- Email: [email protected]
The relationship between lifestyle factors and clinical symptoms of bipolar disorder patients in a Chinese population
Accepted Manuscript
The relationship between lifestyle factors and clinical symptoms of bipolar disorder patients in a Chinese population Jia Huang , Cheng Mei Yuan , Xian Rong Xu , Yong Wang , Wu Hong , Zuo Wei Wang , You song Su , Ying Yan Hu , Lan Cao , Yu Wang , Jun Chen , Yi Ru Fang PII: DOI: Reference:
S0165-1781(17)30273-1 10.1016/j.psychres.2018.04.059 PSY 11390
To appear in:
Psychiatry Research
Received date: Revised date: Accepted date:
20 February 2017 7 March 2018 22 April 2018
Please cite this article as: Jia Huang , Cheng Mei Yuan , Xian Rong Xu , Yong Wang , Wu Hong , Zuo Wei Wang , You song Su , Ying Yan Hu , Lan Cao , Yu Wang , Jun Chen , Yi Ru Fang , The relationship between lifestyle factors and clinical symptoms of bipolar disorder patients in a Chinese population, Psychiatry Research (2018), doi: 10.1016/j.psychres.2018.04.059
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Highlights
We explored the associations between Life style factors and clinical symptoms in BD patients with depressive symptoms, using the
17 item Hamilton
Rating Scale for Depression (HAMD-17) to assess symptom severity. The consumption of whole grain and dairy products showed an independent
CR IP T
negative correlation with the clinical symptoms and HAMD-17 total scores in BD patients.
Our results highlight the importance of Life style management in the
AN US
AC
CE
PT
ED
M
improvement of clinical symptoms in BD patients.
ACCEPTED MANUSCRIPT
The relationship between lifestyle factors and clinical symptoms of bipolar disorder patients in a Chinese population
CR IP T
Running title: Bipolar disorder symptoms and lifestyle
Authors: Jia Huang a, Cheng Mei Yuan b, Xian Rong Xu c, Yong Wang a, Wu Hong a, Zuo Wei Wang d, You song Su a, Ying Yan Hu a, Lan Cao a, Yu Wang d, Jun Chen a*, Yi
a
AN US
Ru Fang a, e, f,**
Department of Mood Disorder, Shanghai Mental Health Center, Shanghai Jiao Tong
b
M
University School of Medicine, 600 Wan Ping Nan Road, Shanghai 200030, PR China First Department of General Psychiatry, Shanghai Mental Health Center, Shanghai
Department of Prevention Medicine, School of Medicine, Hangzhou Normal
PT
c
ED
Jiao Tong University School of Medicine, Shanghai 200030, PR China
University, 16 Xue Lin Road, Hangzhou, PR China Division of Mood Disorders, Hongkou District Mental Health Center of Shanghai,
CE
d
AC
Shanghai 200083, PR China e
State Key Laboratory of Neuroscience, Shanghai Institute for Biological Sciences,
CAS, 320 Yue Yang Road Shanghai, 200031 PR China f
Shanghai Key Laboratory of Psychotic Disorders, 600 Wan Ping Nan Road,
Shanghai 200030, PR China
ACCEPTED MANUSCRIPT
*Corresponding author. Tel.: +86 021 34773367; Fax: +86 021 34773367 **Corresponding author. Tel.: +86 21 64387250x73529; Fax: +86 21 64387986 E-mail addresses: [email protected] (Jun Chen); [email protected]
CR IP T
(Yi Ru Fang)
AN US
Abstract
There is evidence that bipolar disorder (BD) patients with an unhealthy lifestyle have a worse course of illness. This study was designed to examine the extent to which
M
lifestyle could influence the severity of clinical symptoms associated with BD. A total of 113 BD patients were recruited in this study. The lifestyle information including
ED
data on dietary patterns, physical activity, and sleep quality were collected using a
PT
self-rated questionnaire. The results showed that the consumption of whole grain, seafood, and dairy products were significantly negatively correlated with the 17-item
CE
Hamilton Rating Scale for Depression (HAMD-17) total score. The consumption of
AC
sugar, soft drinks, and alcohol as well as being a current smoker were positively correlated with the severity of clinical symptoms. Multiple linear regression and binary logistic regression analyses demonstrated an independent negative correlation between both whole grain and dairy product consumption with the HAMD-17 score. The results from the current study suggested that lifestyle factors, especially dietary patterns, might be associated with clinical symptoms of BD. The association between
ACCEPTED MANUSCRIPT
the consumption of specific foods and severity of depressive symptoms may offer some useful information and further understanding of the role of lifestyle factors in
CR IP T
the development of BD.
Keywords: bipolar disorder; symptom assessment; depression; lifestyle; diet
AN US
1. Introduction
Bipolar disorder (BD) is a severe and recurrent psychiatric illness associated with substantial functional impairments and premature mortality (Belmaker, 2004).
M
Globally, the overall point-prevalence of BD is estimated to be 0.741% (Ferrari et al., 2011) and BD accounts for nearly 1,320,000 excess deaths each year (Patel et al.,
ED
2016). Data from a nationally representative survey in the U.S. demonstrated that the
PT
BD affected approximately 4.4% of people in United States (Merikangas et al., 2007) and incurred disproportionately higher costs for both patients and the mental health
CE
care system than most other mental illnesses do (Kessler et al., 2012).In China, the
AC
estimated point, 12-month and lifetime prevalence of BD in China was 0.09% (95% confidence interval [CI]: 0.06-0.12%), 0.17% (95% CI: 0.10-0.29%) and 0.11% (95% CI: 0.07-0.17%), respectively (Zhang et al., 2017). These results suggest that more than 1.37 million adults in the country may have BD, and even worse, most of them never receive any type of professional interventions for their conditions (Phillips et al., 2009). Therefore, it is important for China to identify the factors involved in the
ACCEPTED MANUSCRIPT
initiation and development of BD. BD is a complex disease, and its etiology has not been fully elucidated. Although genetics plays a significant role in the susceptibility to this illness, evidence from several studies suggests that lifestyle factors, such as diet patterns, exercise and sleep
CR IP T
quality also appear to be important (De Hert et al., 2011). Sylvia et al. found that less frequent weekly exercise was associated with higher BMI, more time spent depressed, more depressive symptoms, and lower quality of life and functioning in BD patients
AN US
(Sylvia et al., 2013). Compared to those with no psychopathology, BD patients had a lower quality of life and a higher glycemic load in their diet (Jacka et al., 2011b). Elmslie et al. found that BD patients consumed more total carbohydrates, sucrose,
M
nonalcoholic beverages, sweetened drinks, cakes and sweets (Elmslie et al., 2001). The authors also found that female BD patients, but not male patients, also had greater
ED
total energy intake. Noaghiul et al. demonstrated that high rates of seafood
PT
consumption were associated with lower lifetime prevalence rates of bipolar I disorder, bipolar II disorder, and bipolar spectrum disorder (Noaghiul and Hibbeln, 2003).
CE
Noguchi et al. found that in BD patients, plant-based food and fish product
AC
consumption patterns were inversely related to physical and psychiatric symptoms, and in men, this pattern demonstrated an inverse relationship with psychiatric symptoms (Noguchi et al., 2013). Lifestyle factors, such as tobacco use and weight gain/obesity, were associated with an increased severity of BD symptoms (Cerimele and Katon, 2013). Despite these promising data on the importance for lifestyle factors in bipolar disorder, little is known about the influence of lifestyle on the remission of
ACCEPTED MANUSCRIPT
the disease and how lifestyle factors might affect bipolar specific mood symptoms. To address this issue, we examined the association between lifestyle factors and clinical characters in BD patients. We hypothesized that a higher quality lifestyle would be associated with a decreased severity of clinical symptoms and improved
CR IP T
odds of clinical remission of patients with BD.
2. Methods
AN US
2.1 Patient population
Participants were recruited from the Shanghai Mental Health Center between July 2015 and May 2016. Subjects in this analysis met the following inclusion criteria: (1)
M
18-65 years of age; (2) inpatients and outpatients meeting DSM-IV criteria for bipolar I or II disorder with a current major depressive episode; and (3) willingness and
ED
ability to perform the necessary clinical tests. Subjects were excluded if they met any
PT
of the following criteria: (1) were unable to provide informed consent (assessed by using a short questionnaire asking key questions about the study); (2) were with any
CE
history or presence of serious physical disease, endocrine or autoimmune disease; (3)
AC
were pregnant or lactating women; (4) had an infection or fever within the last 2 months; or (5) had received any prior systemic hormonal therapy or cytokine therapy. All of the patients with bipolar disorder in this study were diagnosed by at least two chief physicians. The study protocol and methodologies were reviewed and approved by the Institutional Review Board of Shanghai Mental Health Center, and all protocols related to human experiments were conducted in accordance with the
ACCEPTED MANUSCRIPT
Declaration of Helsinki. We ensured that all participants were given an adequate understanding of the study, and written informed consent was obtained from all individuals prior to their inclusion in the study.
CR IP T
2.2 Study measures Demographic data on age, gender, marital status, educational level, income, body mass index (BMI), the duration of major depressive episodes, frequency of exercise,
AN US
history of smoking, and alcohol and other drug use were collected from the participants. A self-administered physical activity questionnaire was used to assess physical activity levels in patients. The patients were asked about the frequency and
M
time spent walking for transport and walking for recreation and leisure as well as about moderate- and vigorous-intensity activity over the previous year. The family
ED
history of psychiatric disorders was collected according to the Family History
PT
Research Diagnostic Criteria (HF-DRC), which has been reported to have a specificity of 96% and sensitivity of 52% (Andreasen et al., 1986). The Hamilton
CE
Rating Scale for Depression-17 (HAMD-17) was used to assess the severity of
AC
depressive symptoms. Clinical remission in patients with bipolar disorder was defined as having a HAMD-17 total score ≤7, which is the current widely accepted cut-off value. The Pittsburgh Sleep Quality Index (PSQI) was used to measure the quality of sleep. All assessments were conducted in the afternoon to avoid the effects of diurnal variation in cortisol levels and mood. All assessments were performed by the same trained rater.
ACCEPTED MANUSCRIPT
2.3 Dietary survey Participants were visited at hospital by specifically trained dieticians. A self-administered semi-quantitative food frequency questionnaire (FFQ) was used to measure the habitual consumption items from 13 different food groups (fruits and
CR IP T
vegetables, whole grains, dairy products, red meat, poultry, fish, eggs, nuts, cooking oils, milk, soft drinks, alcoholic beverages, and fruit and vegetable juices), for a total of 41 items. The questionnaire was adapted from the questionnaire designed by the
AN US
Chinese Nutrition Society (CNS) in 2010 (Zhang et al., 2009). Before administering the diet survey, the patients were given oral and written information of how to carry out the survey. The survey collected information including the frequency of
M
consumption (daily, weekly or monthly) and the quantity of the foods consumed. Patients were asked about their habitual consumption of the listed foods within the
ED
past 1 year. The intake of energy, protein, fat, and carbohydrates were calculated
et al., 2004).
PT
based on the Standardized Tables of Food Composition of China (2004 edition) (Yang
CE
2.4 Statistical analysis
AC
All of the data were analyzed using the software SPSS 19.0 (SPSS, Chicago, IL, USA). Continuous variables such as age, BMI, serum lipids and HAMD-17 total score are presented as the means ± SD. Categorical variables, such as gender, education level, marital status, and family history of psychiatric disorders, are presented as frequencies and percentages. Spearman’s correlation analyses were used to determine the relationship between clinical characters and lifestyle factors in
ACCEPTED MANUSCRIPT
bipolar disorder patients. Multiple linear regression models were used to examine the relationships between the HAMD-17 total score and lifestyle factors after adjusting for potential confounders, which included age, gender, number of depressive episodes, and age at first onset. A backward binary logistic regression analysis was performed to
CR IP T
independently evaluate the effects of lifestyle factors on the clinical remission of the disease, using “remission status” as a dependent variable and lifestyle factors as independent variables. The continuous variables, including the intake of different
AN US
kinds of foods, sleep time, and age, were converted into categorical variables to more accurately evaluate the associations between lifestyle factors and the clinical
M
remission of disease. For all analyses, the significance level was set as P< 0.05.
3. Results
ED
3.1 Patient characteristics
PT
We included 113 patients with bipolar disorder in this study; 65 (57.5%) subjects were inpatients, and 48 (42.5%) subjects were outpatients. The patients had a mean
CE
age of 36.6±15.7; 36 (32%) were male. The mean age at onset of the first mood
AC
episode was 29.9±14.6, with a mean duration of illness of 7.78±8.93 years. The mean age at first contact with a mental health professional was 29.4±15.0 years, whereas the mean age at the correct diagnosis and treatment was 31.5±13.9 years; the mean duration of untreated illness was 2.1±1.1 years. Based on the HAMD-17 scores, 58 (51.3%) patients were categorized as being in clinical remission (values below 7). A range of mood-stabilizers, antipsychotics, antidepressants and sedative hypnotic agent
ACCEPTED MANUSCRIPT
medications were prescribed to the patients. Overall, 74.8% of total patients received at least one kind of drug. All sociodemographic and clinical variables are presented in Table 1. The lifestyle characteristics of the patients are shown in Table 2. The mean caloric
CR IP T
intake was 1939.3±772.5 kcal/d. The daily intakes of carbohydrates, fat, and protein were 258.3±114.0 g, 55.2±35.1 g and 74.3±33.3 g, respectively. The mean or median intake of 13 food items, physical activity and sleep quality are also shown in Table 2.
AN US
3.2 Clinical characteristics and lifestyle
Simple correlation analysis indicated a significant association between lifestyle and clinical characters in BD patients (Table 3). The HAMD-17 score was negatively
M
associated with whole grains, seafood and dairy product consumption and total sleep time. The number of depressive episodes was positively associated with egg,
ED
vegetable, soft drink and alcoholic consumption. The number of manic episodes was
PT
positively associated with red meat, soft drink and alcohol consumption as well as with smoking. The age at first onset was negatively associated with red meat, soft
CE
drink, and sugar consumption and positively associated with nut consumption.
AC
Multiple linear regression analysis supported the association between lifestyle and the HAMD-17 score after controlling for potential confounding factors (Table 4). The results showed that whole grain and red meat consumption and total sleep time were negatively associated with the baseline HAMD-17 total score (P<0.05). The intake of dairy products was also found to be negatively associated with the HAMD-17 score, though this relationship was not statistically significant (P=0.055).
ACCEPTED MANUSCRIPT
We used a logistic regression model to assess the association between lifestyle factors and clinical remission of BD after controlling for potential confounding factors, including age, gender, number of depressive episodes, age at first onset of BD, BMI, total energy, intakes of fruits, vegetables, seafood, amount of physical activity,
CR IP T
smoking, alcohol consumption, family income, and education level (Table 5). The results showed that the intake of whole grains and dairy products was associated with the clinical remission of disease after controlling for potential confounding factors.
AN US
Compared to the reference level (<200 g/d), the intake of >400 g/d whole grain was associated with a statistically higher probability of clinical remission of the disease (OR=14.25, 95% CI: 1.31, 80.23). Compared to those who did not consume dairy
M
products, consumption of >100 g/ml of dairy products was associated with a statistically higher probability of clinical remission of the disease after adjusting for
PT
ED
potential confounding factors (OR=9.93, 95% CI: 2.28, 43.36).
4. Discussion
CE
Several studies have explored the association between clinical symptoms and
AC
lifestyle in BD patients (Jacka et al., 2011b; Noaghiul and Hibbeln, 2003; Noguchi et al., 2013; Sylvia et al., 2013). Lifestyle interventions for BD patients targeting diet and exercise have shown promise in reducing the risk of comorbidities associated with BD (Bauer et al., 2016). In our study, lifestyle factors, including diet, sleep quality, smoking and alcohol consumption, were associated with clinical symptoms in BD patients. The intake of whole grains, seafood, dairy products and nuts was found
ACCEPTED MANUSCRIPT
to be negatively associated with the severity of depressive symptoms. Conversely, soft drink, tobacco, and alcohol consumption was positively associated with the severity of the disease. This relationship was likely because whole grains and dairy products are important components of a healthy diet (Jacka et al., 2011b), which might play a
CR IP T
role in the decrease in the severity of depressive moods (Jacka et al., 2011a). On the other hand, seafood and nuts are rich in omega-3 fatty acids, nutrients that have been proven to have positive effects on depressive symptoms in bipolar disorder patients
AN US
(Montgomery and Richardson, 2008). In contrast, soft drink and alcohol consumption might represent an unhealthy dietary pattern and have been shown to be associated with increased odds of depressive symptoms (Jacka et al., 2010; Jacka et al., 2011b).
M
These findings suggest that lifestyle, particularly diet patterns, may play a role in the clinical symptoms of BD patients, highlighting the importance of lifestyle
ED
management in the improvement of clinical symptoms and outcomes of BD.
PT
To further analyze the association between depressive symptoms and lifestyle in BD patients, we conducted multiple linear regression and binary logistic regression
CE
analyses to adjust for potential confounding factors and examine the extent to which
AC
the lifestyle factors could affect the HAMD-17 score. The results suggest that the intake of whole grains and dairy products was inversely associated with the HAMD-17 score. These two food groups were also associated with a higher probability of clinical remission of depressive symptoms. Although the mechanisms underlying those associations are still unclear, evidence from several studies have suggested that in BD patients, emotional dysregulation/impulsivity is associated with
ACCEPTED MANUSCRIPT
poor nutritional behaviors, such as stress-induced eating (Martin et al., 2016), disinhibition and the perception of hunger (Bernstein et al., 2015). This may lead to a difficulty in eating healthy foods, including whole grains and dairy products, for this population. In addition, given that whole grains and dairy products are an important
CR IP T
source of prebiotics, including lignans, polyphenols, dietary fiber, and lactose (Jost et al., 2015; Keim and Martin, 2014), it might be suspected that these foods may exert depression-reducing effects by modulating the gut microbiome. Recent evidence from
AN US
both animal and human experiments indicates that whole grain interventions could alter the intestinal microbiota composition and parameters of health (Vanegas et al., 2017; Xia et al., 2017). Dairy interventions also led to a significant alteration in the
M
intestinal bacterial community and gut microbial activity in humans (Odamaki et al., 2016; Unno et al., 2015). The gut microbiota is important for mental health (Schmidt,
ED
2015), and the impairment of the gut microbiota may play a key role in the
PT
development of mood disorders, such as major disorders and bipolar disorder (Mangiola et al., 2016). Therefore, the involvement of healthy foods, such as whole
CE
grain and dairy products, in the severity of depressive symptoms might be partly due
AC
to their effects on the gut-brain axis (Lang et al., 2015). 4.1 Limitations Some limitations of the present study should also be considered. First, the
cross-sectional study design precludes any determinations regarding the direction of the relationship between lifestyle and clinical symptoms in BD patients. The absence of a healthy comparison group in our study makes it difficult to reach definitive
ACCEPTED MANUSCRIPT
conclusions. Additionally, the current sample was composed of an entirely Chinese population. Therefore, the results derived from the present study might not apply to populations in Western countries or populations with other dietary patterns. As a result, the extrapolation of the conclusions to other population should be done with caution.
CR IP T
Although some covariates were adjusted for during the data analysis, some residual confounding was possible. Furthermore, 74.8% patients were medicated, and the potential confounding effects of psychotropic medications should be not ruled out. We
AN US
included the information on psychotropic medications evaluate the medication confounding effects by correlating clinical symptoms to the amount of medication taken, measured in mg of drugs. In addition, relying on the memory or cognitive
M
ability of patients for the FFQ method did not allow us to estimate nutrient intake precisely in the present study. Thus, a longer follow-up survey is necessary to
ED
elucidate the relationship between lifestyle factors and clinical symptoms in BD
PT
patients.
CE
5. Conclusion
AC
In spite of the limitations, the results from our study suggest that the lifestyle factors, particularly diet, are associated with clinical symptoms in BD. The associations between the consumption of specific foods and depressive symptom may offer some useful information for the further understanding of the role of lifestyle factors in the development of BD. Well-designed cohort and intervention studies are needed to identify whether or not the lifestyle of those with BD is purely a function of
ACCEPTED MANUSCRIPT
disease status and symptomatology or if lifestyle contributes in any way to the pathophysiology of bipolar illness.
Conflicts of interest
CR IP T
None.
Role of the funding source
AN US
This work was supported by the Research Program for Young Scientists of Shanghai Municipal Bureau of health (grant No.201344127, No.20174Y0223) and College Level International Cooperation Projects of Shanghai Mental Health Center
ED
Contributors
M
(grant No. 2013-YJGJ-06).
PT
The contribution of each of the authors is as follows: Jia Huang, conception and design of the study, interpretation of the data; Chen Mei Yuan, drafting the manuscript;
CE
Xian Rong Xu, analysis and interpretation of the data; Yong Wang, Wu Hong, Zuo
AC
Wei Wang, You-song Su, Ying Yan Hu, Lan Cao and Yu Wang, recruitment of patients and collection of the data; Jun Chen and Yi Ru Fang, conception and design of the study and substantial revision of the manuscript. We confirm that the manuscript was read and approved by all named authors, and no other people satisfied the criteria for authorship.
ACCEPTED MANUSCRIPT
Acknowledgements The authors would like to thank the Department of Clinical Laboratory of Shanghai Mental Health Center for their technical support. We also would like to acknowledge
AN US
CR IP T
the help and support of all of patients who participated in this study.
References
Andreasen, N.C., Rice, J., Endicott, J., Reich, T., Coryell, W., 1986. The family history approach to diagnosis. How useful is it? Arch Gen Psychiatry 43 (5), 421-429.
Bauer, I.E., Galvez, J.F., Hamilton, J.E., Balanza-Martinez, V., Zunta-Soares, G.B., Soares, J.C., et al., 2016. Lifestyle interventions targeting dietary habits and exercise in bipolar disorder: A
M
systematic review. J Psychiatr Res 74, 1-7.
Belmaker, R.H., 2004. Bipolar disorder. N Engl J Med 351 (5), 476-486. Bernstein, E.E., Nierenberg, A.A., Deckersbach, T., Sylvia, L.G., 2015. Eating behavior and obesity in
ED
bipolar disorder. Aust N Z J Psychiatry 49 (6), 566-572. Cerimele, J.M., Katon, W.J., 2013. Associations between health risk behaviors and symptoms of schizophrenia and bipolar disorder: a systematic review. Gen Hosp Psychiatry 35 (1), 16-22.
PT
De Hert, M., Cohen, D., Bobes, J., Cetkovich-Bakmas, M., Leucht, S., Ndetei, D.M., et al., 2011. Physical illness in patients with severe mental disorders. II. Barriers to care, monitoring and
CE
treatment guidelines, plus recommendations at the system and individual level. World Psychiatry 10 (2), 138-151.
Elmslie, J.L., Mann, J.I., Silverstone, J.T., Williams, S.M., Romans, S.E., 2001. Determinants of
AC
overweight and obesity in patients with bipolar disorder. J Clin Psychiatry 62 (6), 486-491; quiz 492-483.
Ferrari, A.J., Baxter, A.J., Whiteford, H.A., 2011. A systematic review of the global distribution and availability of prevalence data for bipolar disorder. J Affect Disord 134 (1-3), 1-13.
Jacka, F.N., Kremer, P.J., Leslie, E.R., Berk, M., Patton, G.C., Toumbourou, J.W., et al., 2010. Associations between diet quality and depressed mood in adolescents: results from the Australian Healthy Neighbourhoods Study. Aust N Z J Psychiatry 44 (5), 435-442. Jacka, F.N., Mykletun, A., Berk, M., Bjelland, I., Tell, G.S., 2011a. The association between habitual diet quality and the common mental disorders in community-dwelling adults: the Hordaland Health study. Psychosom Med 73 (6), 483-490. Jacka, F.N., Pasco, J.A., Mykletun, A., Williams, L.J., Nicholson, G.C., Kotowicz, M.A., et al., 2011b.
ACCEPTED MANUSCRIPT
Diet quality in bipolar disorder in a population-based sample of women. J Affect Disord 129 (1-3), 332-337. Jost, T., Lacroix, C., Braegger, C., Chassard, C., 2015. Impact of human milk bacteria and oligosaccharides on neonatal gut microbiota establishment and gut health. Nutr Rev 73 (7), 426-437. Keim, N.L., Martin, R.J., 2014. Dietary whole grain-microbiota interactions: insights into mechanisms for human health. Adv Nutr 5 (5), 556-557. Kessler, R.C., Petukhova, M., Sampson, N.A., Zaslavsky, A.M., Wittchen, H.U., 2012. Twelve-month States. Int J Methods Psychiatr Res 21 (3), 169-184.
CR IP T
and lifetime prevalence and lifetime morbid risk of anxiety and mood disorders in the United Lang, U.E., Beglinger, C., Schweinfurth, N., Walter, M., Borgwardt, S., 2015. Nutritional aspects of depression. Cell Physiol Biochem 37 (3), 1029-1043.
Mangiola, F., Ianiro, G., Franceschi, F., Fagiuoli, S., Gasbarrini, G., Gasbarrini, A., 2016. Gut microbiota in autism and mood disorders. World J Gastroenterol 22 (1), 361-368.
Martin, K., Woo, J., Timmins, V., Collins, J., Islam, A., Newton, D., et al., 2016. Binge eating and
AN US
emotional eating behaviors among adolescents and young adults with bipolar disorder. J Affect Disord 195, 88-95.
Merikangas, K.R., Akiskal, H.S., Angst, J., Greenberg, P.E., Hirschfeld, R.M., Petukhova, M., et al., 2007. Lifetime and 12-month prevalence of bipolar spectrum disorder in the National Comorbidity Survey replication. Arch Gen Psychiatry 64 (5), 543-552.
Montgomery, P., Richardson, A.J., 2008. Omega-3 fatty acids for bipolar disorder. Cochrane Database Syst Rev(2), CD005169.
M
Noaghiul, S., Hibbeln, J.R., 2003. Cross-national comparisons of seafood consumption and rates of bipolar disorders. Am J Psychiatry 160 (12), 2222-2227. Noguchi, R., Hiraoka, M., Watanabe, Y., Kagawa, Y., 2013. Relationship between dietary patterns and
ED
depressive symptoms: difference by gender, and unipolar and bipolar depression. J Nutr Sci Vitaminol (Tokyo) 59 (2), 115-122.
Odamaki, T., Kato, K., Sugahara, H., Xiao, J.Z., Abe, F., Benno, Y., 2016. Effect of probiotic yoghurt
PT
on animal-based diet-induced change in gut microbiota: an open, randomised, parallel-group study. Benef Microbes 7 (4), 473-484.
CE
Patel, V., Chisholm, D., Parikh, R., Charlson, F.J., Degenhardt, L., Dua, T., et al., 2016. Addressing the burden of mental, neurological, and substance use disorders: key messages from Disease Control Priorities, 3rd edition. Lancet 387 (10028), 1672-1685.
AC
Phillips, M.R., Zhang, J., Shi, Q., Song, Z., Ding, Z., Pang, S., et al., 2009. Prevalence, treatment, and associated disability of mental disorders in four provinces in China during 2001-05: an epidemiological survey. Lancet 373 (9680), 2041-2053.
Schmidt, C., 2015. Mental health: thinking from the gut. Nature 518 (7540), S12-15. Sylvia, L.G., Friedman, E.S., Kocsis, J.H., Bernstein, E.E., Brody, B.D., Kinrys, G., et al., 2013. Association of exercise with quality of life and mood symptoms in a comparative effectiveness study of bipolar disorder. J Affect Disord 151 (2), 722-727. Unno, T., Choi, J.H., Hur, H.G., Sadowsky, M.J., Ahn, Y.T., Huh, C.S., et al., 2015. Changes in human gut microbiota influenced by probiotic fermented milk ingestion. J Dairy Sci 98 (6), 3568-3576. Vanegas, S.M., Meydani, M., Barnett, J.B., Goldin, B., Kane, A., Rasmussen, H., et al., 2017. Substituting whole grains for refined grains in a 6-wk randomized trial has a modest effect on gut
ACCEPTED MANUSCRIPT
microbiota and immune and inflammatory markers of healthy adults. Am J Clin Nutr 105 (3), 635-650. Xia, X., Li, G., Ding, Y., Ren, T., Zheng, J., Kan, J., 2017. Effect of Whole Grain Qingke (Tibetan Hordeum vulgare L. Zangqing 320) on the Serum Lipid Levels and Intestinal Microbiota of Rats under High-Fat Diet. J Agric Food Chem 65 (13), 2686-2693. Yang, Y.X., Wang, G.Y., Pan, X.C., 2004. The Food Composition of China,2004. Peking University Medical Press, Beijing Zhang, L., Cao, X.L., Wang, S.B., Zheng, W., Ungvari, G.S., Ng, C.H., et al., 2017. The prevalence of bipolar disorder in China: A meta-analysis. J Affect Disord 207, 413-421.
CR IP T
Zhang, W., Li, Q., Shi, L., Lu, K., Shang, Q., Yao, L., et al., 2009. [Investigation of dietary intake of
AC
CE
PT
ED
M
AN US
cadmium in certain polluted area of south in China]. Wei Sheng Yan Jiu 38 (5), 552-554, 557.
ACCEPTED MANUSCRIPT
Table 1. Demographic and illness characteristics of BD patients Variables 36.6±15.7 36 (32.1) 68 (60.2)
AC
CE
PT
ED
M
44 (38.9) 57 (56.5) 6 (5.3) 6 (5.3)
CR IP T
AN US
Demographics Age at screen Male, n (%) Education:≥college or technical degree, n(%) Marital status, n(%) Single Married Divorced Widow NO.of individual in the household ≤3 >3 Per capita monthly income of family ≤ USD $750 >USD $750 Clinical characteristics, mean±SD Inpatient, n(%) Age at first onset(years), (n=111) Duration of illness(years) Duration of untreated illness(years) Bipolar disorder type I, n(%) Bipolar disorder type II, n(%) Baseline HAMD-17 Total Score Clinical remission(HAMD-17 score <7 ), n(%) Outpatients, n(%) Inpatients, n(%) Baseline HAMA Total Score Family history of psychiatric disorders, n(%) Number of depressive episode Number of mania episode Psychotropic medications(n=111), n(%) Mood-stabilizer Lithium Sodium vaporate Antipsychotic Quetiapine Olanzapine Aripiprazole Antidepressant Citalopram Venlafaxine; Fluoxetine
90 (79.6) 23 (20.4)
77 (68.1) 36 (31.9)
65 (57.5) 29.9±14.6 7.8±8.9 2.1±1.1 84 (74.3) 25 (25.7) 13.3±7.3 58 (51.3) 40 (35.4) 18 (15.9) 9.5±6.0 15 (13.3) 2.8±3.2 1.6±2.2 83 (74.8) 14 (12.6) 18 (16.2) 37 (33.3) 13 (11.7) 5 (4.5) 11 (9.9) 12 (10.8) 9 (8.1)
ACCEPTED MANUSCRIPT
3 (2.7) 13 (11.7) 13 (11.7) 5 (4.5) 6 (5.4)
AN US
CR IP T
10 (9.0) 6 (5.4) 86.5 (0,349)*
PT
ED
M
Fluvoxamine Mirtazapine Sertraline Paroxetine Duloxetine Sedative hypnotic agents Lorazepam Alprazolam Psychotropic medications, mg per day Metabolic parameters BMI (Kg/m2) Obesity, n(%) Waist circumvent(cm) Hip circumvent(cm) Waist/Hip ratio Body fat rate(%) Fasting blood glucose(μmmol/L) HbA1c, % Serum insulin (pmol/L) Triglyceride (mmol/L) Total cholesterol (mmol/L) LDL-cholesterol (mmol/L) HDL- cholesterol (mmol/L) Alcohol consumption, n(%) Current smoker, n(%) C-reactive protein (mmol/L) Homocysteine (mmol/L) Folic acid (mmol/L) Vitamin B12 (mmol/L)
AC
CE
*, expressed as median (interquartile)
22.2±3.2 17 (15.0) 84.4±12.8 94.1±8.8 0.9±0.1 27.5±6.8 5.3±1.8 5.7±1.4 107.7±101.7 1.2±0.9 4.7±1.3 2.8±1.0 1.4±0.3 13 (11.6) 13 (11.6) 3.4±6.3 14.3±5.7 17.1±7.6 421.7±171.0
ACCEPTED MANUSCRIPT
Table 2. The dietary intake and lifestyle measures among BD patients Variables 1939.3±772.5 258.3±114.0 55.2±35.1 74.3±33.3
AC
CE
PT
ED
CR IP T
368.6±172.8 46.7 (16.3,82.2) 7.1 (0.2,28.6) 53.6 (8.9,100.0) 50.0 (21.42,100.0) 21.4 (7.1, 42.8) 0.6 (0.0,14.3) 156.6±173.9 60.5±35.6 100.0 (50.0, 150.0)
AN US
M
Energy and nutrients Energy(kcal/day) Carbohydrates(g/day) Fat(g/day) Protein(g/day ) Kinds of food Whole grain(g/day ) Unprocessed red meat(g/d) Processed meat(g/d) Dairy products(ml/day ) Eggs(g/day ) Bean and bean products(g/day ) Nuts(g/day) Vegetables(g/day ) Seafood(g/day ) Fruit(g/day ) Soft drink consumption(ml/d), n(%) 0 < 100 ≥100 Sugar(g/day) Alcohol(g/day), n(%) 0 < 50 ≥50 Exercise and sleep Physical activity No activity Activity not sufficient (<150min/week) Sufficient activity(≥150min/week) Sleep quality Total sleep time (minutes) Mean sleep onset time (minutes)
90(79.6) 19(16.8) 4(3.6) 8.8±6.3 100 (88.5) 7 (6.2) 6 (5.3)
34 (30.1) 35 (31.0) 44 (38.9) 532.2 ±77.3 20.3±2.5
ACCEPTED MANUSCRIPT
Table 3. Correlation of lifestyle characteristics and bipolar symptoms Number of depressive episode
Number of Manic episode
Age at first onset
Whole grain
-0.245*
-0.170
-0.079
0.164
Red meat
-0.069
0.014
0.213*
-0.241*
Seafood
-0.214*
0.178
0.095
0.090
Eggs
0.037
0.203*
-0.018
0.101
Dairy products
-0.229*
0.032
-0.046
0.055
Nuts
-0.086
0.061
-0.046
0.222*
Soft drink
-0.017
0.454**
0.701**
-0.310**
Sugar
0.115
-0.117
-0.134
-0.198*
Total sleep time
-0.336**
0.035
0.182
-0.106
Current smoking
0.019
0.114
0.252*
-0.172
Alcohol consumption
-0.042
0.459**
CE
PT
AN US
ED
M
*,P<0.05; **,P<0.01
AC
CR IP T
Variables
HAMD-17 total score
0.792**
-0.127
ACCEPTED MANUSCRIPT
Table 4.Multiple linear regression analysis of HAMD-17 score and lifestyle characteristics in BD$ Variables
Unstandardized
Coefficients
Standardized coefficients
B
Std. Error
Beta
CR IP T
Independent variable: HAMD-17 score
P
t
-0.998 0.580
0.236
0.256
2.458
0.029 0.01 6
Whole grain
-0.012
0.005
-0.305
-2.338
0.02
Red meat
-0.036
0.017
0.324
-2.141
0.115
0.052
0.311 0.028
Vegetables
0.008
-0.019
CE
Fruits
AC
Exercise time (min) Sleep time （min）
0.184
PT
Soft drink
Seafood
0.011
-0.333
M
Sugar
-0.021
ED
Dairy products Nuts
0.448
AN US
Number of manic episode BMI
0.001 -0.007
-0.228 0.263 0.205
0.014
0.441
0.004
0.223
0.011
-0.177
-2.225
3 0.03 7 -1.950
0.05 5
2.21 6 1.68 6 1.99 0 1.80 4 -1.761
0.03 1 0.09 8 0.05 2 0.07 7 0.08 1
0.006
0.023
0.198
0.84 3
0.009
-0.083
-0.832
0.40 8
-0.019
0.009
-0.225
-2.128
0.03 7
$,controlled by age, gender, number of depressive episode, age at first onset, income of family, education level .
Table 5 The association between whole grain, dairy products intake and remission by Binary logistic analysis
ACCEPTED MANUSCRIPT
Variables
Unadjusted OR(95%CI)
Adjusted OR (95%CI)$
P
<200
Ref.
200-400
10.37 (1.242,86.62)
10.74（1.19,55.51)
0.035
>400
19.09 (2.155,169.09)
14.25 (1.31,80.23)
0.029
Independent variable: Clinical remission (0=HAMD-17 score≥17; 1= HAMD-17< score 17)
P for trend
CR IP T
Whole grain intake (g/day)
0.003
Dairy products intake(ml/day) Ref.
<100
1.08 (0.36,3.29)
2.35 (0.56,9.82)
0.119
>100
4.17 (1.41,20.71)
9.93 (2.28,43.16)
0.002
AN US
0
P for trend
0.003
AC
CE
PT
ED
M
$,adjusted by age, gender, number of depressive episode, age at first onset ,BMI, total energy, intake of fruits, vegetables, seafood, exercise time, smoking and alcohol consumption, income of family , education level.
The relationship between lifestyle factors and clinical symptoms of bipolar disorder patients in a Chinese population Jia Huang , Cheng Mei Yuan , Xian Rong Xu , Yong Wang , Wu Hong , Zuo Wei Wang , You song Su , Ying Yan Hu , Lan Cao , Yu Wang , Jun Chen , Yi Ru Fang PII: DOI: Reference:
S0165-1781(17)30273-1 10.1016/j.psychres.2018.04.059 PSY 11390
To appear in:
Psychiatry Research
Received date: Revised date: Accepted date:
20 February 2017 7 March 2018 22 April 2018
Please cite this article as: Jia Huang , Cheng Mei Yuan , Xian Rong Xu , Yong Wang , Wu Hong , Zuo Wei Wang , You song Su , Ying Yan Hu , Lan Cao , Yu Wang , Jun Chen , Yi Ru Fang , The relationship between lifestyle factors and clinical symptoms of bipolar disorder patients in a Chinese population, Psychiatry Research (2018), doi: 10.1016/j.psychres.2018.04.059
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Highlights
We explored the associations between Life style factors and clinical symptoms in BD patients with depressive symptoms, using the
17 item Hamilton
Rating Scale for Depression (HAMD-17) to assess symptom severity. The consumption of whole grain and dairy products showed an independent
CR IP T
negative correlation with the clinical symptoms and HAMD-17 total scores in BD patients.
Our results highlight the importance of Life style management in the
AN US
AC
CE
PT
ED
M
improvement of clinical symptoms in BD patients.
ACCEPTED MANUSCRIPT
The relationship between lifestyle factors and clinical symptoms of bipolar disorder patients in a Chinese population
CR IP T
Running title: Bipolar disorder symptoms and lifestyle
Authors: Jia Huang a, Cheng Mei Yuan b, Xian Rong Xu c, Yong Wang a, Wu Hong a, Zuo Wei Wang d, You song Su a, Ying Yan Hu a, Lan Cao a, Yu Wang d, Jun Chen a*, Yi
a
AN US
Ru Fang a, e, f,**
Department of Mood Disorder, Shanghai Mental Health Center, Shanghai Jiao Tong
b
M
University School of Medicine, 600 Wan Ping Nan Road, Shanghai 200030, PR China First Department of General Psychiatry, Shanghai Mental Health Center, Shanghai
Department of Prevention Medicine, School of Medicine, Hangzhou Normal
PT
c
ED
Jiao Tong University School of Medicine, Shanghai 200030, PR China
University, 16 Xue Lin Road, Hangzhou, PR China Division of Mood Disorders, Hongkou District Mental Health Center of Shanghai,
CE
d
AC
Shanghai 200083, PR China e
State Key Laboratory of Neuroscience, Shanghai Institute for Biological Sciences,
CAS, 320 Yue Yang Road Shanghai, 200031 PR China f
Shanghai Key Laboratory of Psychotic Disorders, 600 Wan Ping Nan Road,
Shanghai 200030, PR China
ACCEPTED MANUSCRIPT
*Corresponding author. Tel.: +86 021 34773367; Fax: +86 021 34773367 **Corresponding author. Tel.: +86 21 64387250x73529; Fax: +86 21 64387986 E-mail addresses: [email protected] (Jun Chen); [email protected]
CR IP T
(Yi Ru Fang)
AN US
Abstract
There is evidence that bipolar disorder (BD) patients with an unhealthy lifestyle have a worse course of illness. This study was designed to examine the extent to which
M
lifestyle could influence the severity of clinical symptoms associated with BD. A total of 113 BD patients were recruited in this study. The lifestyle information including
ED
data on dietary patterns, physical activity, and sleep quality were collected using a
PT
self-rated questionnaire. The results showed that the consumption of whole grain, seafood, and dairy products were significantly negatively correlated with the 17-item
CE
Hamilton Rating Scale for Depression (HAMD-17) total score. The consumption of
AC
sugar, soft drinks, and alcohol as well as being a current smoker were positively correlated with the severity of clinical symptoms. Multiple linear regression and binary logistic regression analyses demonstrated an independent negative correlation between both whole grain and dairy product consumption with the HAMD-17 score. The results from the current study suggested that lifestyle factors, especially dietary patterns, might be associated with clinical symptoms of BD. The association between
ACCEPTED MANUSCRIPT
the consumption of specific foods and severity of depressive symptoms may offer some useful information and further understanding of the role of lifestyle factors in
CR IP T
the development of BD.
Keywords: bipolar disorder; symptom assessment; depression; lifestyle; diet
AN US
1. Introduction
Bipolar disorder (BD) is a severe and recurrent psychiatric illness associated with substantial functional impairments and premature mortality (Belmaker, 2004).
M
Globally, the overall point-prevalence of BD is estimated to be 0.741% (Ferrari et al., 2011) and BD accounts for nearly 1,320,000 excess deaths each year (Patel et al.,
ED
2016). Data from a nationally representative survey in the U.S. demonstrated that the
PT
BD affected approximately 4.4% of people in United States (Merikangas et al., 2007) and incurred disproportionately higher costs for both patients and the mental health
CE
care system than most other mental illnesses do (Kessler et al., 2012).In China, the
AC
estimated point, 12-month and lifetime prevalence of BD in China was 0.09% (95% confidence interval [CI]: 0.06-0.12%), 0.17% (95% CI: 0.10-0.29%) and 0.11% (95% CI: 0.07-0.17%), respectively (Zhang et al., 2017). These results suggest that more than 1.37 million adults in the country may have BD, and even worse, most of them never receive any type of professional interventions for their conditions (Phillips et al., 2009). Therefore, it is important for China to identify the factors involved in the
ACCEPTED MANUSCRIPT
initiation and development of BD. BD is a complex disease, and its etiology has not been fully elucidated. Although genetics plays a significant role in the susceptibility to this illness, evidence from several studies suggests that lifestyle factors, such as diet patterns, exercise and sleep
CR IP T
quality also appear to be important (De Hert et al., 2011). Sylvia et al. found that less frequent weekly exercise was associated with higher BMI, more time spent depressed, more depressive symptoms, and lower quality of life and functioning in BD patients
AN US
(Sylvia et al., 2013). Compared to those with no psychopathology, BD patients had a lower quality of life and a higher glycemic load in their diet (Jacka et al., 2011b). Elmslie et al. found that BD patients consumed more total carbohydrates, sucrose,
M
nonalcoholic beverages, sweetened drinks, cakes and sweets (Elmslie et al., 2001). The authors also found that female BD patients, but not male patients, also had greater
ED
total energy intake. Noaghiul et al. demonstrated that high rates of seafood
PT
consumption were associated with lower lifetime prevalence rates of bipolar I disorder, bipolar II disorder, and bipolar spectrum disorder (Noaghiul and Hibbeln, 2003).
CE
Noguchi et al. found that in BD patients, plant-based food and fish product
AC
consumption patterns were inversely related to physical and psychiatric symptoms, and in men, this pattern demonstrated an inverse relationship with psychiatric symptoms (Noguchi et al., 2013). Lifestyle factors, such as tobacco use and weight gain/obesity, were associated with an increased severity of BD symptoms (Cerimele and Katon, 2013). Despite these promising data on the importance for lifestyle factors in bipolar disorder, little is known about the influence of lifestyle on the remission of
ACCEPTED MANUSCRIPT
the disease and how lifestyle factors might affect bipolar specific mood symptoms. To address this issue, we examined the association between lifestyle factors and clinical characters in BD patients. We hypothesized that a higher quality lifestyle would be associated with a decreased severity of clinical symptoms and improved
CR IP T
odds of clinical remission of patients with BD.
2. Methods
AN US
2.1 Patient population
Participants were recruited from the Shanghai Mental Health Center between July 2015 and May 2016. Subjects in this analysis met the following inclusion criteria: (1)
M
18-65 years of age; (2) inpatients and outpatients meeting DSM-IV criteria for bipolar I or II disorder with a current major depressive episode; and (3) willingness and
ED
ability to perform the necessary clinical tests. Subjects were excluded if they met any
PT
of the following criteria: (1) were unable to provide informed consent (assessed by using a short questionnaire asking key questions about the study); (2) were with any
CE
history or presence of serious physical disease, endocrine or autoimmune disease; (3)
AC
were pregnant or lactating women; (4) had an infection or fever within the last 2 months; or (5) had received any prior systemic hormonal therapy or cytokine therapy. All of the patients with bipolar disorder in this study were diagnosed by at least two chief physicians. The study protocol and methodologies were reviewed and approved by the Institutional Review Board of Shanghai Mental Health Center, and all protocols related to human experiments were conducted in accordance with the
ACCEPTED MANUSCRIPT
Declaration of Helsinki. We ensured that all participants were given an adequate understanding of the study, and written informed consent was obtained from all individuals prior to their inclusion in the study.
CR IP T
2.2 Study measures Demographic data on age, gender, marital status, educational level, income, body mass index (BMI), the duration of major depressive episodes, frequency of exercise,
AN US
history of smoking, and alcohol and other drug use were collected from the participants. A self-administered physical activity questionnaire was used to assess physical activity levels in patients. The patients were asked about the frequency and
M
time spent walking for transport and walking for recreation and leisure as well as about moderate- and vigorous-intensity activity over the previous year. The family
ED
history of psychiatric disorders was collected according to the Family History
PT
Research Diagnostic Criteria (HF-DRC), which has been reported to have a specificity of 96% and sensitivity of 52% (Andreasen et al., 1986). The Hamilton
CE
Rating Scale for Depression-17 (HAMD-17) was used to assess the severity of
AC
depressive symptoms. Clinical remission in patients with bipolar disorder was defined as having a HAMD-17 total score ≤7, which is the current widely accepted cut-off value. The Pittsburgh Sleep Quality Index (PSQI) was used to measure the quality of sleep. All assessments were conducted in the afternoon to avoid the effects of diurnal variation in cortisol levels and mood. All assessments were performed by the same trained rater.
ACCEPTED MANUSCRIPT
2.3 Dietary survey Participants were visited at hospital by specifically trained dieticians. A self-administered semi-quantitative food frequency questionnaire (FFQ) was used to measure the habitual consumption items from 13 different food groups (fruits and
CR IP T
vegetables, whole grains, dairy products, red meat, poultry, fish, eggs, nuts, cooking oils, milk, soft drinks, alcoholic beverages, and fruit and vegetable juices), for a total of 41 items. The questionnaire was adapted from the questionnaire designed by the
AN US
Chinese Nutrition Society (CNS) in 2010 (Zhang et al., 2009). Before administering the diet survey, the patients were given oral and written information of how to carry out the survey. The survey collected information including the frequency of
M
consumption (daily, weekly or monthly) and the quantity of the foods consumed. Patients were asked about their habitual consumption of the listed foods within the
ED
past 1 year. The intake of energy, protein, fat, and carbohydrates were calculated
et al., 2004).
PT
based on the Standardized Tables of Food Composition of China (2004 edition) (Yang
CE
2.4 Statistical analysis
AC
All of the data were analyzed using the software SPSS 19.0 (SPSS, Chicago, IL, USA). Continuous variables such as age, BMI, serum lipids and HAMD-17 total score are presented as the means ± SD. Categorical variables, such as gender, education level, marital status, and family history of psychiatric disorders, are presented as frequencies and percentages. Spearman’s correlation analyses were used to determine the relationship between clinical characters and lifestyle factors in
ACCEPTED MANUSCRIPT
bipolar disorder patients. Multiple linear regression models were used to examine the relationships between the HAMD-17 total score and lifestyle factors after adjusting for potential confounders, which included age, gender, number of depressive episodes, and age at first onset. A backward binary logistic regression analysis was performed to
CR IP T
independently evaluate the effects of lifestyle factors on the clinical remission of the disease, using “remission status” as a dependent variable and lifestyle factors as independent variables. The continuous variables, including the intake of different
AN US
kinds of foods, sleep time, and age, were converted into categorical variables to more accurately evaluate the associations between lifestyle factors and the clinical
M
remission of disease. For all analyses, the significance level was set as P< 0.05.
3. Results
ED
3.1 Patient characteristics
PT
We included 113 patients with bipolar disorder in this study; 65 (57.5%) subjects were inpatients, and 48 (42.5%) subjects were outpatients. The patients had a mean
CE
age of 36.6±15.7; 36 (32%) were male. The mean age at onset of the first mood
AC
episode was 29.9±14.6, with a mean duration of illness of 7.78±8.93 years. The mean age at first contact with a mental health professional was 29.4±15.0 years, whereas the mean age at the correct diagnosis and treatment was 31.5±13.9 years; the mean duration of untreated illness was 2.1±1.1 years. Based on the HAMD-17 scores, 58 (51.3%) patients were categorized as being in clinical remission (values below 7). A range of mood-stabilizers, antipsychotics, antidepressants and sedative hypnotic agent
ACCEPTED MANUSCRIPT
medications were prescribed to the patients. Overall, 74.8% of total patients received at least one kind of drug. All sociodemographic and clinical variables are presented in Table 1. The lifestyle characteristics of the patients are shown in Table 2. The mean caloric
CR IP T
intake was 1939.3±772.5 kcal/d. The daily intakes of carbohydrates, fat, and protein were 258.3±114.0 g, 55.2±35.1 g and 74.3±33.3 g, respectively. The mean or median intake of 13 food items, physical activity and sleep quality are also shown in Table 2.
AN US
3.2 Clinical characteristics and lifestyle
Simple correlation analysis indicated a significant association between lifestyle and clinical characters in BD patients (Table 3). The HAMD-17 score was negatively
M
associated with whole grains, seafood and dairy product consumption and total sleep time. The number of depressive episodes was positively associated with egg,
ED
vegetable, soft drink and alcoholic consumption. The number of manic episodes was
PT
positively associated with red meat, soft drink and alcohol consumption as well as with smoking. The age at first onset was negatively associated with red meat, soft
CE
drink, and sugar consumption and positively associated with nut consumption.
AC
Multiple linear regression analysis supported the association between lifestyle and the HAMD-17 score after controlling for potential confounding factors (Table 4). The results showed that whole grain and red meat consumption and total sleep time were negatively associated with the baseline HAMD-17 total score (P<0.05). The intake of dairy products was also found to be negatively associated with the HAMD-17 score, though this relationship was not statistically significant (P=0.055).
ACCEPTED MANUSCRIPT
We used a logistic regression model to assess the association between lifestyle factors and clinical remission of BD after controlling for potential confounding factors, including age, gender, number of depressive episodes, age at first onset of BD, BMI, total energy, intakes of fruits, vegetables, seafood, amount of physical activity,
CR IP T
smoking, alcohol consumption, family income, and education level (Table 5). The results showed that the intake of whole grains and dairy products was associated with the clinical remission of disease after controlling for potential confounding factors.
AN US
Compared to the reference level (<200 g/d), the intake of >400 g/d whole grain was associated with a statistically higher probability of clinical remission of the disease (OR=14.25, 95% CI: 1.31, 80.23). Compared to those who did not consume dairy
M
products, consumption of >100 g/ml of dairy products was associated with a statistically higher probability of clinical remission of the disease after adjusting for
PT
ED
potential confounding factors (OR=9.93, 95% CI: 2.28, 43.36).
4. Discussion
CE
Several studies have explored the association between clinical symptoms and
AC
lifestyle in BD patients (Jacka et al., 2011b; Noaghiul and Hibbeln, 2003; Noguchi et al., 2013; Sylvia et al., 2013). Lifestyle interventions for BD patients targeting diet and exercise have shown promise in reducing the risk of comorbidities associated with BD (Bauer et al., 2016). In our study, lifestyle factors, including diet, sleep quality, smoking and alcohol consumption, were associated with clinical symptoms in BD patients. The intake of whole grains, seafood, dairy products and nuts was found
ACCEPTED MANUSCRIPT
to be negatively associated with the severity of depressive symptoms. Conversely, soft drink, tobacco, and alcohol consumption was positively associated with the severity of the disease. This relationship was likely because whole grains and dairy products are important components of a healthy diet (Jacka et al., 2011b), which might play a
CR IP T
role in the decrease in the severity of depressive moods (Jacka et al., 2011a). On the other hand, seafood and nuts are rich in omega-3 fatty acids, nutrients that have been proven to have positive effects on depressive symptoms in bipolar disorder patients
AN US
(Montgomery and Richardson, 2008). In contrast, soft drink and alcohol consumption might represent an unhealthy dietary pattern and have been shown to be associated with increased odds of depressive symptoms (Jacka et al., 2010; Jacka et al., 2011b).
M
These findings suggest that lifestyle, particularly diet patterns, may play a role in the clinical symptoms of BD patients, highlighting the importance of lifestyle
ED
management in the improvement of clinical symptoms and outcomes of BD.
PT
To further analyze the association between depressive symptoms and lifestyle in BD patients, we conducted multiple linear regression and binary logistic regression
CE
analyses to adjust for potential confounding factors and examine the extent to which
AC
the lifestyle factors could affect the HAMD-17 score. The results suggest that the intake of whole grains and dairy products was inversely associated with the HAMD-17 score. These two food groups were also associated with a higher probability of clinical remission of depressive symptoms. Although the mechanisms underlying those associations are still unclear, evidence from several studies have suggested that in BD patients, emotional dysregulation/impulsivity is associated with
ACCEPTED MANUSCRIPT
poor nutritional behaviors, such as stress-induced eating (Martin et al., 2016), disinhibition and the perception of hunger (Bernstein et al., 2015). This may lead to a difficulty in eating healthy foods, including whole grains and dairy products, for this population. In addition, given that whole grains and dairy products are an important
CR IP T
source of prebiotics, including lignans, polyphenols, dietary fiber, and lactose (Jost et al., 2015; Keim and Martin, 2014), it might be suspected that these foods may exert depression-reducing effects by modulating the gut microbiome. Recent evidence from
AN US
both animal and human experiments indicates that whole grain interventions could alter the intestinal microbiota composition and parameters of health (Vanegas et al., 2017; Xia et al., 2017). Dairy interventions also led to a significant alteration in the
M
intestinal bacterial community and gut microbial activity in humans (Odamaki et al., 2016; Unno et al., 2015). The gut microbiota is important for mental health (Schmidt,
ED
2015), and the impairment of the gut microbiota may play a key role in the
PT
development of mood disorders, such as major disorders and bipolar disorder (Mangiola et al., 2016). Therefore, the involvement of healthy foods, such as whole
CE
grain and dairy products, in the severity of depressive symptoms might be partly due
AC
to their effects on the gut-brain axis (Lang et al., 2015). 4.1 Limitations Some limitations of the present study should also be considered. First, the
cross-sectional study design precludes any determinations regarding the direction of the relationship between lifestyle and clinical symptoms in BD patients. The absence of a healthy comparison group in our study makes it difficult to reach definitive
ACCEPTED MANUSCRIPT
conclusions. Additionally, the current sample was composed of an entirely Chinese population. Therefore, the results derived from the present study might not apply to populations in Western countries or populations with other dietary patterns. As a result, the extrapolation of the conclusions to other population should be done with caution.
CR IP T
Although some covariates were adjusted for during the data analysis, some residual confounding was possible. Furthermore, 74.8% patients were medicated, and the potential confounding effects of psychotropic medications should be not ruled out. We
AN US
included the information on psychotropic medications evaluate the medication confounding effects by correlating clinical symptoms to the amount of medication taken, measured in mg of drugs. In addition, relying on the memory or cognitive
M
ability of patients for the FFQ method did not allow us to estimate nutrient intake precisely in the present study. Thus, a longer follow-up survey is necessary to
ED
elucidate the relationship between lifestyle factors and clinical symptoms in BD
PT
patients.
CE
5. Conclusion
AC
In spite of the limitations, the results from our study suggest that the lifestyle factors, particularly diet, are associated with clinical symptoms in BD. The associations between the consumption of specific foods and depressive symptom may offer some useful information for the further understanding of the role of lifestyle factors in the development of BD. Well-designed cohort and intervention studies are needed to identify whether or not the lifestyle of those with BD is purely a function of
ACCEPTED MANUSCRIPT
disease status and symptomatology or if lifestyle contributes in any way to the pathophysiology of bipolar illness.
Conflicts of interest
CR IP T
None.
Role of the funding source
AN US
This work was supported by the Research Program for Young Scientists of Shanghai Municipal Bureau of health (grant No.201344127, No.20174Y0223) and College Level International Cooperation Projects of Shanghai Mental Health Center
ED
Contributors
M
(grant No. 2013-YJGJ-06).
PT
The contribution of each of the authors is as follows: Jia Huang, conception and design of the study, interpretation of the data; Chen Mei Yuan, drafting the manuscript;
CE
Xian Rong Xu, analysis and interpretation of the data; Yong Wang, Wu Hong, Zuo
AC
Wei Wang, You-song Su, Ying Yan Hu, Lan Cao and Yu Wang, recruitment of patients and collection of the data; Jun Chen and Yi Ru Fang, conception and design of the study and substantial revision of the manuscript. We confirm that the manuscript was read and approved by all named authors, and no other people satisfied the criteria for authorship.
ACCEPTED MANUSCRIPT
Acknowledgements The authors would like to thank the Department of Clinical Laboratory of Shanghai Mental Health Center for their technical support. We also would like to acknowledge
AN US
CR IP T
the help and support of all of patients who participated in this study.
References
Andreasen, N.C., Rice, J., Endicott, J., Reich, T., Coryell, W., 1986. The family history approach to diagnosis. How useful is it? Arch Gen Psychiatry 43 (5), 421-429.
Bauer, I.E., Galvez, J.F., Hamilton, J.E., Balanza-Martinez, V., Zunta-Soares, G.B., Soares, J.C., et al., 2016. Lifestyle interventions targeting dietary habits and exercise in bipolar disorder: A
M
systematic review. J Psychiatr Res 74, 1-7.
Belmaker, R.H., 2004. Bipolar disorder. N Engl J Med 351 (5), 476-486. Bernstein, E.E., Nierenberg, A.A., Deckersbach, T., Sylvia, L.G., 2015. Eating behavior and obesity in
ED
bipolar disorder. Aust N Z J Psychiatry 49 (6), 566-572. Cerimele, J.M., Katon, W.J., 2013. Associations between health risk behaviors and symptoms of schizophrenia and bipolar disorder: a systematic review. Gen Hosp Psychiatry 35 (1), 16-22.
PT
De Hert, M., Cohen, D., Bobes, J., Cetkovich-Bakmas, M., Leucht, S., Ndetei, D.M., et al., 2011. Physical illness in patients with severe mental disorders. II. Barriers to care, monitoring and
CE
treatment guidelines, plus recommendations at the system and individual level. World Psychiatry 10 (2), 138-151.
Elmslie, J.L., Mann, J.I., Silverstone, J.T., Williams, S.M., Romans, S.E., 2001. Determinants of
AC
overweight and obesity in patients with bipolar disorder. J Clin Psychiatry 62 (6), 486-491; quiz 492-483.
Ferrari, A.J., Baxter, A.J., Whiteford, H.A., 2011. A systematic review of the global distribution and availability of prevalence data for bipolar disorder. J Affect Disord 134 (1-3), 1-13.
Jacka, F.N., Kremer, P.J., Leslie, E.R., Berk, M., Patton, G.C., Toumbourou, J.W., et al., 2010. Associations between diet quality and depressed mood in adolescents: results from the Australian Healthy Neighbourhoods Study. Aust N Z J Psychiatry 44 (5), 435-442. Jacka, F.N., Mykletun, A., Berk, M., Bjelland, I., Tell, G.S., 2011a. The association between habitual diet quality and the common mental disorders in community-dwelling adults: the Hordaland Health study. Psychosom Med 73 (6), 483-490. Jacka, F.N., Pasco, J.A., Mykletun, A., Williams, L.J., Nicholson, G.C., Kotowicz, M.A., et al., 2011b.
ACCEPTED MANUSCRIPT
Diet quality in bipolar disorder in a population-based sample of women. J Affect Disord 129 (1-3), 332-337. Jost, T., Lacroix, C., Braegger, C., Chassard, C., 2015. Impact of human milk bacteria and oligosaccharides on neonatal gut microbiota establishment and gut health. Nutr Rev 73 (7), 426-437. Keim, N.L., Martin, R.J., 2014. Dietary whole grain-microbiota interactions: insights into mechanisms for human health. Adv Nutr 5 (5), 556-557. Kessler, R.C., Petukhova, M., Sampson, N.A., Zaslavsky, A.M., Wittchen, H.U., 2012. Twelve-month States. Int J Methods Psychiatr Res 21 (3), 169-184.
CR IP T
and lifetime prevalence and lifetime morbid risk of anxiety and mood disorders in the United Lang, U.E., Beglinger, C., Schweinfurth, N., Walter, M., Borgwardt, S., 2015. Nutritional aspects of depression. Cell Physiol Biochem 37 (3), 1029-1043.
Mangiola, F., Ianiro, G., Franceschi, F., Fagiuoli, S., Gasbarrini, G., Gasbarrini, A., 2016. Gut microbiota in autism and mood disorders. World J Gastroenterol 22 (1), 361-368.
Martin, K., Woo, J., Timmins, V., Collins, J., Islam, A., Newton, D., et al., 2016. Binge eating and
AN US
emotional eating behaviors among adolescents and young adults with bipolar disorder. J Affect Disord 195, 88-95.
Merikangas, K.R., Akiskal, H.S., Angst, J., Greenberg, P.E., Hirschfeld, R.M., Petukhova, M., et al., 2007. Lifetime and 12-month prevalence of bipolar spectrum disorder in the National Comorbidity Survey replication. Arch Gen Psychiatry 64 (5), 543-552.
Montgomery, P., Richardson, A.J., 2008. Omega-3 fatty acids for bipolar disorder. Cochrane Database Syst Rev(2), CD005169.
M
Noaghiul, S., Hibbeln, J.R., 2003. Cross-national comparisons of seafood consumption and rates of bipolar disorders. Am J Psychiatry 160 (12), 2222-2227. Noguchi, R., Hiraoka, M., Watanabe, Y., Kagawa, Y., 2013. Relationship between dietary patterns and
ED
depressive symptoms: difference by gender, and unipolar and bipolar depression. J Nutr Sci Vitaminol (Tokyo) 59 (2), 115-122.
Odamaki, T., Kato, K., Sugahara, H., Xiao, J.Z., Abe, F., Benno, Y., 2016. Effect of probiotic yoghurt
PT
on animal-based diet-induced change in gut microbiota: an open, randomised, parallel-group study. Benef Microbes 7 (4), 473-484.
CE
Patel, V., Chisholm, D., Parikh, R., Charlson, F.J., Degenhardt, L., Dua, T., et al., 2016. Addressing the burden of mental, neurological, and substance use disorders: key messages from Disease Control Priorities, 3rd edition. Lancet 387 (10028), 1672-1685.
AC
Phillips, M.R., Zhang, J., Shi, Q., Song, Z., Ding, Z., Pang, S., et al., 2009. Prevalence, treatment, and associated disability of mental disorders in four provinces in China during 2001-05: an epidemiological survey. Lancet 373 (9680), 2041-2053.
Schmidt, C., 2015. Mental health: thinking from the gut. Nature 518 (7540), S12-15. Sylvia, L.G., Friedman, E.S., Kocsis, J.H., Bernstein, E.E., Brody, B.D., Kinrys, G., et al., 2013. Association of exercise with quality of life and mood symptoms in a comparative effectiveness study of bipolar disorder. J Affect Disord 151 (2), 722-727. Unno, T., Choi, J.H., Hur, H.G., Sadowsky, M.J., Ahn, Y.T., Huh, C.S., et al., 2015. Changes in human gut microbiota influenced by probiotic fermented milk ingestion. J Dairy Sci 98 (6), 3568-3576. Vanegas, S.M., Meydani, M., Barnett, J.B., Goldin, B., Kane, A., Rasmussen, H., et al., 2017. Substituting whole grains for refined grains in a 6-wk randomized trial has a modest effect on gut
ACCEPTED MANUSCRIPT
microbiota and immune and inflammatory markers of healthy adults. Am J Clin Nutr 105 (3), 635-650. Xia, X., Li, G., Ding, Y., Ren, T., Zheng, J., Kan, J., 2017. Effect of Whole Grain Qingke (Tibetan Hordeum vulgare L. Zangqing 320) on the Serum Lipid Levels and Intestinal Microbiota of Rats under High-Fat Diet. J Agric Food Chem 65 (13), 2686-2693. Yang, Y.X., Wang, G.Y., Pan, X.C., 2004. The Food Composition of China,2004. Peking University Medical Press, Beijing Zhang, L., Cao, X.L., Wang, S.B., Zheng, W., Ungvari, G.S., Ng, C.H., et al., 2017. The prevalence of bipolar disorder in China: A meta-analysis. J Affect Disord 207, 413-421.
CR IP T
Zhang, W., Li, Q., Shi, L., Lu, K., Shang, Q., Yao, L., et al., 2009. [Investigation of dietary intake of
AC
CE
PT
ED
M
AN US
cadmium in certain polluted area of south in China]. Wei Sheng Yan Jiu 38 (5), 552-554, 557.
ACCEPTED MANUSCRIPT
Table 1. Demographic and illness characteristics of BD patients Variables 36.6±15.7 36 (32.1) 68 (60.2)
AC
CE
PT
ED
M
44 (38.9) 57 (56.5) 6 (5.3) 6 (5.3)
CR IP T
AN US
Demographics Age at screen Male, n (%) Education:≥college or technical degree, n(%) Marital status, n(%) Single Married Divorced Widow NO.of individual in the household ≤3 >3 Per capita monthly income of family ≤ USD $750 >USD $750 Clinical characteristics, mean±SD Inpatient, n(%) Age at first onset(years), (n=111) Duration of illness(years) Duration of untreated illness(years) Bipolar disorder type I, n(%) Bipolar disorder type II, n(%) Baseline HAMD-17 Total Score Clinical remission(HAMD-17 score <7 ), n(%) Outpatients, n(%) Inpatients, n(%) Baseline HAMA Total Score Family history of psychiatric disorders, n(%) Number of depressive episode Number of mania episode Psychotropic medications(n=111), n(%) Mood-stabilizer Lithium Sodium vaporate Antipsychotic Quetiapine Olanzapine Aripiprazole Antidepressant Citalopram Venlafaxine; Fluoxetine
90 (79.6) 23 (20.4)
77 (68.1) 36 (31.9)
65 (57.5) 29.9±14.6 7.8±8.9 2.1±1.1 84 (74.3) 25 (25.7) 13.3±7.3 58 (51.3) 40 (35.4) 18 (15.9) 9.5±6.0 15 (13.3) 2.8±3.2 1.6±2.2 83 (74.8) 14 (12.6) 18 (16.2) 37 (33.3) 13 (11.7) 5 (4.5) 11 (9.9) 12 (10.8) 9 (8.1)
ACCEPTED MANUSCRIPT
3 (2.7) 13 (11.7) 13 (11.7) 5 (4.5) 6 (5.4)
AN US
CR IP T
10 (9.0) 6 (5.4) 86.5 (0,349)*
PT
ED
M
Fluvoxamine Mirtazapine Sertraline Paroxetine Duloxetine Sedative hypnotic agents Lorazepam Alprazolam Psychotropic medications, mg per day Metabolic parameters BMI (Kg/m2) Obesity, n(%) Waist circumvent(cm) Hip circumvent(cm) Waist/Hip ratio Body fat rate(%) Fasting blood glucose(μmmol/L) HbA1c, % Serum insulin (pmol/L) Triglyceride (mmol/L) Total cholesterol (mmol/L) LDL-cholesterol (mmol/L) HDL- cholesterol (mmol/L) Alcohol consumption, n(%) Current smoker, n(%) C-reactive protein (mmol/L) Homocysteine (mmol/L) Folic acid (mmol/L) Vitamin B12 (mmol/L)
AC
CE
*, expressed as median (interquartile)
22.2±3.2 17 (15.0) 84.4±12.8 94.1±8.8 0.9±0.1 27.5±6.8 5.3±1.8 5.7±1.4 107.7±101.7 1.2±0.9 4.7±1.3 2.8±1.0 1.4±0.3 13 (11.6) 13 (11.6) 3.4±6.3 14.3±5.7 17.1±7.6 421.7±171.0
ACCEPTED MANUSCRIPT
Table 2. The dietary intake and lifestyle measures among BD patients Variables 1939.3±772.5 258.3±114.0 55.2±35.1 74.3±33.3
AC
CE
PT
ED
CR IP T
368.6±172.8 46.7 (16.3,82.2) 7.1 (0.2,28.6) 53.6 (8.9,100.0) 50.0 (21.42,100.0) 21.4 (7.1, 42.8) 0.6 (0.0,14.3) 156.6±173.9 60.5±35.6 100.0 (50.0, 150.0)
AN US
M
Energy and nutrients Energy(kcal/day) Carbohydrates(g/day) Fat(g/day) Protein(g/day ) Kinds of food Whole grain(g/day ) Unprocessed red meat(g/d) Processed meat(g/d) Dairy products(ml/day ) Eggs(g/day ) Bean and bean products(g/day ) Nuts(g/day) Vegetables(g/day ) Seafood(g/day ) Fruit(g/day ) Soft drink consumption(ml/d), n(%) 0 < 100 ≥100 Sugar(g/day) Alcohol(g/day), n(%) 0 < 50 ≥50 Exercise and sleep Physical activity No activity Activity not sufficient (<150min/week) Sufficient activity(≥150min/week) Sleep quality Total sleep time (minutes) Mean sleep onset time (minutes)
90(79.6) 19(16.8) 4(3.6) 8.8±6.3 100 (88.5) 7 (6.2) 6 (5.3)
34 (30.1) 35 (31.0) 44 (38.9) 532.2 ±77.3 20.3±2.5
ACCEPTED MANUSCRIPT
Table 3. Correlation of lifestyle characteristics and bipolar symptoms Number of depressive episode
Number of Manic episode
Age at first onset
Whole grain
-0.245*
-0.170
-0.079
0.164
Red meat
-0.069
0.014
0.213*
-0.241*
Seafood
-0.214*
0.178
0.095
0.090
Eggs
0.037
0.203*
-0.018
0.101
Dairy products
-0.229*
0.032
-0.046
0.055
Nuts
-0.086
0.061
-0.046
0.222*
Soft drink
-0.017
0.454**
0.701**
-0.310**
Sugar
0.115
-0.117
-0.134
-0.198*
Total sleep time
-0.336**
0.035
0.182
-0.106
Current smoking
0.019
0.114
0.252*
-0.172
Alcohol consumption
-0.042
0.459**
CE
PT
AN US
ED
M
*,P<0.05; **,P<0.01
AC
CR IP T
Variables
HAMD-17 total score
0.792**
-0.127
ACCEPTED MANUSCRIPT
Table 4.Multiple linear regression analysis of HAMD-17 score and lifestyle characteristics in BD$ Variables
Unstandardized
Coefficients
Standardized coefficients
B
Std. Error
Beta
CR IP T
Independent variable: HAMD-17 score
P
t
-0.998 0.580
0.236
0.256
2.458
0.029 0.01 6
Whole grain
-0.012
0.005
-0.305
-2.338
0.02
Red meat
-0.036
0.017
0.324
-2.141
0.115
0.052
0.311 0.028
Vegetables
0.008
-0.019
CE
Fruits
AC
Exercise time (min) Sleep time （min）
0.184
PT
Soft drink
Seafood
0.011
-0.333
M
Sugar
-0.021
ED
Dairy products Nuts
0.448
AN US
Number of manic episode BMI
0.001 -0.007
-0.228 0.263 0.205
0.014
0.441
0.004
0.223
0.011
-0.177
-2.225
3 0.03 7 -1.950
0.05 5
2.21 6 1.68 6 1.99 0 1.80 4 -1.761
0.03 1 0.09 8 0.05 2 0.07 7 0.08 1
0.006
0.023
0.198
0.84 3
0.009
-0.083
-0.832
0.40 8
-0.019
0.009
-0.225
-2.128
0.03 7
$,controlled by age, gender, number of depressive episode, age at first onset, income of family, education level .
Table 5 The association between whole grain, dairy products intake and remission by Binary logistic analysis
ACCEPTED MANUSCRIPT
Variables
Unadjusted OR(95%CI)
Adjusted OR (95%CI)$
P
<200
Ref.
200-400
10.37 (1.242,86.62)
10.74（1.19,55.51)
0.035
>400
19.09 (2.155,169.09)
14.25 (1.31,80.23)
0.029
Independent variable: Clinical remission (0=HAMD-17 score≥17; 1= HAMD-17< score 17)
P for trend
CR IP T
Whole grain intake (g/day)
0.003
Dairy products intake(ml/day) Ref.
<100
1.08 (0.36,3.29)
2.35 (0.56,9.82)
0.119
>100
4.17 (1.41,20.71)
9.93 (2.28,43.16)
0.002
AN US
0
P for trend
0.003
AC
CE
PT
ED
M
$,adjusted by age, gender, number of depressive episode, age at first onset ,BMI, total energy, intake of fruits, vegetables, seafood, exercise time, smoking and alcohol consumption, income of family , education level.