (BIC), sample size adjusted BIC (Vrieze, 2012), and Entropy values. Smaller values of AIC and BIC, and larger values of Entropy, indicate better model fit (Nylund et al., 2007). Gender was included in the multiple-group latent class analysis.
Table 4 also presents the results of logistics analysis. Considering social demographic factors and childhood adversities, childhood adversities were statistically significant risk factors (OR = 2.7). 3.5. Multiple-group latent class analysis LCA revealed that four classes provided the best fit statistics. Fit statistics for the final 4-class model were: AIC = 329.63; BIC = 583.31; Adjusted BIC = 336.23; Entropy = 0.84. Table 5 depicts the observed class membership and endorsement frequencies of the 9 disaggregated depressive symptoms in the best-fitting LCA solution. Class 1 (24.7% of respondents with anxiety) was characterized by very high endorsement rates for most of the 9 disaggregated symptoms of major depression. Class 1 was termed as “severe typical” because of the nearly universal presence of major depression along with classical depressive symptoms (i.e., weight and appetite loss, insomnia, psychomotor retardation, anergia, and poor concentration). Class 2 (31.3% of respondents with anxiety) was characterized by prominent symptoms of low mood and loss of interest in females but nearly no symptoms in males. Thus class 2 was termed as “not depressed”. Like class 1, class 3 (25.5% of respondents with anxiety) was characterized by many depressive symptoms, however, the symptom pattern was more characterized by weight decrease/increase and insomnia/hypersomnia but less characterized by feelings of guilt or worthlessness and thoughts of death/suicide. Class 3 was termed “moderate typical”. Class 4 (18.5% of respondents with anxiety) was characterized by atypical depressive symptoms, being termed as “mild atypical.” From the results of multiple-group latent class analysis, it can be seen that the latent class probabilities were different between males and females. In male respondents with anxiety, the proportions of these four classes were 36.9%, 24.7%, 17.2%, 21.2% respectively while in female respondents with anxiety, the proportions of these four classes were 29.1%, 18.6%, 23.6%,28.7% respectively. The female group in class 2 was characterized by a very high endorsement rate of prominent symptoms while the male group was not. Besides, in class 4, the female group was characterized by thoughts of death while the male group was not. (Table 5)
3. Results 3.1. Sample description A total of 4528 respondents were interviewed in this study. The response rate was 71.2%. More participants are women (53.9%), married or cohabiting (93.1%), of low educational level (87.6%), lived in rural area (67.2%), and reported no steady job (71.1%). Table 1 shows more information of socio-demographic characteristics of the Chifeng sample by gender. 3.2. Prevalence of anxiety disorders and comorbid depression The prevalence estimates for anxiety in the total sample were 5.70% for lifetime and 3.96% for the 12 months before the interview. Among residents who had ever had anxiety, 17.59% of them had depression while 15.79% of them met the criteria of MDD. Prevalence rates of comorbidity between people with different sociodemographic characters were shown in Table 2. There were significant differences between males and females. 3.3. Comorbid depressive symptom Among respondents who met the diagnosis for anxiety disorders, most of them had more than one kind of depressive symptom. There was no significant difference between males and females. 64.31% of those respondents had more than 4 kinds of symptoms but did not meet the criteria of major depression disorder since the duration of symptoms was less than 2 weeks. (Table 3)
4. Discussion 4.1. Discussion Using a representative data of Chifeng City in Inner Mongolia Autonomous Region of China, results confirmed high lifetime comorbid depression disorders rates in anxiety disorders. Of persons with anxiety disorder, 17.59% had a lifetime comorbid depressive disorder in which rates of comorbid anxiety disorder were higher for those with MDD than for those with MND. The results were similar to the reports of some previous regional surveys in China, but were lower than some CIDI surveys in other counties (King-Kallimanis et al., 2009; Johansson et al., 2013; Schuch et al., 2014). Besides, there are statistically significant differences of comorbidity between males and females: as is consistent with previous studies, women reported more comorbid depression when compared to men. In one longitudinal epidemiological study, the prevalence of MDD and of one or two anxiety disorders was about two-fold higher in women than in men (Breslau et al., 1995). Besides, there were 62.08% of the respondents with anxiety who also had more than 4 kinds of symptoms but did not meet the criteria of depression, as the duration of symptoms was less than 2 weeks. These results showed that females with more than 4 kinds of depressive symptoms were more likely to meet the criteria of MDD than male due to the low level of emotion-regulation ability. Besides, although they did not meet criteria for depression, these symptoms still made them
Table 1 Socio-demographic characteristics of the Chifeng sample by gender. Characteristic
Age
Area Marital status
Education Income
Male
18–34 35–49 50–64 65Rural Urban Married/cohabitating Divorced/separated/widowed/ never married Lowera Higherb No Yes
Female
N
%
N
%
303 763 759 260 1445 640 1946 139
30.12 34.94 24.98 9.96 68.54 31.46 86.15 13.85
432 961 771 279 1596 847 2270 173
33.43 36.58 21.39 8.60 66.24 33.76 89.35 10.65
1675 410 479 1606
77.23 22.77 23.93 76.08
2058 385 1048 1395
80.09 19.91 43.13 56.87
a The respondents who received less than or equal to nine years were categorized as having lower education level. b The respondents who completed more than nine years of school were categorized as having a higher education level.
362
Journal of Affective Disorders 243 (2019) 360–365
H. Chen et al.
Table 2 Prevalence of depression disorders in anxiety disorder patients. Characteristic
Gender Age
Area Education Income
a
MNDa
MDD
Male Female 18–34 35–49 50–64 65Rural Urban Lower Higher No Yes
%
Se
P
χ
8.96 19.77 12.92 12.19 18.37 27.90 13.82 23.80 15.70 16.20 18.43 13.90
2.79 3.29 5.43 3.34 3.80 8.98 2.70 5.97 2.70 6.46 3.87 2.84
0.010
6.569
0.234
4.273
0.089
2.887
0.944
0.005
0.293
1.106
2
Depression
%
se
P
χ
0.83 2.98 – 2.80 2.28 5.50 2.00 2.95 2.35 1.44 1.52 2.67
0.84 1.19 – 1.63 1.61 3.76 0.85 2.07 0.92 1.45 1.05 1.20
0.215
1.535
–
–
0.632
0.229
0.644
0.213
0.498
0.459
2
%
Se
P
χ2
9.79 22.13 12.92 13.84 20.66 33.40 15.33 26.75 17.58 17.64 19.94 15.89
2.90 3.34 5.43 3.70 3.96 9.12 2.60 6.68 2.60 7.10 3.84 2.88
0.006
7.478
0.098
6.297
0.071
3.269
0.993
<0.001
0.352
0.868
Minor depression disorder.
programs should target not only the people who met the diagnostic criteria but also the people with subthreshold mental disorder. This paper reported gender differences in anxiety comorbid symptomatic profiles of depression. The pattern of comorbidity is similar with previous study (Sheila M Marcus, 2011). It is the same in class 1 and class 3, indicating that typical depressive symptoms of females and males showed no significant differences. GordonParker's study indicated that female unipolar groups scored only marginally (and nonsignificantly) higher than male groups on the depression severity measures. Gender had minimal if any impact on unipolar depression (Parker et al., 2014). While in class 2, females were more likely to have feelings of depression and loss of interest, and to have more kinds of symptoms than males. One explanation is that compared to males, females with anxiety disorders are more sensitive (Park et al., 2015). Gonadal hormones may be a major factor in this disparity, given that women are more likely to experience mood disturbances during times of hormonal flux, and testosterone may have protective benefits against anxiety and depression (McHenry et al., 2014). Besides, when being stressed, females seek support and express their feelings more than males (Rose and Rudolph, 2006). Therefore, they may receive more relationship provisions, which may later contribute to a positive emotional adjustment. This might be the reason why females had major symptoms that were categorized in “not depressed” class. In class 4, the female group was characterized by thoughts of death but had less symptoms than males. This result was similar with previous study. Ruth's study (Derdikman-Eiron et al., 2012) showed that gender was a moderator variable in the associations between symptoms of anxiety and depression and impairment, meaning that boys' functioning was impaired to a larger extent than girls' functioning. These results suggest that precision prevention should be targeted based on the specific kind of population. Any given prevention strategy may not be equally effective in all individuals because of biological differences in risk and response to the preventive modality. Unlike precision medicine, which has traditionally focused on molecular data to target specific treatments to an individual, precision prevention requires a broader conceptual framework. Consequently, the integration of a broad range of research domains is required, which includes cancer biology, epidemiology, biomedical informatics, bioinformatics, biostatistics, risk estimation, genetics, biomarkers, disparities, health services research, clinical trials, geospatial analysis, and dissemination and implementation research. These domains should be integrated under a community engagement framework to achieve the future precision prevention goals.
Table 3 Anxiety disorders comorbid depressive symptoms by gender. Total
Feelings of depression Loss of interest Weight decrease/increase insomnia/hypersomnia psychomotor agitation/ retardation Tiredness/lack of energy Feelings of guilt or worthlessness Trouble concentrating Thoughts of death/suicide More than 4 kinds of symptoms Depression
Male
P
Female
N
%
N
%
N
%
131 120 105 118 66
68.59 62.83 74.47 83.69 46.81
38 32 31 34 22
62.62 57.93 81.01 84.60 56.77
93 88 74 84 44
71.14 66.83 71.13 84.89 44.34
0.336 0.331 0.296 0.970 0.251
111 64 120 69 116 53
78.72 45.39 85.11 48.94 38.28 22.56
30 20 35 18 32 11
75.47 52.40 86.30 49.14 37.03 9.79
81 44 85 51 84 42
82.23 40.33 87.43 46.18 49.73 22.12
0.426 0.313 0.857 0.775 0.067 0.006
Table 4 Logistic models of anxiety disorders comorbid more than 4 kinds of depressive symptoms. Effect Gender Area Age
Marital status
Employment Income Education Childhood adversities
OR Male Female Rural Urban 18–34 35–49 50–64 65Married/cohabitating Divorced/separated/ widowed/ never married No Yes No Yes Lowera Higherb No Yes
1 1.48 1 0.73 1 2.24 1.27 1.93 1 1.23 1 1.16 1 0.65 1 2.12 1 2.7
(95%CI)
(0.85,2.58) (0.37,1.47) (0.87,5.75) (0.49,3.29) (0.65,5.71) (0.46,3.33)
(0.58,2.33) (0.38,1.11) (0.86,5.24) (1.10,6.60)
a The respondents who received less than or equal to nine years were categorized as having a lower education level. b The respondents who completed more than nine years of school were categorized as having a higher education level.
feel stressed most of the day at that time and influenced their daily life. In light of these findings, not only case-level depression but also symptom-level depressive state should be taken into consideration in future studies on comorbidity. Prevention and early intervention
4.2. Limitations Limitations of this study include the cross-sectional character of this 363
Journal of Affective Disorders 243 (2019) 360–365
H. Chen et al.
Table 5 Multiple-group latent class analysis of depressive symptoms among respondents with anxiety by gender. Male
Feelings of depression Loss of interest Weight decrease/increase Insomnia/hypersomnia Psychomotor agitation/retardation Tiredness/lack of energy Feelings of guilt or worthlessness Trouble concentrating Thoughts of death/suicide Percent of sample by gender Average number of symptoms
Female
Class1 Severe typical
Class2 Not depressed
Class3 Moderate typical
Class4 Mild atypical
Class1 Severe typical
Class2 Not depressed
Class3 Moderate typical
Class4 Mild atypical
0.928 0.994 0.948 0.853 0.726 0.851 0.710 0.998 0.826 0.369 7.92
0.233 0.086 0.044 0.052 0.028 0.519 0.267 0.524 0.036 0.247 0.40
0.992 0.990 0.686 0.996 0.015 0.380 0.192 0.585 0.013 0.172 4.69
0.568 0.116 0.725 0.994 0.979 0.991 0.384 0.995 0.165 0.212 6.33
0.983 0.997 0.824 0.923 0.522 0.939 0.713 0.998 0.969 0.291 7.97
0.895 0.690 0.315 0.475 0.053 0.385 0.133 0.674 0.184 0.186 4.02
0.852 0.959 0.993 0.998 0.567 0.954 0.360 0.783 0.131 0.236 6.58
0.105 0.008 0.525 0.991 0.968 0.988 0.287 0.990 0.967 0.287 0.80
References
study, which does not allow for making causal inferences and the data was based on retrospective self-reports, which is subject to recall bias. So it is important to be cautious in interpreting the results because recall bias almost certainly exists in retrospective study. Furthermore, mental health was possibly underestimated because of the respondents’ stigma of psychological questions.
Angst, J., Merikangas, K., 1997. The depressive spectrum: diagnostic classification and course. J. Affect. Disord. 45 (1-2), 31–39 discussion 39-40. Breslau, N., Schultz, L., et al., 1995. Sex differences in depression: a role for preexisting anxiety. Psychiatry Res. 58 (1), 1–12. Chen, H.L., Bin, H.U., et al., 2004. Epidemiological investigation on mental disorders in Jiangxi Province. Chin. J. Psychiatry 37 (3), 172–175. Cyranowski, J.M., Schott, L.L., et al., 2012. Psychosocial features associated with lifetime comorbidity of major depression and anxiety disorders among a community sample of mid-life women: the SWAN mental health study. Depress Anxiety 29 (12), 1050–1057. Derdikman-Eiron, R., Indredavik, M.S., et al., 2012. Gender differences in psychosocial functioning of adolescents with symptoms of anxiety and depression: longitudinal findings from the Nord-Trøndelag Health Study. Social Psychiatry Psychiatric Epidemiol. 47 (11), 1855–1863. Hutchinson, T.P., Tang, K.Y., 1987. "The value of latent class analysis in medical diagnosis. Stat. Med. 6 (4), 529–531. Jian, Y., Qin, H.Y., et al., 2010. Epidemiological survey of affective disorder in Kunming City. Chin. Mental Health J. 24 (2), 110–115. Johansson, R., Carlbring, P., et al., 2013. Depression, anxiety and their comorbidity in the Swedish general population: point prevalence and the effect on health-related quality of life. Peer J. 1, e98. Kessler, R.C., McGonagle, K.A., et al., 1994. Sex and depression in the national comorbidity survey. II: cohort effects. J. Affect. Disord. 30 (1), 15–26. Kessler, R.C., Ustun, T.B., 2004. The world mental health (WMH) survey initiative version of the world health organization (WHO) composite international diagnostic interview (CIDI). Int. J. Methods Psychiatr. Res. 13 (2), 93–121. Khan, A.A., Gardner, C.O., et al., 2002. Gender differences in the symptoms of major depression in opposite-sex dizygotic twin pairs. Am. J. Psychiatry 159 (8), 1427–1429. King-Kallimanis, B., Gum, A.M., et al., 2009. Comorbidity of depressive and anxiety disorders for older Americans in the national comorbidity survey-replication. Am. J. Geriatr. Psychiatry 17 (9), 782–792. Klein, D.N., Shankman, S.A., et al., 2009. Subthreshold depressive disorder in adolescents: predictors of escalation to full-syndrome depressive disorders. J. Am. Acad. Child Adolesc. Psychiatry 48 (7), 703–710. Lamers, F., Oppen, P.V., et al., 2011. Comorbidity patterns of anxiety and depressive disorders in a large cohort study: the Netherlands study of depression and anxiety (NESDA). J. Clin. Psychiatry 72 (3), 341. Marcus, S.M., Kerber, K.B., et al., 2011. Gender differences in depression symptoms in treatment-seeking adults: STAR*D confirmatory analyses. Comprehensive Psychiatry 49 (3). McHenry, J., Carrier, N., et al., 2014. Sex differences in anxiety and depression: role of testosterone. Front. Neuroendocrinol. 35 (1), 42–57. Nylund, K.L., et al., 2007. Deciding on the number of classes in latent class analysis and growth mixture modeling: a Monte Carlo simulation study. Struct. Eq. Model. Multidisciplin. J. 4 (14), 535–569. Pan, G., Chao, J., et al., 2006. Epidemiological survey of mental disorders in urban and rural areas of Liaoning province. Chin. J. Public Health 73 (2), 13–18. Park, S.C., Lee, M.S., et al., 2015. Gender differences in depressive symptom profiles and patterns of psychotropic drug usage in Asian patients with depression: findings from the research on Asian psychotropic prescription patterns for antidepressants study. Aust. N. Z. J. Psychiatry 49 (9), 833–841. Parker, G., Fletcher, K., et al., 2014. Gender differences in depression severity and symptoms across depressive sub-types. J. Affect. Disord. 167, 351–357. Prince, M., Patel, V., et al., 2007. No health without mental health. Lancet 370 (9590), 859–877. Qin, H.Y., Xie, S.F., et al., 2010. Community-based evaluation of the reliability and validity of Chinese version of composite international diagnostic interview-3.0. Chin. Mental Health J. 24 (1) 21-20. Rose, A.J., Rudolph, K.D., 2006. A review of sex differences in peer relationship processes: potential trade-offs for the emotional and behavioral development of girls and boys.
5. Conclusion This is the first regional representative cross-sectional survey of mental disorders in Chifeng City, the Inner Mongolia Autonomous Region of China. The prevalence rates of comorbidity were similar to the reports of previous regional surveys in China and there are statistically significant differences of comorbidity between males and females. Bearing in mind the importance of precision prevention, our findings may contribute to improved assessment and intervention methods tailored differently to males and females with symptoms of anxiety and depression. Competing of Interests The authors have declared that no competing interests exist. Disclaimers The views expressed in the submitted article are our own and not an official position of the institution or funder. Source of support This research was supported by the Bureau public health of Chifeng, China and The National Key Technology R&D Program of China (No. 2015BAI13B00). Acknowledgments This research was supported by the Bureau public health of Chifeng, China and The National Key Technology R&D Program of China (No. 2015BAI13B00). We thank all of the colleagues in the Chifeng Anding Hospital and Peking University Sixth Hospital for providing data and assisting the data analysis. We also thank Dr. William A Jefferson from Chinese Academy of Science for help with English grammar and wording in our manuscripts. Supplementary materials Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.jad.2018.09.053. 364
Journal of Affective Disorders 243 (2019) 360–365
H. Chen et al.
Commission of Health and Function 2016-present Vice-president, China Disabled Persons’ Federation 2013-present President, Chinese Mental Health Journal 2013-present Honorary Professor in the University of Hong Kong 2003-present Director, Division of Social Psychiatry and Behavior Medicine, Institute of Mental Health, Peking University 2002-present 6. President, Society of Crisis Intervention of Chinese Association for Mental Health
Psychol. Bull. 132 (1), 98–131. Rui, L.Z., Qin, H.Y., et al., 2013. The prevalence of mood disorder, anxiety disorder and substance use disorder in community residents in Beijing: a cross-sectional study. Chin. Mental Health J. 27 (2), 102–110. Schuch, J.J., Roest, A.M., et al., 2014. Gender differences in major depressive disorder: results from the Netherlands study of depression and anxiety. J. Affect. Disord. 156, 156–163. Sheila, MMarcus,K.B.K.A., 2011. Gender differences in depression symptoms in treatment-seeking adults: STAR*D confirmatory analyses. Comprehens. Psychiatry 49 (3). Shen, Y.C., Zhang, M.Y., et al., 2006. Twelve-month prevalence, severity, and unmet need for treatment of mental disorders in metropolitan China. Psychol. Med. 36 (2), 257–267. Silverstein, B., 1999. Gender difference in the prevalence of clinical depression: the role played by depression associated with somatic symptoms. Am. J. Psychiatry 156 (3), 480–482. van Loo, H.M., Schoevers, R.A., et al., 2016. Psychiatric comorbidity does not only depend on diagnostic thresholds: an illustration with major depressive disorder and generalized anxiety disorder. Depression Anxiety 33 (2), 143–152. Vrieze, S.I., 2012. Model selection and psychological theory: a discussion of the differences between the Akaike information criterion (AIC) and the Bayesian information criterion (BIC). Psychol. Methods 17 (2), 228–243. Wang, X., Liu, Z., et al., 2018. Association of comorbidity of mood and anxiety disorders with suicidal behaviors. J. Affect. Disord. 227, 810–816. Whiteford, H.A., Degenhardt, L., et al., 2013. Global burden of disease attributable to mental and substance use disorders: findings from the global burden of disease study 2010. Lancet 382 (9904), 1575–1586. Yang, I., Becker, M.P., 1997. Latent variable modeling of diagnostic accuracy. Biometrics 53 (3), 948–958. Zhao, Z., Huang, Y., et al., 2009. An epidemiological survey of mental disorders in Guangzhou area. Chin. J. Nervous Mental Dis.
2. 3. 4. 5.
Clinical Posts: Psychiatrist, The Sixth Hospital, Peking University 2002-present Psychiatrist, Institute of Mental Health, Beijing Medical University 1987–1996
Current and Recent (Relevant) Research Support 1. Chinese Ministry of Technology. Multilevel analysis on predictors of mental health service use among patients with affective and cognitive disorders. (2015–2017) Key personnel, ¥23,540,000 2. European Research Commission. 10/66 ten years on – monitoring and improving health expectancy by targeting frailty among older people in middle income countries. (2014–2019) Country PI, €137,900 3. Beijing Health Bureau. Community based intervention trail on families with dementia patients, (2012–2015) co-PI, ¥104,000 4. Chinese Ministry of Technology. National epidemiological survey of mental disorders. (2012–2015) PI. ¥4,760,000 5. Chinese Ministry of Health. Study on disease burden of mental disorders and mental health service. (2012–2015) PI.¥20,220,000 6. WHO WPRO. Study on suicide surveillance in Kunming City, Yunnan Province, China. (2011–2014). PI, $25,024 7. Beijing Scientific and Technology Committee. Optimization of treatment of rTMS on depression. (2010–2012). PI, ¥1,600,000 8. Suicide study and prevention: China Collaborative Suicide Research Training Program– (No.D43TW007273-05)(2006.1 −2010.12)
Yueqin Huang:
Education: Bachelor of Medicine (MD equivalent), Beijing Medical University, 1984 Master of Medicine (MPH equivalent), Beijing Medical University, 1987 Ph.D, Peking University, 2004
Publications The editor-in-Chief of six books, and 276 papers including 142 first-author and correspondent-author papers
Academic Posts: 1. Member, Executive Committee of Rehabilitation International and Chair of
365