- Email: [email protected]
The contribution of lifestyle factors to depressive symptoms: A cross-sectional study in Chinese college students
Author’s Accepted Manuscript The contribution of lifestyle factors to depressive symptoms: a cross-sectional study in Chinese college students Ying Xu, Juan Qi, Xiaozhong Wen, Yi Yang www.elsevier.com/locate/psychres
PII: DOI: Reference:
S0165-1781(15)30085-8 http://dx.doi.org/10.1016/j.psychres.2016.03.009 PSY9517
To appear in: Psychiatry Research Received date: 31 July 2015 Revised date: 9 December 2015 Accepted date: 4 March 2016 Cite this article as: Ying Xu, Juan Qi, Xiaozhong Wen and Yi Yang, The contribution of lifestyle factors to depressive symptoms: a cross-sectional study in Chinese college students, Psychiatry Research, http://dx.doi.org/10.1016/j.psychres.2016.03.009 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 galley proof before it is published in its final citable 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.
1
The contribution of lifestyle factors to depressive symptoms: a cross-sectional study in Chinese college students Ying Xua, Juan Qia, Xiaozhong Wenb*, Yi Yanga* bDepartment
of Epidemiology and Biostatistics, School of Public Health, Guangdong Pharmaceutical
University, Guangzhou, China bDivision
of Behavioral Medicine, Department of Pediatrics, School of Medicine and Biomedical
Sciences, State University of New York at Buffalo, Buffalo, USA [email protected] [email protected]
*Corresponding
authors. Division of Behavioral Medicine, Department of Pediatrics, School of Medicine
and Biomedical Sciences, State University of New York at Buffalo, 3435 Main St., G56 Farber Hall, Buffalo, NY 14214-3000, USA. Tel.: 1 716 829-6811; fax: 1 716 829 3993. *Corresponding
authors. Department of Epidemiology and Biostatistics, School of Public Health,
GuangDong Pharmaceutical University, Jianghai Road 283, Haizhu District of Guangzhou, China. Tel.: 0086 135 6045 0852; fax: 0086-20 3405 5802. Abstract It is well known that some lifestyle factors are related to depression, but their cumulative contribution to the depression remains unclear. This study aimed to assess the importance of multiple lifestyle factors in contributing to depressive symptoms among Chinese college students. Between September and December in 2012, we conducted a cross-sectional study among 1907 Chinese college students from Guangzhou, Southern China. College students completed self-administered questionnaires and reported their lifestyle factors including sleep quality and duration, Internet use, smoking, drinking, exercise, outdoor activity or sunlight exposure, and eating breakfast. Depression was measured using the Center for Epidemiologic Studies Depression Scale (CES-D), and mild-to-moderate depressive symptoms were defined as the CES-D score ≥16. Among all the students, 29.7% reported mild-to-moderate depressive symptoms. Higher quality and longer duration of sleep, more exercises, more outdoor activities or sunlight exposures, and eating breakfast daily were associated with a lower CES-D score, which could explain 11.3% of variance of the CES-D score, after adjusting for
2
socio-demographics, family history, interpersonal relationship, and academic characteristics using hierarchical multivariable linear regression. These associations were comparable between males and females. The protective role of healthy lifestyles should be considered in intervention programs for improving mental health among college students. Keywords: Center for Epidemiologic Studies Depression Scale (CES-D), Depressive symptoms, Lifestyles, College students, Early adulthood 1. Introduction Depression is a common mental disorder among college students. The overall prevalence of depression among college students is 30.6% with a large variation across populations (10–85%) (Ibrahim et al., 2013). Depression has numerous adverse impacts on college students such as poor quality of life, poor academic performance, early withdrawal from college, and even self-injurious behaviors such as suicide (Buchanan, 2012). College students are characterized by 1) experiencing stressful transition from adolescence to adulthood, 2) relatively healthier physical function, 3) high level of stress from multiple sources such as academic performance and competitive job seeking, and 4) spending maximum time in schools and having a relatively monotonous daily life routine. These unique characteristics not only make college students more vulnerable to being depressed but also make the research findings especially those on risk factors from general population not be generalized to college students. Lifestyle factors such as diet, physical activity, and sleep are often perceived modifiable and have been increasingly targeted to prevent or treat depression (Garcia-Toro et al., 2012). During the transition from adolescence to adulthood, college students often experience some unhealthy change in lifestyles, such as decreasing quality of diet, being less physically active, initiation or habitualization of smoking and drinking, shorter sleep duration, and fewer outdoor activities or less sunlight exposure (Butler et al., 2004; Wengreen and Moncur, 2009). In addition, some unhealthy lifestyles such as shorter sleep duration, less sunlight exposure, smoking, and drinking have become more popular among college students in recent years (Keating et al., 2005; Lund et al., 2010; Reed et al., 2007). These unfavorable changes in lifestyles have been shown to disturb brain physiology and thus may increase their vulnerability to depression (Bourre, 2006; Sarbadhikari and Saha, 2006). However, the existing evidence on the associations of these lifestyle factors with depression is quite inconsistent (Averina et al., 2005; Demura and Sato, 2003; Song et al., 2012; Tanaka et al., 2011). In addition, most of the previous studies in this field focused on community adult residents or individuals at high risk of cardiovascular diseases who usually have very different lifestyles from college students (Tanaka et al.,
3
2011) (Demura and Sato, 2003), and less is known on the independent or combined contribution of lifestyle factors such as diet, physical activity, and sleep to depression among college students, compared with other well-established factors including less advantageous socio-demographics, academic stress, neuroticism personality features, less use of social network, an ineffective coping strategy, and less family support associated with high risk of depression (Kato, 2015; Harris and Molock, 2000; Ibrahim et al., 2012, 2013; Jorm et al., 2000; Rice et al., 2014). Given the modifiability of these lifestyle factors and their good potential to be used to prevent or treat depression (Garcia-Toro et al., 2012), it is crucial to better understand their roles in the development and progression of depression in college students. Therefore, in this cross-sectional study we aimed to identify the most important lifestyle factors for depressive symptoms among Chinese college students, by evaluating the independent contributions of lifestyle factors adjusting for each other and potential confounders. 2. Participants and methods 2.1. Sample and participants We conducted this cross-sectional study among college students aged 17–27 years in Guangzhou, Southern China, from September to December in 2012. We enrolled study participants by a three-stage sampling strategy to get a representative sample: 1) six of 21 eligible universities were selected from Guangzhou through simple random sampling, 2) one school was randomly selected from each of the six selected universities through simple random sampling, and 3) all the classes in those selected schools were included in the study. A total of 2134 students in these selected classes were invited to participate in the study. Among them, 191 were absent in the classes, 33 refused, and 1910 were willing to participate in the survey. 2.2. Data collection College students completed self-administrated questionnaires in their classrooms during regular class time. At the beginning of the survey, trained research staff introduced the study purposes and also ensured students about data confidentiality. Of the 1910 participating students, three did not complete the questionnaire and were thus excluded, which left 1,907 participants with an average age of 19.5 an a years (890 boys and 1017 girls) in the final analytic sample. The study was approved by the Medical Ethical Committee of the Guangdong Pharmaceutical University and written informed consent was obtained from all participants. 2.3. Questionnaire and measures
4
2.3.1. Outcome measures The Center for Epidemiologic Studies Depression Scale (CES-D) was used to measure depression (Radloff, 1991). Designed for use in epidemiological studies on depression, the CES-D is fairly reliable among college students (Shean and Baldwin, 2008). Briefly, CES-D comprises 20 self-reported items on depression-related symptoms. The respondent rates his or her frequency of each symptom in the past week based on a four-point Likert scale ranging from 0 (rarely or none of the time (less than 1 day)) to 3 (most or all of the time (5–7 days)). After reverse-coding four items, we calculated the total score of the 20 items (range 0–60) with a higher score indicating higher severity of depressive symptoms. We also defined mild-to-moderate level of depression as the CES-D score equal to 16 or higher. In our sample, the CES-D scale showed good internal consistency reliability as measured by Spearman-Brown (0.829) and Cronbach’s Alpha (0.848). 2.3.2. Exposure measures Students self-reported their lifestyle factors in the past year including their quality (high, medium, and poor) and duration (hours per day, including napping) of sleep, the time of Internet use (hours per day), the frequency of eating breakfast (times per week), cigarette smoking status (current smokers and non-smokers), alcohol drinking status (current drinkers and non-drinkers), the frequency of physical activity (times per week), and the time of outdoor activity or sunlight exposure (hours per day). Table 2 shows the detailed categories for each exposure measure. Smokers were defined as those smoking more than five packs of cigarettes so far, and drinkers were defined as those drinking at least once a week for six consecutive months. 2.3.3. Confounder measures Based on the literature and prior knowledge, the potential confounders we considered included sex, single child status, grade (freshman vs. higher), major (medicine, arts, and science and engineering), academic stress (low, medium, and high), interest in the current major (low, medium, and strong), interpersonal relationship (bad, fair, and good), family monthly income (per capita), and family history of depression. 2.4. Statistical analysis Mean and standard deviation for continuous variables and percentage for categorical variables were used to describe the distributions of lifestyle factors, depression, and confounders. Analysis of variance along with Student–Newman–Keuls (SNK, post-hoc pairwise comparisons) was used to compare mean CES-D scores (continuous) across different levels of categorical lifestyle factors and
5
confounders. Hierarchical multivariable linear regression models were employed to examine the association between lifestyle factors and CES-D score, adjusting for potential confounders. Specifically, significant potential confounders were firstly included into the regression model as block 1 by the enter method, and then lifestyle factors were added as block 2 by the stepwise method (Pin<0.05, Pout>0.10). Given the dose–response associations between levels of lifestyle factors and depression, we treated lifestyle factors as ordinal variables (1–3 or 1–4) in the multivariable linear regression model. The product terms of sex and lifestyle factors as independent variables were included in the linear regression models to analyze the interaction between them on the CES-D score. All analyses were conducted in IBM SPSS (Statistical Package for the Social Sciences) for Windows, version 20.0 (IBM Inc, New York, NY). The significance level (α) was set as 0.05. 3. Results 3.1. Sample characteristics Table 1 shows socio-demographic, academic, and family characteristics of the final analytic sample (N=1907). Of them, 46.7% were male; 24.3% were the single child in their family; 74.1% were freshmen; and 31.0% majored in medicine, 22.6% in arts, and 46.4% in science and engineering. Most students reported strong (53.4%) or medium (40.5%) levels of interest in their current majors and high (39.9%) or medium (55.1%) level of academic stress. Their family monthly income per capita was distributed as follows: 30.4% <1000 RMB, 42.0% between 1000 and 2999 RMB, and 27.6% ≥3000 RMB. The majority of them had good (53.5%) and fair (43.8%) interpersonal relationships. Very few students (2.5%) reported family history of depression. 3.2. CES-D depression score In our analytic sample, the mean CES-D score was 12.2±7.7 (12.4±8.0 in males vs. 11.9±7.5 in females). Using CES-D score ≥16 as the cut-off point, 29.7% (31.1% for males vs. 28.4% for females) had mild-to-moderate depression. The mean CES-D score varied significantly by students’ major, level of interest in the current major, academic stress, interpersonal relationship, family monthly income, and family history of depression (Table 1). These characteristics were thus included in the following hierarchical multivariable regression model. However, sex, single child status, and grade were not associated with the CES-D score. 3.3. Lifestyles and depressive symptoms Significant interactions between sex and lifestyle factors on the depression score in our sample
6
were not found (p-value for interaction of sex and lifestyle factors ranged from 0.08 to 0.638). Therefore, we only reported the analytic results for the total sample. Table 2 shows the distribution of the eight lifestyle factors of our interest and also their associations with CES-D scores in both univariate and multivariable analyses. Those who reported poor quality of sleep, shorter sleep duration (i.e., < 6 h), eating breakfast less than 2 days per week, currently smoked, and had fewer physical activities and fewer outdoor activities or sunlight exposures had a higher mean CES-D score, and there was a “dose– response” relationship between the levels of these factors and the mean CES-D score. These associations remained significant in the hierarchical multivariable regression model after controlling for students’ major, level of interest in the current major, academic stress, interpersonal relationship, family monthly income per capital, and family history of depression. The corresponding mean differences in the CES-D score for one level of increment (e.g., low to medium or medium to high) in each lifestyle factor were −3.92 (95% confidence interval or CI, −4.49, −3.35) for higher sleep quality, −1.79 (−2.49, −1.09) for longer sleep duration, −1.24 (−1.70, −0.78) for eating breakfast more frequently, −0.83 (−1.27, −0.40) for more frequent physical activity, and −1.07 (−1.64, −0.51) for more outdoor activities or sunlight exposures, respectively. These significant lifestyle factors as a whole could explain 11.3% of the variance in CES-D depression score after partialling out major, level of interest in current major, academic stress, interpersonal relationship, family monthly income per capital, and family history of depression (F change=53.483, p<0.001). The distribution of eight lifestyle factors was similar between males and females, except for time of Internet use (males < females) and outdoor activity or sunlight exposure (males > females) (Table 3). Overall, the sex-stratified analysis yielded similar results of the associations between lifestyle factors and depression to sex-combined analysis (among the total sample), except that drinking was significantly associated with the CES-D score among females only (Table 4). The percentages of variance in CES-D score explained by regression models were 13.1% for males (F change=27.056, p<0.001) and 10.3% for females (F change=36.264, p<0.001). 4. Discussion In this cross-sectional study, we examined the associations between multiple lifestyle factors and depression among Chinese college students. We found that about 3 out of 10 of participants were experiencing mild-to-moderate depression. Poor quality and short duration of sleep, fewer physical activities and outdoor activities, and eating breakfast less frequently were significantly associated with higher CES-D depression score and as a whole accounted for 11.3% of the total variance in the score. Our findings added to the limited evidence on the role of lifestyle factors in depression among college students. We concluded that Chinese college students with multiple unhealthy lifestyles have higher
7
risk for depression and need special attention from college administrators and healthcare providers. In the present study, we found that the prevalence of depressive symptoms was 29.7%, a similar rate to the 27.1% among Turkish university students (Bayram and Bilgel, 2008), but much higher than the 11.7% found among the Chinese university students (Chen et al., 2013) and much lower than the 85% among American university students (Garlow et al., 2008). The wide variations appear to originate from cultural differences of the studied population, different methods of assessment, and different study methods. Despite these, more attention should be given to the identification and management of depression in university settings to decrease the negative effects resulting from mental disorders. Consistent with the literature (Moo-Estrella et al., 2005), we found that poor quality of sleep and short duration of sleep were associated with higher depression score among college students. Although poor sleep is one of the depressive symptoms, some prospective studies indicate that sleep disturbance itself could lead to the development or progression of depression (Breslau et al., 1996; Chang et al., 1997). For example, Naomi et al. (Breslau et al., 1996) reported that the relative risk for major depression among young adults aged 21–30 years was nearly four times higher for those with a history of insomnia compared with others who did not. Although our cross-sectional study does not allow causal interpretation for the sleep–depression association, we suspect that sleep problems are likely to precede depression in our sample because students reported the quality and duration of their sleep in the past year, while depressive symptoms in the past week. Furthermore, our sensitivity analysis by excluding the sleep item from the CES-D scale suggested that the mean CES-D score remained significantly different across students with different levels of quality and sleep duration (data not shown). Therefore, it is necessary to educate students to form healthy sleep habits and encourage those students with sleep problems to attend sleep training programs or seek professional treatment if necessary, all of which might help to prevent and/or reduce depression among college students. Observational research indicates that physical activity can promote emotional well-being, relieve depressive mood, and even reduce risk of suicide among college students (Elliot et al., 2012; Taliaferro et al., 2009b; Tyson et al., 2010). Also, a review of randomized controlled trials concludes that exercises seem to be able to reduce depressive symptoms in adults aged 18 and above, who are diagnosed with depression (Rimer et al., 2012). The present study adds to the evidence on the protective effect of regular exercises on depression among college students. However, most of the college students in our sample were physically inactive (less than two times per week). So it is urgent to develop effective interventions that help college students to achieve sufficient level of physical activity. Similarly, more outdoor activities or sunlight exposures were associated with a lower depression score in our sample. Increasing evidence suggests that outdoor activities and sunlight exposure are effective supplemental treatments for patients with mild-to–moderate depression (Garcia-Toro et al.,
8
2012). The possible mechanisms include that 1) sunlight exposure may change the secretion of some brain neurochemicals such as cortical secretion, thyroid stimulating hormone, melatonin, and monoamines (Kent et al., 2009; Soria and Urretavizcaya, 2009) and 2) sunlight exposure may increase the plasma level of inflammatory cytokines such as interleukin 6 (IL-6) that can counteract depression (Levandovski et al., 2013). Eating breakfast regularly leads to improved mood, better memory, more energy, and feelings of calmness (Lombard, 2000). In our sample, eating breakfast regularly was associated with lower CES-D scores, which was consistent with previous studies among college students and adolescents (Tanihata et al., 2012). Notably, univariate analysis showed that smokers had a higher mean CES-D score than non-smokers in our study, but the difference was not significant after adjustment for some confounders and other lifestyle factors. The existing evidence on the association between smoking and depression is inconsistent. Some studies (Boden et al., 2010; Dos et al., 2010), but not others (Rohde et al., 2004; Roy et al., 2001), suggested that smoking was associated with high risk of depressive symptoms. The reasons for this inconsistency are unclear, but it may be partially explained by their complex bi-directional relationship; for example, smoking may be used as a coping strategy for depressive symptoms. Furthermore, we found no association between the time of Internet use and depressive symptoms. This null association may be because the Internet use for different purposes (e.g., learning vs. entertainment) has different effects on mental health (Selfhout et al., 2009). Unfortunately, we could not assess this possibility in our sample, as we did not collect the information on the purposes of Internet use. We did not find significant sex difference in either the mean CES-D depression score or the associations between lifestyle factors and depression, although males and females had quite different lifestyles. These findings suggested that lifestyle interventions for depression should be recommended to college students regardless of sex. Limitations Our study had several notable limitations. Firstly, findings from our cross-sectional study are only suggestive (not confirmative) for causal associations between lifestyles and depressive symptoms. Secondly, all lifestyle factors were based on the college students’ self-reports and were thus subject to recall bias. But our assurance of data confidentiality may partially overcome under-reporting of unfavorable lifestyles. Thirdly, we did not use standardized scales to assess sleep quality (e.g., Pittsburgh Sleep Quality Index) and physical activity (e.g., International Physical Activity Questionnaire), due to limited survey time. This could limit the comparison of our findings with previous studies. Lastly, we might miss some habitual truants during this classroom-based survey. On average, these
9
absentees have higher risk of depression so that we probably underestimated the prevalence of depressive symptoms. Conclusion In summary, higher quality and longer duration of sleep, physical exercises, outdoor activities, and eating breakfast regularly are associated with a lower CES-D score among a sample of Chinese college students. Taking into account these lifestyle factors may help to better design programs to prevent or treat depression among college students. Implications for school health Depression as a psychiatric disorder characterized by symptoms of persistent feelings of hopelessness and dejection has been an epidemic mental problem among college students worldwide and has been shown to have detrimental effects. A better understanding of depression and its correlates is essential in planning for appropriate interventions in this population group. The information and results from this article may help inform school-based health promotion practices and make targeted measures to improve lifestyle factors in addition to conventional factors, including factors related to stress from study, interpersonal relationship and social supports, and so on. The results showed that these lifestyle factors as a whole could explain 11.6% of variance of the CES-D score, after adjusting for socio-demographics, family history, interpersonal relationship, and academic characteristics. For example, higher quality and longer duration of sleep, no smoking, more exercises, more outdoor activities or sunlight exposures, and eating breakfast daily were significantly associated with lower CES-D scores. Therefore, the mental health education workers in the colleges should advise the students about some methods to improve the quality of sleep, including avoidance of excitant food, trying to relax themselves, doing some reading and listening to good music, and conducting health education on the harms of smoking and the benefits of proper exercise habits, especially to encourage students to participating in outdoor activities. Furthermore, this study suggested that future research for evaluating the effect of the ways to improve lifestyle factors on decreasing depression should be considered. Acknowledgments This project was funded by the Foundation of the Ministry of Education of China for Outstanding Young Teachers in University (Grant no: 11YJCZH204, awarded to Ying Xu). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors would like to thank the involved college teachers and research assistants for their help in the
10
investigation and also the participating college students for their time.
References Averina, M., Nilssen, O., Brenn, T., Brox, J., Arkhipovsky, V. L., Kalinin, A. G., 2005. Social and lifestyle determinants of depression, anxiety, sleeping disorders and self-evaluated quality of life in Russia--a population-based study in Arkhangelsk. Soc Psychiatry Psychiatr Epidemiol 40, 511-518. Bayram, N., Bilgel, N., 2008. The prevalence and socio-demographic correlations of depression, anxiety and stress among a group of university students. Soc Psychiatry Psychiatr Epidemiol 43, 667-672. Boden, J. M., Fergusson, D. M., Horwood, L. J., 2010. Cigarette smoking and depression: tests of causal linkages using a longitudinal birth cohort. Br J Psychiatry 196, 440-446. Bourre, J. M., 2006. Effects of nutrients (in food) on the structure and function of the nervous system: update on dietary requirements for brain. Part 2 : macronutrients. J NUTR HEALTH AGING 10, 386-399. Breslau, N., Roth, T., Rosenthal, L., Andreski, P., 1996. Sleep disturbance and psychiatric disorders: a longitudinal epidemiological study of young adults. Biol Psychiatry 39, 411-418. Buchanan, J. L., 2012. Prevention of depression in the college student population: a review of the literature. Arch Psychiatr Nurs 26, 21-42. Butler, S. M., Black, D. R., Blue, C. L., Gretebeck, R. J., 2004. Change in diet, physical activity, and body weight in female college freshman. Am J Health Behav 28, 24-32. Chang, P. P., Ford, D. E., Mead, L. A., Cooper-Patrick, L., Klag, M. J., 1997. Insomnia in young men and subsequent depression. The Johns Hopkins Precursors Study. AM J EPIDEMIOL 146, 105-114. Chen, L., Wang, L., Qiu, X. H., Yang, X. X., Qiao, Z. X., Yang, Y. J., Liang, Y., 2013. Depression among Chinese university students: prevalence and socio-demographic correlates. PLOS ONE 8, e58379. Demura, S., Sato, S., 2003. Relationships between depression, lifestyle and quality of life in the community dwelling elderly: a comparison between gender and age groups. J Physiol Anthropol Appl Human Sci 22, 159-166. Dos, S. V., Migott, A. M., Bau, C. H., Chatkin, J. M., 2010. Tobacco smoking and depression: results of a cross-sectional study. Br J Psychiatry 197, 413-414. Elliot, C. A., Kennedy, C., Morgan, G., Anderson, S. K., Morris, D., 2012. Undergraduate physical activity and depressive symptoms: a national study. Am J Health Behav 36, 230-241. Garcia-Toro, M., Ibarra, O., Gili, M., Serrano, M. J., Olivan, B., Vicens, E., Roca, M., 2012. Four hygienic-dietary recommendations as add-on treatment in depression: a randomized-controlled trial. J Affect Disord 140, 200-203. Garcia-Toro, M., Ibarra, O., Gili, M., Serrano, M. J., Vives, M., Monzon, S., Bauza, N., Olivan, B., Vicens, E., Roca, M., 2012. Adherence to lifestyle recommendations by patients with depression. Rev Psiquiatr Salud Ment 5, 236-240. Garlow, S. J., Rosenberg, J., Moore, J. D., Haas, A. P., Koestner, B., Hendin, H., Nemeroff, C. B., 2008. Depression, desperation, and suicidal ideation in college students: results from the American Foundation for Suicide Prevention College Screening Project at Emory University. DEPRESS ANXIETY 25, 482-488. Harris, T. L., Molock, S. D., 2000. Cultural orientation, family cohesion, and family support in suicide ideation
11
and depression among African American college students. Suicide Life Threat Behav 30, 341-353. Ibrahim, A. K., Kelly, S. J., Adams, C. E., Glazebrook, C., 2013. A systematic review of studies of depression prevalence in university students. J PSYCHIATR RES 47, 391-400. Ibrahim, A. K., Kelly, S. J., Glazebrook, C., 2012. Analysis of an Egyptian study on the socioeconomic distribution of depressive symptoms among undergraduates. Soc Psychiatry Psychiatr Epidemiol 47, 927-937. Ibrahim, A. K., Kelly, S. J., Glazebrook, C., 2013. Socioeconomic status and the risk of depression among U.K. higher education students. Soc Psychiatry Psychiatr Epidemiol 48, 1491-1501. Jorm, A. F., Christensen, H., Henderson, A. S., Jacomb, P. A., Korten, A. E., Rodgers, B., 2000. Predicting anxiety and depression from personality: is there a synergistic effect of neuroticism and extraversion? J Abnorm Psychol 109, 145-149. Kato, T., 2015. The impact of coping flexibility on the risk of depressive symptoms. PLOS ONE 10, e128307. Keating, X. D., Guan, J., Pinero, J. C., Bridges, D. M., 2005. A meta-analysis of college students' physical activity behaviors. J Am Coll Health 54, 116-125. Kent, S. T., McClure, L. A., Crosson, W. L., Arnett, D. K., Wadley, V. G., Sathiakumar, N., 2009. Effect of sunlight exposure on cognitive function among depressed and non-depressed participants: a REGARDS cross-sectional study. Environ Health 8, 34. Levandovski, R., Pfaffenseller, B., Carissimi, A., Gama, C. S., Hidalgo, M. P., 2013. The effect of sunlight exposure on interleukin-6 levels in depressive and non-depressive subjects. BMC PSYCHIATRY 13, 75. Lombard, C. B., 2000. What is the role of food in preventing depression and improving mood, performance and cognitive function? Med J Aust 173 Suppl, S104-S105. Lund, H. G., Reider, B. D., Whiting, A. B., Prichard, J. R., 2010. Sleep patterns and predictors of disturbed sleep in a large population of college students. J Adolesc Health 46, 124-132. Moo-Estrella, J., Perez-Benitez, H., Solis-Rodriguez, F., Arankowsky-Sandoval, G., 2005. Evaluation of depressive symptoms and sleep alterations in college students. ARCH MED RES 36, 393-398. Naninck, E. F., Lucassen, P. J., Bakker, J., 2011. Sex differences in adolescent depression: do sex hormones determine vulnerability? J NEUROENDOCRINOL 23, 383-392. Radloff, L. S., 1991. The use of the Center for Epidemiologic Studies Depression Scale in adolescents and young adults. J Youth Adolesc 20, 149-166. Reed, M. B., Wang, R., Shillington, A. M., Clapp, J. D., Lange, J. E., 2007. The relationship between alcohol use and cigarette smoking in a sample of undergraduate college students. ADDICT BEHAV 32, 449-464. Rice, S. M., Goodall, J., Hetrick, S. E., Parker, A. G., Gilbertson, T., Amminger, G. P., Davey, C. G., McGorry, P. D., Gleeson, J., Alvarez-Jimenez, M., 2014. Online and social networking interventions for the treatment of depression in young people: a systematic review. J MED INTERNET RES 16, e206. Rimer, J., Dwan, K., Lawlor, D. A., Greig, C. A., McMurdo, M., Morley, W., Mead, G. E., 2012. Exercise for depression. Cochrane Database Syst Rev 7, D4366. Rohde, P., Kahler, C. W., Lewinsohn, P. M., Brown, R. A., 2004. Psychiatric disorders, familial factors, and cigarette smoking: II. Associations with progression to daily smoking. NICOTINE TOB RES 6, 119-132. Roy, K., Parker, G., Mitchell, P., Wilhelm, K., 2001. Depression and smoking: examining correlates in a subset of depressed patients. Aust N Z J Psychiatry 35, 329-335. Sarbadhikari, S. N., Saha, A. K., 2006. Moderate exercise and chronic stress produce counteractive effects on
12
different areas of the brain by acting through various neurotransmitter receptor subtypes: a hypothesis. THEOR BIOL MED MODEL 3, 33. Selfhout, M. H., Branje, S. J., Delsing, M., ter Bogt, T. F., Meeus, W. H., 2009. Different types of Internet use, depression, and social anxiety: the role of perceived friendship quality. J Adolesc 32, 819-833. Shean, G., Baldwin, G., 2008. Sensitivity and specificity of depression questionnaires in a college-age sample. J GENET PSYCHOL 169, 281-288. Song, M. R., Lee, Y. S., Baek, J. D., Miller, M., 2012. Physical activity status in adults with depression in the National Health and Nutrition Examination Survey, 2005-2006. PUBLIC HEALTH NURS 29, 208-217. Soria, V., Urretavizcaya, M., 2009. [Circadian rhythms and depression]. Actas Esp Psiquiatr 37, 222-232. Taliaferro, L. A., Rienzo, B. A., Pigg, R. J., Miller, M. D., Dodd, V. J., 2009. Associations between physical activity and reduced rates of hopelessness, depression, and suicidal behavior among college students. J Am Coll Health 57, 427-436. Tanaka, H., Sasazawa, Y., Suzuki, S., Nakazawa, M., Koyama, H., 2011. Health status and lifestyle factors as predictors of depression in middle-aged and elderly Japanese adults: a seven-year follow-up of the Komo-Ise cohort study. BMC PSYCHIATRY 11, 20. Tanihata, T., Kanda, H., Osaki, Y., Ohida, T., Minowa, M., Wada, K., Suzuki, K., Hayashi, K., 2012. Unhealthy Lifestyle, Poor Mental Health , and Its Correlation Among Adolescents: A Nationwide Cross-Sectional Survey. Asia Pac J Public Health. Tyson, P., Wilson, K., Crone, D., Brailsford, R., Laws, K., 2010. Physical activity and mental health in a student population. J Ment Health 19, 492-499. Wengreen, H. J., Moncur, C., 2009. Change in diet, physical activity, and body weight among young-adults during the transition from high school to college. NUTR J 8, 32.
TABLES Table 1. Univariate analysis on socio-demographic, academic, and family factors for depression among Chinese college students (N=1,907) CES-D Factor Categories n (%) score F P (mean±SD) Sex 1.798 0.180 Males 890 (46.7) 12.4±8.0 Females 1017(53.3) 11.9±7.5 Single child in the family 0.420 0.517 Yes 463(24.3) 11.9±8.0 No 1444(75.7) 12.2±7.7 Major 6.298 0.002 Medicine 592(31.0) 11.2±7.0a Arts 431(22.6) 12.3±8.0b Science or 884(46.4) 12.7±8.1b engineering Level of interest in the current major 26.146 <0.001 Strong 1018(53.4) 11.2±7.5 a Medium 773(40.5) 12.8±7.5 b Low 116(6.1) 16.1±9.5 c Grade 0.087 0.768 Freshmen 1413(74.1) 12.2±7.9
13
Non-freshmen Academic stress
494(25.9)
12.1±7.4
52.836 0.000 High 760(39.9) 14.3±8.2a Medium 1051(55.1) 10.6±6.9b Low 96( 5.0) 11.9±9.3b Family monthly income (RMB per capita) 3.972 0.019 <1000 579(30.4) 12.8±8.3a 1000-2999 801(42.0) 12.1±7.4a ≥3000 527(27.6) 11.5±7.6b Family history 42.366 0.000 Yes 47( 2.5) 14.0±8.2a No 1573(82.5) 11.4±7.3b Unknown 287(15.0) 15.8±8.8c Interpersonal relationship 125.865 0.000 Good 1019(53.5) 10.2±6.9a Fair 836(43.8) 13.8±7.7b Bad 52( 2.7) 24.0±9.0c CES-D, Center for Epidemiologic Studies Depression Scale; SD, standard deviation. a, b, c The same letter indicates no significant difference in the mean CES-D score in post-hoc pairwise comparisons.
14
Table 2 Univariate and hierarchical multivariable regression analysis on the associations between lifestyle factors and depression among Chinese college students (N=1,907) Univariate analysis Multivariable analysis Lifestyle factors n (%) CES-D score P Code in the b (95%CI) * P (mean±SD) model Quality of sleep <0.001 −3.92 (−4.49, −3.35) <0.001 Poor 36(1.9) 23.4±11.0c 1 Medium 721(37.8) 14.9±7.9b 2 High 1150(60.3) 10.1±6.6a 3 Sleep duration <0.001 −1.79(−2.49, −1.09) <0.001 a <6 h/day 130( 6.8) 17.3±9.4 1 6–7 h 1563(82.0) 11.9±7.4b 2 ≥8 h 214(11.2) 10.7±7.6b 3 Time of the 0.361 ---Internet use <3 h/day 1174(61.6) 12.0±7.8 1 3–h 557(29.2) 12.5±7.7 2 ≥5 h 176( 9.2) 12.4±7.8 3 Eating breakfast <0.001 −1.24(−1.70, −0.78) <0.001 (per week) ≤2 d 163( 8.5) 15.3±9.2a 1 3–6 d 624(32.7) 13.1±7.7b 2 Every day 1120(58.7) 11.1±7.3c 3 Smoking 0.020 ---Non-smokers 1878(98.5) 12.1±7.7a 1 Current 2 29( 1.5) 15.5±11.6b smokers Drinking <0.001 ---a Non-drinkers 1804(94.6) 12.0±7.6 1 Current 2 103( 5.4) 14.8±9.9b drinkers Physical activity <0.001 −0.83(−1.27, −0.40) <0.001 (per week) No 175( 9.2) 16.0±9.5a 1 1–2 d 1333(69.9) 11.9±7.3b 2 3–5 d 271(14.2) 11.4±7.3b 3 >5 d 128( 6.7) 10.8±8.6b 4 Outdoor <0.001 −1.07(−1.64, −0.51) <0.001 activities(per day) No 75( 3.9) 16.5±10.7a 1 <2 h 1248(65.4) 12.5±7.7b 2 ≥2 h 584(30.6) 10.8±7.1c 3 a, b, c The same letter indicates no significant difference in the mean CES-D score. ----: Not included in the model due to Pout>0.10. * Adjust for other lifestyle factors and potential confounders (major, level of interest in the current major, academic stress, interpersonal relationship, family monthly income, and family history of depression).
15
Table 3 Distribution of lifestyle factors among male and female college students (N=1,907), n (%) Lifestyle factors Males (N=890) Females(N=1017) Quality of sleep Poor 23( 2.6) 13(1.3) Medium 355(39.9) 366(36.0) High 512(57.5) 638(62.7) Sleep duration <6 h 78( 8.8) 52( 5.1) 6–7 h 703(79.0) 860(84.6) ≥8 h 109(12.2) 105(10.3) Time of the Internet use (per day) <3 h 570(64.0) 604(59.4) 3–h 240(27.0) 317(31.2) ≥5 h 80( 9.0) 96( 9.4) Eating breakfast (per week) ≤2 d 109(12.2) 54( 5.3) 3–6 d 316(35.5) 308(30.3) Every day 465(52.2) 655(64.4) Smoking Non-smokers 865(97.2) 1013(99.6) Current 25( 2.8) 4(0.4) smokers Drinking Non-drinkers 811(91.1) 993(97.7) Current 79( 8.9) 24( 2.4) drinkers Physical activity (per week) No 54 (6.1) 121(11.9) 1–2 d 608(68.3) 725(71.3) 3–5 d 149(16.7) 122(12.0) >5 d 79( 8.9) 49( 4.8) Outdoor activity and sunlight exposure (per day) No 33( 3.7) 42( 4.1) <2 h 564(63.4) 684(67.3) ≥2 h 293(32.9) 291(28.6)
P for sex-difference 0.016
0.002
0.098
<0.001
<0.001
<0.001
<0.001
0.123
16
Table 4 Sex-stratified analysis on the associations between lifestyle factors and CES-D depression score among college students (N=1,907) Males (n=890) Females (n=1017) Lifestyle factors# b (95% CI)* P b (95% CI) * P Quality of sleep −4.29(−5.11, −3.46,) <0.001 −3.52(−4.31, −2.74) <0.001 Sleep duration(h) −2.03(−3.01, −1.05) <0.001 −1.46(-2.48, −0.45) 0.005 Time of Internet use (per day) ------Eating breakfast (per week) −1.19(−1.83, −0.55) <0.001 −1.35(−2.03, −0.68) 0.001 Smoking ------Drinking ---2.84(0.26,5.43) 0.031 Physical activity (per week) −0.80(−1.45, −0.16) 0.008 −0.87(−1.48, −0.26) 0.005 Outdoor activity and sunlight −1.16(-2.02, −0.31) 0.015 −0.99(−1.75, −0.22) 0.011 exposure (per day) # Lifestyle factors were included as ordinal variables in the models (see coding in Table 2). ----: Not included in the model due to Pout>0.10. * Adjust for other lifestyle factors and potential confounders (major, level of interest in the current major, academic stress, interpersonal relationship, family monthly income, and family history of depression).
Highlights
Better quality and longer duration of sleep, no smoking, more exercises, more outdoor activities or sunlight exposures, and eating breakfast daily were significantly associated with lower Center for Epidemiologic Studies Depression Scale (CES-D) scores. These lifestyle factors as a whole could explain 11.6% of variance of the CES-D score, after adjusting for socio-demographics, family history, interpersonal relationship, and academic characteristics. Considering these lifestyle factors may help to better design programs to prevent or treat depression among college students.
PII: DOI: Reference:
S0165-1781(15)30085-8 http://dx.doi.org/10.1016/j.psychres.2016.03.009 PSY9517
To appear in: Psychiatry Research Received date: 31 July 2015 Revised date: 9 December 2015 Accepted date: 4 March 2016 Cite this article as: Ying Xu, Juan Qi, Xiaozhong Wen and Yi Yang, The contribution of lifestyle factors to depressive symptoms: a cross-sectional study in Chinese college students, Psychiatry Research, http://dx.doi.org/10.1016/j.psychres.2016.03.009 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 galley proof before it is published in its final citable 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.
1
The contribution of lifestyle factors to depressive symptoms: a cross-sectional study in Chinese college students Ying Xua, Juan Qia, Xiaozhong Wenb*, Yi Yanga* bDepartment
of Epidemiology and Biostatistics, School of Public Health, Guangdong Pharmaceutical
University, Guangzhou, China bDivision
of Behavioral Medicine, Department of Pediatrics, School of Medicine and Biomedical
Sciences, State University of New York at Buffalo, Buffalo, USA [email protected] [email protected]
*Corresponding
authors. Division of Behavioral Medicine, Department of Pediatrics, School of Medicine
and Biomedical Sciences, State University of New York at Buffalo, 3435 Main St., G56 Farber Hall, Buffalo, NY 14214-3000, USA. Tel.: 1 716 829-6811; fax: 1 716 829 3993. *Corresponding
authors. Department of Epidemiology and Biostatistics, School of Public Health,
GuangDong Pharmaceutical University, Jianghai Road 283, Haizhu District of Guangzhou, China. Tel.: 0086 135 6045 0852; fax: 0086-20 3405 5802. Abstract It is well known that some lifestyle factors are related to depression, but their cumulative contribution to the depression remains unclear. This study aimed to assess the importance of multiple lifestyle factors in contributing to depressive symptoms among Chinese college students. Between September and December in 2012, we conducted a cross-sectional study among 1907 Chinese college students from Guangzhou, Southern China. College students completed self-administered questionnaires and reported their lifestyle factors including sleep quality and duration, Internet use, smoking, drinking, exercise, outdoor activity or sunlight exposure, and eating breakfast. Depression was measured using the Center for Epidemiologic Studies Depression Scale (CES-D), and mild-to-moderate depressive symptoms were defined as the CES-D score ≥16. Among all the students, 29.7% reported mild-to-moderate depressive symptoms. Higher quality and longer duration of sleep, more exercises, more outdoor activities or sunlight exposures, and eating breakfast daily were associated with a lower CES-D score, which could explain 11.3% of variance of the CES-D score, after adjusting for
2
socio-demographics, family history, interpersonal relationship, and academic characteristics using hierarchical multivariable linear regression. These associations were comparable between males and females. The protective role of healthy lifestyles should be considered in intervention programs for improving mental health among college students. Keywords: Center for Epidemiologic Studies Depression Scale (CES-D), Depressive symptoms, Lifestyles, College students, Early adulthood 1. Introduction Depression is a common mental disorder among college students. The overall prevalence of depression among college students is 30.6% with a large variation across populations (10–85%) (Ibrahim et al., 2013). Depression has numerous adverse impacts on college students such as poor quality of life, poor academic performance, early withdrawal from college, and even self-injurious behaviors such as suicide (Buchanan, 2012). College students are characterized by 1) experiencing stressful transition from adolescence to adulthood, 2) relatively healthier physical function, 3) high level of stress from multiple sources such as academic performance and competitive job seeking, and 4) spending maximum time in schools and having a relatively monotonous daily life routine. These unique characteristics not only make college students more vulnerable to being depressed but also make the research findings especially those on risk factors from general population not be generalized to college students. Lifestyle factors such as diet, physical activity, and sleep are often perceived modifiable and have been increasingly targeted to prevent or treat depression (Garcia-Toro et al., 2012). During the transition from adolescence to adulthood, college students often experience some unhealthy change in lifestyles, such as decreasing quality of diet, being less physically active, initiation or habitualization of smoking and drinking, shorter sleep duration, and fewer outdoor activities or less sunlight exposure (Butler et al., 2004; Wengreen and Moncur, 2009). In addition, some unhealthy lifestyles such as shorter sleep duration, less sunlight exposure, smoking, and drinking have become more popular among college students in recent years (Keating et al., 2005; Lund et al., 2010; Reed et al., 2007). These unfavorable changes in lifestyles have been shown to disturb brain physiology and thus may increase their vulnerability to depression (Bourre, 2006; Sarbadhikari and Saha, 2006). However, the existing evidence on the associations of these lifestyle factors with depression is quite inconsistent (Averina et al., 2005; Demura and Sato, 2003; Song et al., 2012; Tanaka et al., 2011). In addition, most of the previous studies in this field focused on community adult residents or individuals at high risk of cardiovascular diseases who usually have very different lifestyles from college students (Tanaka et al.,
3
2011) (Demura and Sato, 2003), and less is known on the independent or combined contribution of lifestyle factors such as diet, physical activity, and sleep to depression among college students, compared with other well-established factors including less advantageous socio-demographics, academic stress, neuroticism personality features, less use of social network, an ineffective coping strategy, and less family support associated with high risk of depression (Kato, 2015; Harris and Molock, 2000; Ibrahim et al., 2012, 2013; Jorm et al., 2000; Rice et al., 2014). Given the modifiability of these lifestyle factors and their good potential to be used to prevent or treat depression (Garcia-Toro et al., 2012), it is crucial to better understand their roles in the development and progression of depression in college students. Therefore, in this cross-sectional study we aimed to identify the most important lifestyle factors for depressive symptoms among Chinese college students, by evaluating the independent contributions of lifestyle factors adjusting for each other and potential confounders. 2. Participants and methods 2.1. Sample and participants We conducted this cross-sectional study among college students aged 17–27 years in Guangzhou, Southern China, from September to December in 2012. We enrolled study participants by a three-stage sampling strategy to get a representative sample: 1) six of 21 eligible universities were selected from Guangzhou through simple random sampling, 2) one school was randomly selected from each of the six selected universities through simple random sampling, and 3) all the classes in those selected schools were included in the study. A total of 2134 students in these selected classes were invited to participate in the study. Among them, 191 were absent in the classes, 33 refused, and 1910 were willing to participate in the survey. 2.2. Data collection College students completed self-administrated questionnaires in their classrooms during regular class time. At the beginning of the survey, trained research staff introduced the study purposes and also ensured students about data confidentiality. Of the 1910 participating students, three did not complete the questionnaire and were thus excluded, which left 1,907 participants with an average age of 19.5 an a years (890 boys and 1017 girls) in the final analytic sample. The study was approved by the Medical Ethical Committee of the Guangdong Pharmaceutical University and written informed consent was obtained from all participants. 2.3. Questionnaire and measures
4
2.3.1. Outcome measures The Center for Epidemiologic Studies Depression Scale (CES-D) was used to measure depression (Radloff, 1991). Designed for use in epidemiological studies on depression, the CES-D is fairly reliable among college students (Shean and Baldwin, 2008). Briefly, CES-D comprises 20 self-reported items on depression-related symptoms. The respondent rates his or her frequency of each symptom in the past week based on a four-point Likert scale ranging from 0 (rarely or none of the time (less than 1 day)) to 3 (most or all of the time (5–7 days)). After reverse-coding four items, we calculated the total score of the 20 items (range 0–60) with a higher score indicating higher severity of depressive symptoms. We also defined mild-to-moderate level of depression as the CES-D score equal to 16 or higher. In our sample, the CES-D scale showed good internal consistency reliability as measured by Spearman-Brown (0.829) and Cronbach’s Alpha (0.848). 2.3.2. Exposure measures Students self-reported their lifestyle factors in the past year including their quality (high, medium, and poor) and duration (hours per day, including napping) of sleep, the time of Internet use (hours per day), the frequency of eating breakfast (times per week), cigarette smoking status (current smokers and non-smokers), alcohol drinking status (current drinkers and non-drinkers), the frequency of physical activity (times per week), and the time of outdoor activity or sunlight exposure (hours per day). Table 2 shows the detailed categories for each exposure measure. Smokers were defined as those smoking more than five packs of cigarettes so far, and drinkers were defined as those drinking at least once a week for six consecutive months. 2.3.3. Confounder measures Based on the literature and prior knowledge, the potential confounders we considered included sex, single child status, grade (freshman vs. higher), major (medicine, arts, and science and engineering), academic stress (low, medium, and high), interest in the current major (low, medium, and strong), interpersonal relationship (bad, fair, and good), family monthly income (per capita), and family history of depression. 2.4. Statistical analysis Mean and standard deviation for continuous variables and percentage for categorical variables were used to describe the distributions of lifestyle factors, depression, and confounders. Analysis of variance along with Student–Newman–Keuls (SNK, post-hoc pairwise comparisons) was used to compare mean CES-D scores (continuous) across different levels of categorical lifestyle factors and
5
confounders. Hierarchical multivariable linear regression models were employed to examine the association between lifestyle factors and CES-D score, adjusting for potential confounders. Specifically, significant potential confounders were firstly included into the regression model as block 1 by the enter method, and then lifestyle factors were added as block 2 by the stepwise method (Pin<0.05, Pout>0.10). Given the dose–response associations between levels of lifestyle factors and depression, we treated lifestyle factors as ordinal variables (1–3 or 1–4) in the multivariable linear regression model. The product terms of sex and lifestyle factors as independent variables were included in the linear regression models to analyze the interaction between them on the CES-D score. All analyses were conducted in IBM SPSS (Statistical Package for the Social Sciences) for Windows, version 20.0 (IBM Inc, New York, NY). The significance level (α) was set as 0.05. 3. Results 3.1. Sample characteristics Table 1 shows socio-demographic, academic, and family characteristics of the final analytic sample (N=1907). Of them, 46.7% were male; 24.3% were the single child in their family; 74.1% were freshmen; and 31.0% majored in medicine, 22.6% in arts, and 46.4% in science and engineering. Most students reported strong (53.4%) or medium (40.5%) levels of interest in their current majors and high (39.9%) or medium (55.1%) level of academic stress. Their family monthly income per capita was distributed as follows: 30.4% <1000 RMB, 42.0% between 1000 and 2999 RMB, and 27.6% ≥3000 RMB. The majority of them had good (53.5%) and fair (43.8%) interpersonal relationships. Very few students (2.5%) reported family history of depression. 3.2. CES-D depression score In our analytic sample, the mean CES-D score was 12.2±7.7 (12.4±8.0 in males vs. 11.9±7.5 in females). Using CES-D score ≥16 as the cut-off point, 29.7% (31.1% for males vs. 28.4% for females) had mild-to-moderate depression. The mean CES-D score varied significantly by students’ major, level of interest in the current major, academic stress, interpersonal relationship, family monthly income, and family history of depression (Table 1). These characteristics were thus included in the following hierarchical multivariable regression model. However, sex, single child status, and grade were not associated with the CES-D score. 3.3. Lifestyles and depressive symptoms Significant interactions between sex and lifestyle factors on the depression score in our sample
6
were not found (p-value for interaction of sex and lifestyle factors ranged from 0.08 to 0.638). Therefore, we only reported the analytic results for the total sample. Table 2 shows the distribution of the eight lifestyle factors of our interest and also their associations with CES-D scores in both univariate and multivariable analyses. Those who reported poor quality of sleep, shorter sleep duration (i.e., < 6 h), eating breakfast less than 2 days per week, currently smoked, and had fewer physical activities and fewer outdoor activities or sunlight exposures had a higher mean CES-D score, and there was a “dose– response” relationship between the levels of these factors and the mean CES-D score. These associations remained significant in the hierarchical multivariable regression model after controlling for students’ major, level of interest in the current major, academic stress, interpersonal relationship, family monthly income per capital, and family history of depression. The corresponding mean differences in the CES-D score for one level of increment (e.g., low to medium or medium to high) in each lifestyle factor were −3.92 (95% confidence interval or CI, −4.49, −3.35) for higher sleep quality, −1.79 (−2.49, −1.09) for longer sleep duration, −1.24 (−1.70, −0.78) for eating breakfast more frequently, −0.83 (−1.27, −0.40) for more frequent physical activity, and −1.07 (−1.64, −0.51) for more outdoor activities or sunlight exposures, respectively. These significant lifestyle factors as a whole could explain 11.3% of the variance in CES-D depression score after partialling out major, level of interest in current major, academic stress, interpersonal relationship, family monthly income per capital, and family history of depression (F change=53.483, p<0.001). The distribution of eight lifestyle factors was similar between males and females, except for time of Internet use (males < females) and outdoor activity or sunlight exposure (males > females) (Table 3). Overall, the sex-stratified analysis yielded similar results of the associations between lifestyle factors and depression to sex-combined analysis (among the total sample), except that drinking was significantly associated with the CES-D score among females only (Table 4). The percentages of variance in CES-D score explained by regression models were 13.1% for males (F change=27.056, p<0.001) and 10.3% for females (F change=36.264, p<0.001). 4. Discussion In this cross-sectional study, we examined the associations between multiple lifestyle factors and depression among Chinese college students. We found that about 3 out of 10 of participants were experiencing mild-to-moderate depression. Poor quality and short duration of sleep, fewer physical activities and outdoor activities, and eating breakfast less frequently were significantly associated with higher CES-D depression score and as a whole accounted for 11.3% of the total variance in the score. Our findings added to the limited evidence on the role of lifestyle factors in depression among college students. We concluded that Chinese college students with multiple unhealthy lifestyles have higher
7
risk for depression and need special attention from college administrators and healthcare providers. In the present study, we found that the prevalence of depressive symptoms was 29.7%, a similar rate to the 27.1% among Turkish university students (Bayram and Bilgel, 2008), but much higher than the 11.7% found among the Chinese university students (Chen et al., 2013) and much lower than the 85% among American university students (Garlow et al., 2008). The wide variations appear to originate from cultural differences of the studied population, different methods of assessment, and different study methods. Despite these, more attention should be given to the identification and management of depression in university settings to decrease the negative effects resulting from mental disorders. Consistent with the literature (Moo-Estrella et al., 2005), we found that poor quality of sleep and short duration of sleep were associated with higher depression score among college students. Although poor sleep is one of the depressive symptoms, some prospective studies indicate that sleep disturbance itself could lead to the development or progression of depression (Breslau et al., 1996; Chang et al., 1997). For example, Naomi et al. (Breslau et al., 1996) reported that the relative risk for major depression among young adults aged 21–30 years was nearly four times higher for those with a history of insomnia compared with others who did not. Although our cross-sectional study does not allow causal interpretation for the sleep–depression association, we suspect that sleep problems are likely to precede depression in our sample because students reported the quality and duration of their sleep in the past year, while depressive symptoms in the past week. Furthermore, our sensitivity analysis by excluding the sleep item from the CES-D scale suggested that the mean CES-D score remained significantly different across students with different levels of quality and sleep duration (data not shown). Therefore, it is necessary to educate students to form healthy sleep habits and encourage those students with sleep problems to attend sleep training programs or seek professional treatment if necessary, all of which might help to prevent and/or reduce depression among college students. Observational research indicates that physical activity can promote emotional well-being, relieve depressive mood, and even reduce risk of suicide among college students (Elliot et al., 2012; Taliaferro et al., 2009b; Tyson et al., 2010). Also, a review of randomized controlled trials concludes that exercises seem to be able to reduce depressive symptoms in adults aged 18 and above, who are diagnosed with depression (Rimer et al., 2012). The present study adds to the evidence on the protective effect of regular exercises on depression among college students. However, most of the college students in our sample were physically inactive (less than two times per week). So it is urgent to develop effective interventions that help college students to achieve sufficient level of physical activity. Similarly, more outdoor activities or sunlight exposures were associated with a lower depression score in our sample. Increasing evidence suggests that outdoor activities and sunlight exposure are effective supplemental treatments for patients with mild-to–moderate depression (Garcia-Toro et al.,
8
2012). The possible mechanisms include that 1) sunlight exposure may change the secretion of some brain neurochemicals such as cortical secretion, thyroid stimulating hormone, melatonin, and monoamines (Kent et al., 2009; Soria and Urretavizcaya, 2009) and 2) sunlight exposure may increase the plasma level of inflammatory cytokines such as interleukin 6 (IL-6) that can counteract depression (Levandovski et al., 2013). Eating breakfast regularly leads to improved mood, better memory, more energy, and feelings of calmness (Lombard, 2000). In our sample, eating breakfast regularly was associated with lower CES-D scores, which was consistent with previous studies among college students and adolescents (Tanihata et al., 2012). Notably, univariate analysis showed that smokers had a higher mean CES-D score than non-smokers in our study, but the difference was not significant after adjustment for some confounders and other lifestyle factors. The existing evidence on the association between smoking and depression is inconsistent. Some studies (Boden et al., 2010; Dos et al., 2010), but not others (Rohde et al., 2004; Roy et al., 2001), suggested that smoking was associated with high risk of depressive symptoms. The reasons for this inconsistency are unclear, but it may be partially explained by their complex bi-directional relationship; for example, smoking may be used as a coping strategy for depressive symptoms. Furthermore, we found no association between the time of Internet use and depressive symptoms. This null association may be because the Internet use for different purposes (e.g., learning vs. entertainment) has different effects on mental health (Selfhout et al., 2009). Unfortunately, we could not assess this possibility in our sample, as we did not collect the information on the purposes of Internet use. We did not find significant sex difference in either the mean CES-D depression score or the associations between lifestyle factors and depression, although males and females had quite different lifestyles. These findings suggested that lifestyle interventions for depression should be recommended to college students regardless of sex. Limitations Our study had several notable limitations. Firstly, findings from our cross-sectional study are only suggestive (not confirmative) for causal associations between lifestyles and depressive symptoms. Secondly, all lifestyle factors were based on the college students’ self-reports and were thus subject to recall bias. But our assurance of data confidentiality may partially overcome under-reporting of unfavorable lifestyles. Thirdly, we did not use standardized scales to assess sleep quality (e.g., Pittsburgh Sleep Quality Index) and physical activity (e.g., International Physical Activity Questionnaire), due to limited survey time. This could limit the comparison of our findings with previous studies. Lastly, we might miss some habitual truants during this classroom-based survey. On average, these
9
absentees have higher risk of depression so that we probably underestimated the prevalence of depressive symptoms. Conclusion In summary, higher quality and longer duration of sleep, physical exercises, outdoor activities, and eating breakfast regularly are associated with a lower CES-D score among a sample of Chinese college students. Taking into account these lifestyle factors may help to better design programs to prevent or treat depression among college students. Implications for school health Depression as a psychiatric disorder characterized by symptoms of persistent feelings of hopelessness and dejection has been an epidemic mental problem among college students worldwide and has been shown to have detrimental effects. A better understanding of depression and its correlates is essential in planning for appropriate interventions in this population group. The information and results from this article may help inform school-based health promotion practices and make targeted measures to improve lifestyle factors in addition to conventional factors, including factors related to stress from study, interpersonal relationship and social supports, and so on. The results showed that these lifestyle factors as a whole could explain 11.6% of variance of the CES-D score, after adjusting for socio-demographics, family history, interpersonal relationship, and academic characteristics. For example, higher quality and longer duration of sleep, no smoking, more exercises, more outdoor activities or sunlight exposures, and eating breakfast daily were significantly associated with lower CES-D scores. Therefore, the mental health education workers in the colleges should advise the students about some methods to improve the quality of sleep, including avoidance of excitant food, trying to relax themselves, doing some reading and listening to good music, and conducting health education on the harms of smoking and the benefits of proper exercise habits, especially to encourage students to participating in outdoor activities. Furthermore, this study suggested that future research for evaluating the effect of the ways to improve lifestyle factors on decreasing depression should be considered. Acknowledgments This project was funded by the Foundation of the Ministry of Education of China for Outstanding Young Teachers in University (Grant no: 11YJCZH204, awarded to Ying Xu). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The authors would like to thank the involved college teachers and research assistants for their help in the
10
investigation and also the participating college students for their time.
References Averina, M., Nilssen, O., Brenn, T., Brox, J., Arkhipovsky, V. L., Kalinin, A. G., 2005. Social and lifestyle determinants of depression, anxiety, sleeping disorders and self-evaluated quality of life in Russia--a population-based study in Arkhangelsk. Soc Psychiatry Psychiatr Epidemiol 40, 511-518. Bayram, N., Bilgel, N., 2008. The prevalence and socio-demographic correlations of depression, anxiety and stress among a group of university students. Soc Psychiatry Psychiatr Epidemiol 43, 667-672. Boden, J. M., Fergusson, D. M., Horwood, L. J., 2010. Cigarette smoking and depression: tests of causal linkages using a longitudinal birth cohort. Br J Psychiatry 196, 440-446. Bourre, J. M., 2006. Effects of nutrients (in food) on the structure and function of the nervous system: update on dietary requirements for brain. Part 2 : macronutrients. J NUTR HEALTH AGING 10, 386-399. Breslau, N., Roth, T., Rosenthal, L., Andreski, P., 1996. Sleep disturbance and psychiatric disorders: a longitudinal epidemiological study of young adults. Biol Psychiatry 39, 411-418. Buchanan, J. L., 2012. Prevention of depression in the college student population: a review of the literature. Arch Psychiatr Nurs 26, 21-42. Butler, S. M., Black, D. R., Blue, C. L., Gretebeck, R. J., 2004. Change in diet, physical activity, and body weight in female college freshman. Am J Health Behav 28, 24-32. Chang, P. P., Ford, D. E., Mead, L. A., Cooper-Patrick, L., Klag, M. J., 1997. Insomnia in young men and subsequent depression. The Johns Hopkins Precursors Study. AM J EPIDEMIOL 146, 105-114. Chen, L., Wang, L., Qiu, X. H., Yang, X. X., Qiao, Z. X., Yang, Y. J., Liang, Y., 2013. Depression among Chinese university students: prevalence and socio-demographic correlates. PLOS ONE 8, e58379. Demura, S., Sato, S., 2003. Relationships between depression, lifestyle and quality of life in the community dwelling elderly: a comparison between gender and age groups. J Physiol Anthropol Appl Human Sci 22, 159-166. Dos, S. V., Migott, A. M., Bau, C. H., Chatkin, J. M., 2010. Tobacco smoking and depression: results of a cross-sectional study. Br J Psychiatry 197, 413-414. Elliot, C. A., Kennedy, C., Morgan, G., Anderson, S. K., Morris, D., 2012. Undergraduate physical activity and depressive symptoms: a national study. Am J Health Behav 36, 230-241. Garcia-Toro, M., Ibarra, O., Gili, M., Serrano, M. J., Olivan, B., Vicens, E., Roca, M., 2012. Four hygienic-dietary recommendations as add-on treatment in depression: a randomized-controlled trial. J Affect Disord 140, 200-203. Garcia-Toro, M., Ibarra, O., Gili, M., Serrano, M. J., Vives, M., Monzon, S., Bauza, N., Olivan, B., Vicens, E., Roca, M., 2012. Adherence to lifestyle recommendations by patients with depression. Rev Psiquiatr Salud Ment 5, 236-240. Garlow, S. J., Rosenberg, J., Moore, J. D., Haas, A. P., Koestner, B., Hendin, H., Nemeroff, C. B., 2008. Depression, desperation, and suicidal ideation in college students: results from the American Foundation for Suicide Prevention College Screening Project at Emory University. DEPRESS ANXIETY 25, 482-488. Harris, T. L., Molock, S. D., 2000. Cultural orientation, family cohesion, and family support in suicide ideation
11
and depression among African American college students. Suicide Life Threat Behav 30, 341-353. Ibrahim, A. K., Kelly, S. J., Adams, C. E., Glazebrook, C., 2013. A systematic review of studies of depression prevalence in university students. J PSYCHIATR RES 47, 391-400. Ibrahim, A. K., Kelly, S. J., Glazebrook, C., 2012. Analysis of an Egyptian study on the socioeconomic distribution of depressive symptoms among undergraduates. Soc Psychiatry Psychiatr Epidemiol 47, 927-937. Ibrahim, A. K., Kelly, S. J., Glazebrook, C., 2013. Socioeconomic status and the risk of depression among U.K. higher education students. Soc Psychiatry Psychiatr Epidemiol 48, 1491-1501. Jorm, A. F., Christensen, H., Henderson, A. S., Jacomb, P. A., Korten, A. E., Rodgers, B., 2000. Predicting anxiety and depression from personality: is there a synergistic effect of neuroticism and extraversion? J Abnorm Psychol 109, 145-149. Kato, T., 2015. The impact of coping flexibility on the risk of depressive symptoms. PLOS ONE 10, e128307. Keating, X. D., Guan, J., Pinero, J. C., Bridges, D. M., 2005. A meta-analysis of college students' physical activity behaviors. J Am Coll Health 54, 116-125. Kent, S. T., McClure, L. A., Crosson, W. L., Arnett, D. K., Wadley, V. G., Sathiakumar, N., 2009. Effect of sunlight exposure on cognitive function among depressed and non-depressed participants: a REGARDS cross-sectional study. Environ Health 8, 34. Levandovski, R., Pfaffenseller, B., Carissimi, A., Gama, C. S., Hidalgo, M. P., 2013. The effect of sunlight exposure on interleukin-6 levels in depressive and non-depressive subjects. BMC PSYCHIATRY 13, 75. Lombard, C. B., 2000. What is the role of food in preventing depression and improving mood, performance and cognitive function? Med J Aust 173 Suppl, S104-S105. Lund, H. G., Reider, B. D., Whiting, A. B., Prichard, J. R., 2010. Sleep patterns and predictors of disturbed sleep in a large population of college students. J Adolesc Health 46, 124-132. Moo-Estrella, J., Perez-Benitez, H., Solis-Rodriguez, F., Arankowsky-Sandoval, G., 2005. Evaluation of depressive symptoms and sleep alterations in college students. ARCH MED RES 36, 393-398. Naninck, E. F., Lucassen, P. J., Bakker, J., 2011. Sex differences in adolescent depression: do sex hormones determine vulnerability? J NEUROENDOCRINOL 23, 383-392. Radloff, L. S., 1991. The use of the Center for Epidemiologic Studies Depression Scale in adolescents and young adults. J Youth Adolesc 20, 149-166. Reed, M. B., Wang, R., Shillington, A. M., Clapp, J. D., Lange, J. E., 2007. The relationship between alcohol use and cigarette smoking in a sample of undergraduate college students. ADDICT BEHAV 32, 449-464. Rice, S. M., Goodall, J., Hetrick, S. E., Parker, A. G., Gilbertson, T., Amminger, G. P., Davey, C. G., McGorry, P. D., Gleeson, J., Alvarez-Jimenez, M., 2014. Online and social networking interventions for the treatment of depression in young people: a systematic review. J MED INTERNET RES 16, e206. Rimer, J., Dwan, K., Lawlor, D. A., Greig, C. A., McMurdo, M., Morley, W., Mead, G. E., 2012. Exercise for depression. Cochrane Database Syst Rev 7, D4366. Rohde, P., Kahler, C. W., Lewinsohn, P. M., Brown, R. A., 2004. Psychiatric disorders, familial factors, and cigarette smoking: II. Associations with progression to daily smoking. NICOTINE TOB RES 6, 119-132. Roy, K., Parker, G., Mitchell, P., Wilhelm, K., 2001. Depression and smoking: examining correlates in a subset of depressed patients. Aust N Z J Psychiatry 35, 329-335. Sarbadhikari, S. N., Saha, A. K., 2006. Moderate exercise and chronic stress produce counteractive effects on
12
different areas of the brain by acting through various neurotransmitter receptor subtypes: a hypothesis. THEOR BIOL MED MODEL 3, 33. Selfhout, M. H., Branje, S. J., Delsing, M., ter Bogt, T. F., Meeus, W. H., 2009. Different types of Internet use, depression, and social anxiety: the role of perceived friendship quality. J Adolesc 32, 819-833. Shean, G., Baldwin, G., 2008. Sensitivity and specificity of depression questionnaires in a college-age sample. J GENET PSYCHOL 169, 281-288. Song, M. R., Lee, Y. S., Baek, J. D., Miller, M., 2012. Physical activity status in adults with depression in the National Health and Nutrition Examination Survey, 2005-2006. PUBLIC HEALTH NURS 29, 208-217. Soria, V., Urretavizcaya, M., 2009. [Circadian rhythms and depression]. Actas Esp Psiquiatr 37, 222-232. Taliaferro, L. A., Rienzo, B. A., Pigg, R. J., Miller, M. D., Dodd, V. J., 2009. Associations between physical activity and reduced rates of hopelessness, depression, and suicidal behavior among college students. J Am Coll Health 57, 427-436. Tanaka, H., Sasazawa, Y., Suzuki, S., Nakazawa, M., Koyama, H., 2011. Health status and lifestyle factors as predictors of depression in middle-aged and elderly Japanese adults: a seven-year follow-up of the Komo-Ise cohort study. BMC PSYCHIATRY 11, 20. Tanihata, T., Kanda, H., Osaki, Y., Ohida, T., Minowa, M., Wada, K., Suzuki, K., Hayashi, K., 2012. Unhealthy Lifestyle, Poor Mental Health , and Its Correlation Among Adolescents: A Nationwide Cross-Sectional Survey. Asia Pac J Public Health. Tyson, P., Wilson, K., Crone, D., Brailsford, R., Laws, K., 2010. Physical activity and mental health in a student population. J Ment Health 19, 492-499. Wengreen, H. J., Moncur, C., 2009. Change in diet, physical activity, and body weight among young-adults during the transition from high school to college. NUTR J 8, 32.
TABLES Table 1. Univariate analysis on socio-demographic, academic, and family factors for depression among Chinese college students (N=1,907) CES-D Factor Categories n (%) score F P (mean±SD) Sex 1.798 0.180 Males 890 (46.7) 12.4±8.0 Females 1017(53.3) 11.9±7.5 Single child in the family 0.420 0.517 Yes 463(24.3) 11.9±8.0 No 1444(75.7) 12.2±7.7 Major 6.298 0.002 Medicine 592(31.0) 11.2±7.0a Arts 431(22.6) 12.3±8.0b Science or 884(46.4) 12.7±8.1b engineering Level of interest in the current major 26.146 <0.001 Strong 1018(53.4) 11.2±7.5 a Medium 773(40.5) 12.8±7.5 b Low 116(6.1) 16.1±9.5 c Grade 0.087 0.768 Freshmen 1413(74.1) 12.2±7.9
13
Non-freshmen Academic stress
494(25.9)
12.1±7.4
52.836 0.000 High 760(39.9) 14.3±8.2a Medium 1051(55.1) 10.6±6.9b Low 96( 5.0) 11.9±9.3b Family monthly income (RMB per capita) 3.972 0.019 <1000 579(30.4) 12.8±8.3a 1000-2999 801(42.0) 12.1±7.4a ≥3000 527(27.6) 11.5±7.6b Family history 42.366 0.000 Yes 47( 2.5) 14.0±8.2a No 1573(82.5) 11.4±7.3b Unknown 287(15.0) 15.8±8.8c Interpersonal relationship 125.865 0.000 Good 1019(53.5) 10.2±6.9a Fair 836(43.8) 13.8±7.7b Bad 52( 2.7) 24.0±9.0c CES-D, Center for Epidemiologic Studies Depression Scale; SD, standard deviation. a, b, c The same letter indicates no significant difference in the mean CES-D score in post-hoc pairwise comparisons.
14
Table 2 Univariate and hierarchical multivariable regression analysis on the associations between lifestyle factors and depression among Chinese college students (N=1,907) Univariate analysis Multivariable analysis Lifestyle factors n (%) CES-D score P Code in the b (95%CI) * P (mean±SD) model Quality of sleep <0.001 −3.92 (−4.49, −3.35) <0.001 Poor 36(1.9) 23.4±11.0c 1 Medium 721(37.8) 14.9±7.9b 2 High 1150(60.3) 10.1±6.6a 3 Sleep duration <0.001 −1.79(−2.49, −1.09) <0.001 a <6 h/day 130( 6.8) 17.3±9.4 1 6–7 h 1563(82.0) 11.9±7.4b 2 ≥8 h 214(11.2) 10.7±7.6b 3 Time of the 0.361 ---Internet use <3 h/day 1174(61.6) 12.0±7.8 1 3–h 557(29.2) 12.5±7.7 2 ≥5 h 176( 9.2) 12.4±7.8 3 Eating breakfast <0.001 −1.24(−1.70, −0.78) <0.001 (per week) ≤2 d 163( 8.5) 15.3±9.2a 1 3–6 d 624(32.7) 13.1±7.7b 2 Every day 1120(58.7) 11.1±7.3c 3 Smoking 0.020 ---Non-smokers 1878(98.5) 12.1±7.7a 1 Current 2 29( 1.5) 15.5±11.6b smokers Drinking <0.001 ---a Non-drinkers 1804(94.6) 12.0±7.6 1 Current 2 103( 5.4) 14.8±9.9b drinkers Physical activity <0.001 −0.83(−1.27, −0.40) <0.001 (per week) No 175( 9.2) 16.0±9.5a 1 1–2 d 1333(69.9) 11.9±7.3b 2 3–5 d 271(14.2) 11.4±7.3b 3 >5 d 128( 6.7) 10.8±8.6b 4 Outdoor <0.001 −1.07(−1.64, −0.51) <0.001 activities(per day) No 75( 3.9) 16.5±10.7a 1 <2 h 1248(65.4) 12.5±7.7b 2 ≥2 h 584(30.6) 10.8±7.1c 3 a, b, c The same letter indicates no significant difference in the mean CES-D score. ----: Not included in the model due to Pout>0.10. * Adjust for other lifestyle factors and potential confounders (major, level of interest in the current major, academic stress, interpersonal relationship, family monthly income, and family history of depression).
15
Table 3 Distribution of lifestyle factors among male and female college students (N=1,907), n (%) Lifestyle factors Males (N=890) Females(N=1017) Quality of sleep Poor 23( 2.6) 13(1.3) Medium 355(39.9) 366(36.0) High 512(57.5) 638(62.7) Sleep duration <6 h 78( 8.8) 52( 5.1) 6–7 h 703(79.0) 860(84.6) ≥8 h 109(12.2) 105(10.3) Time of the Internet use (per day) <3 h 570(64.0) 604(59.4) 3–h 240(27.0) 317(31.2) ≥5 h 80( 9.0) 96( 9.4) Eating breakfast (per week) ≤2 d 109(12.2) 54( 5.3) 3–6 d 316(35.5) 308(30.3) Every day 465(52.2) 655(64.4) Smoking Non-smokers 865(97.2) 1013(99.6) Current 25( 2.8) 4(0.4) smokers Drinking Non-drinkers 811(91.1) 993(97.7) Current 79( 8.9) 24( 2.4) drinkers Physical activity (per week) No 54 (6.1) 121(11.9) 1–2 d 608(68.3) 725(71.3) 3–5 d 149(16.7) 122(12.0) >5 d 79( 8.9) 49( 4.8) Outdoor activity and sunlight exposure (per day) No 33( 3.7) 42( 4.1) <2 h 564(63.4) 684(67.3) ≥2 h 293(32.9) 291(28.6)
P for sex-difference 0.016
0.002
0.098
<0.001
<0.001
<0.001
<0.001
0.123
16
Table 4 Sex-stratified analysis on the associations between lifestyle factors and CES-D depression score among college students (N=1,907) Males (n=890) Females (n=1017) Lifestyle factors# b (95% CI)* P b (95% CI) * P Quality of sleep −4.29(−5.11, −3.46,) <0.001 −3.52(−4.31, −2.74) <0.001 Sleep duration(h) −2.03(−3.01, −1.05) <0.001 −1.46(-2.48, −0.45) 0.005 Time of Internet use (per day) ------Eating breakfast (per week) −1.19(−1.83, −0.55) <0.001 −1.35(−2.03, −0.68) 0.001 Smoking ------Drinking ---2.84(0.26,5.43) 0.031 Physical activity (per week) −0.80(−1.45, −0.16) 0.008 −0.87(−1.48, −0.26) 0.005 Outdoor activity and sunlight −1.16(-2.02, −0.31) 0.015 −0.99(−1.75, −0.22) 0.011 exposure (per day) # Lifestyle factors were included as ordinal variables in the models (see coding in Table 2). ----: Not included in the model due to Pout>0.10. * Adjust for other lifestyle factors and potential confounders (major, level of interest in the current major, academic stress, interpersonal relationship, family monthly income, and family history of depression).
Highlights
Better quality and longer duration of sleep, no smoking, more exercises, more outdoor activities or sunlight exposures, and eating breakfast daily were significantly associated with lower Center for Epidemiologic Studies Depression Scale (CES-D) scores. These lifestyle factors as a whole could explain 11.6% of variance of the CES-D score, after adjusting for socio-demographics, family history, interpersonal relationship, and academic characteristics. Considering these lifestyle factors may help to better design programs to prevent or treat depression among college students.