关键词: 抑郁, 中年人, 老年人, 女性, 中国健康与养老追踪调查, 影响因素分析

Abstract:

Prevalence and Associated Factors of Depression among Middle-aged and Elderly Women
YE Haichun1, YAN Yajie2,3, WANG Quan2, 3*
1.School of Nursing, Shandong Xiehe University, Jinan 250109, China
2.School of Health Sciences, Wuhan University, Wuhan 430071, China
3.Global Health Institute, Wuhan University, Wuhan 430072, China
*Corresponding author: WANG Quan, Associate professor, Master supervisor. E-mail: wangquan73@whu.edu.cn
【Abstract】Background Depression is a common mental illness threatening physical and psychological health of middle-aged and elderly people. However, there are few large-scale studies focusing on depression and its influencing factors in middle-aged and elderly Chinese women. Objective To investigate the depression prevalence and associated factors in middle-agedand elderly Chinese women, providing evidence for exploring mental health and effective interventions in this population.Methods This study was conducted from January to March 2021. Data were obtained from the China Health and RetirementLongitudinal Study (CHARLS), involving 7963 women at age 45 or over, including demographic characteristics, physicalhealth status, socio-economic features, life satisfaction, regional distribution(eastern, central or western China), and depressive prevalence assessed by the 10-item Centre for Epidemiologic Studies Depression Scale (CES-D-10). The score of CES-D-10 ≥ 10 was considered as depressive symptoms. Robust OLS regression, robust Tobit regression and robust Logit regression were used to identify associated factors of depressive symptoms. Results The median CES-D-10 score of the participants was 8(4, 14)points. Except for 4490 cases(56.39%), the remaining 3473 cases (43.61%) were found with depressive symptoms, including1715(41.52%) aged 45-59 years, and 1758(45.88%) aged 60 or over. Analyses using three regression models indicated that age and age squared(OR=1.099, 0.999), education level (OR=0.897), living in rural or urban areas (OR=0.731), self-rated health (OR=1.245), physical disability (OR=1.332), chronic disease (OR=1.172), troubled with body pains(OR=1.579), BADL (activities of daily living) disability(OR=1.734), IADL(instrumental activities of daily living) disability(OR=1.967), living with spouse(partner)or not(OR=0.763), number of children(OR=1.074), using the internet or not(OR=0.773), having care support or not when needed (OR=1.509), having debt(OR=1.017), life satisfaction (OR=2.150), and regional distribution (OR=1.275) were associated with depression(P<0.05). Conclusion According to the data analysis of this study, the prevalence of depressive symptoms among middle-aged and elderly Chinese women was high, accounting for more than 40%. To control and prevent depressive symptoms to improve mental health status in this population, it is suggested to improve their education level, physical health status, family relations and life satisfaction, reduce their debt-financed consumption, and to offer them more ways to access information, and more social support, as well as to promote urbanization.
【Key words】 Depression; Middle aged; Aged; Femininity; China Health and Retirement Longitudinal Survey; Root cause analysis
Studies have shown that depression, as a common psychological disorder among middle-aged and elderly people, can contribute to a high risk of self-harm and suicide [1-2], as well as the risk of cognitive dysfunction and senile dementia [3], which not only reduces the quality of life of middle-aged and elderly people, but a big contributor to the family economic burden and national medical and health resources burden [4]. A great deal of research indicating that there is a gender gap in depression, which is more common in middle-aged and elderly women [5-7].The transition of family identity of middle-aged womenis highly consistent with the China's reform and opening up, the transition period of family and social ethics, and they are more adversely affected in life and psychology.Therefore, it is of great practical importance to understand the mental health status of middle-aged and elderly women in China, especially to explore as many influencing factors as possible.Given this, the study used the 2018 survey data of the China Health and Retirement Longitudinal Survey (CHARLS) to analyze prevalence and associated factors of depression in middle-aged and elderly Chinese women, providing evidence for exploring mental health and effective interventions in this population.
1.Objectivesand Methods
1.1. Objectives
This study was conducted from January to March 2021. Data were obtained from the fourth wave survey data of the China Health and Retirement Longitudinal Study (CHARLS, wave 4), which was released in September 2020 and involved 19816 respondents from 150 counties/districts and 450 villages/communities, with good sample representation [8]. Inclusion criteria :(1) age ≥ 45 years; (2) female; (3) respondents who clearly responded to the 10-item version of the Centre for Epidemiological Studies Depression Scale. Exclusion criteria: inability to complete the survey or refusal to participate in the survey. After removing samples with missing selected variables, a total of 7963 middle-aged and elderly women were included in the study. The CHARLS was approved by the Ethical Review Committee of Peking University. (IRB00001052-11015) andthe informed consent was signedat the time of participation with all participants.
1.2. Methods
1.2.1.The investigation content of CHARLS related to the study
The contents included demographic characteristics (age, education levels, residency), physical health status (self-reported health, physical disability, chronic disease, troubled with body pains,BADL disability, IADL disability), sociological characteristics (living with spouse/partner or not, number of family members, number of children, number of children who visit their parents at least once a month, caring for grandchildren, number of still alive parents, social activities, using the internet or not, and having care support or not when needed), economic characteristics (having jobs other than self-employed agricultural, individual income, having debt, retirement), life satisfaction, and regional distribution.For BADL, respondents were asked to answer whether they had difficulty in six activities of daily living included dressing, bathing/showering, feeding oneself,
getting in or out of bed, using the toilet, and controlling urination and defecation, while theIADL contained doing household chores, cooking, shopping, managing finances, taking medications, and using telephone calls. For both BADL and IADL, answers were categorized as: “do not have any difficulty”, “have difficulties but still can do it”, “have difficulties and help is needed”, “cannot complete it”. Those respondents who reported any difficulty in any item of BADL/IADL were defined as having BADL disability or IADL disability [9].
1.2.2.Measures of depression
The 10-item version of the Centre for Epidemiological Studies Depression Scale (CES-D-10)was used to assess depression in middle-aged and elderly women.The CES-D-10 was revised by ANDRESEN et al. [10] based on the results of item analysis to overcome the problems of long answer time, sensitive item content and high rejection rate in the original CES-D-20.The CES-D-10 scale included the following: (1) I was bothered by things that do not usually bother me. (2) I had trouble keeping my mind on what I was doing. (3) I felt depressed. (4) I felt that everything I did was an effort. (5) I felt hopeful about the future. (6) I felt fearful. (7) My sleep was restless. (8) I was happy. (9) I felt lonely. (10) I could not get “going”.By asked respondents “How often this past week did you ...” answered the above ten items, each item was scored from 0 to 3:“rarely or none of the time (less than 1 day) was scored as 0,” “some or a little of the time(1–2 days) was scored as 1,”, “occasionally or a moderate amount of the time(3–4 days) was scored as 2,” “most or all of the time (5–7 days) was scored as 3.”The responses the two positive feelings of the item 5 and the item 8 were scored as 3, 2, 1 and 0. The total CES-D 10 score ranged from 0 to 30, respondents were classified as scores 10-30 being having depression symptoms and those with scores 0-9 as being without depression symptoms. With a higher score indicating a greater severity of depressive symptoms [10]. The Cronbach’s alpha of CES-D-10 was 0.788[11].
1.3. Statistical analysis
Statistical analyses were performed using Stata version 14.1 software. The measurement data that did not conform to normal distribution were described by M (P25, P75), and counting data were described in relative numbers. Robust OLS regression, Tobit regression and Logit regression were used to analyze the influencing factors of depression in middle-aged and elderly women. Two sided P<0.05 was considered as statistically significant.
2. Results
2.1. The characteristics of the sample
Among 7963 middle-aged and elderly women, 4131 (51.88%) were aged from 45 to 59, and 3832 (48.12%) were aged 60 and above. Other demographic characteristics, physical health status, sociological characteristics, economic characteristics, life satisfaction and regional distribution are shown in Table 1.
Table 1. Characteristics of 7963 middle-aged and elderly women.

Note: BADL, basic activities of daily living; IADL, instrumental activities of daily living; The number of family members, number of children, number of children who visit their parents at least once a month, the number of still alive parents (including foster parents, father, mother, father-in-law, mother-in-law), individual income, and having debt were recorded as continuous variables and not listed in the table 1.
2.2. Prevalence of depression in middle-aged and elderly women
Among 7963 middle-aged and elderly women, the median CES-D-10 score of the participants was 8(4, 14)points. Except for 4490 cases (56.39%), the remaining 3473 cases (43.61%) were found with depressive symptoms, including 1715(41.52%) aged 45-59 years, and 1758(45.88%) aged 60 or over.
2.3. Regression analysis of the influencing factors of depression in middle-aged and elderly women
Based on the analysis of relevant literature[12-13], we selected demographic characteristics, physical health status, sociological and economic characteristics, life satisfaction and regional distribution of middle-aged and elderly women as independent variables. For age, the quadratic term of age was introduced in regression according to existing literature[14].
The CES-D-10 scoreof middle-aged and elderly women wasseen as dependent variables, and the above independent variables were included for OLS regression analysis (the assignments of variables used in the study are all shown in Table 2).The multi-collinearity test was performed first, and it was found that the maximum VIF of each variable except age and its quadratic term was 1.88, indicating that there was no multi-collinearity problem.Then,heteroscedasticity test was carried out and it was found that there was heteroscedasticity, so robust OLS regression was used. The result of the robust OLS regression analysis showed that age and age squared, education level, residency, self-reported health, physical disability, chronic disease, troubled with body pains, BADL disability, IADL disability, living with spouse (partner) or not, number of family members, number of children, using the internet or not, having care support or not when needed, having jobs other than self-employed agricultural or not, individual income, having debt, life satisfaction, and regional distribution were correlated with depression in middle-aged and elderly women (P<0.05).
Table 2. Assignment of variables possibly associated with depression among middle-aged and elderly women.

Note: CES-D, Center for Epidemiologic Studies Depression Scale; BADL, basic activities of daily living; IADL, instrumental activities of daily living.
The total CES-D-10 score of 10 items ranged from 0 to 30, and does not conform to normal distribution, OLS regression may produce estimation errors, so the dependent variables and independent variables are included for further analysis in robust Tobit regression, and the results show that age and age squared, education level, residency, self-reported health, physical disability, chronic disease, troubled with body pains, BADL disability, IADL disability, living with spouse (partner) or not, number of family members, number of children, using the internet or not, having care support or not when needed, having jobs other than self-employed agricultural or not, having debt, life satisfaction, and regional distribution were associated with depression in middle-aged and elderly women (P<0.05).
With or without depressive symptoms as dependent variables, the above independent variables were included for robust Logit regression. The results indicated that age and age squared, education level, residency, self-reported health, physical disability, chronic disease, troubled with body pains, BADL disability, IADL disability, living with spouse (partner) or not, number of children, using the internet or not, having care support or not when needed, having debt, life satisfaction, and regional distribution have significant impact on depression in middle-aged and elderly women (P<0.05) (Table 3).
The results of the three regression models showed that age and age squared, education level, living in rural or urban areas, self-rated health, physical disability, chronic disease, troubled with body pains, BADL disability, IADL disability, living with spouse (partner) or not, number of children, using the internet or not, having care support or not when needed, having debt, life satisfaction, and regional distribution were associated with depression.
Table 3. Robust OLS regression, robust Tobit regression and robust Logit regression analyses of influencing factors possibly associated with depression among middle-aged and elderly women.

Note: BADL, basic activities of daily living; IADL, instrumental activities of daily living; a, Chi-square statistic.
3. Discussion
Previous studies have found that the rate ofdepression for women was higher than that for men,and the prevalence of depressive symptoms in middle-aged and elderly women in China was
43.2% [15].The results of this study showed that middle-aged and elderly women with depressive symptoms accounted for 43.61% (3473/7963), which was close to the above results.Demographic variables, including age, education levels and residency, can affect the depressive symptoms of middle-aged and elderly women. Among them, the influence of age on depressive symptoms was relatively complex, asthe coefficient of age’s level valuewas significantly positive and coefficient of age squaredwas significantly negative, which was in the shape of inverted U-shaped parabola, and the peak age of onset was 52 years old.This was similar to the results of relevant domestic studies, depression symptoms firstworsened and then alleviated with age, and the peak age of onset was between 50 and 60 years old[14].One study reported that higher education level reduced the correlation between social isolation and depressive symptoms in men, but not in women [16]. However, the results of this study showed that the higher level of education, the lower the risk of depression symptoms in middle-aged and elderly women, which was consistent with the results of Li J S et al. [14].The results of our study showed that compared with living in rural areas, middle-aged and elderly women living in urban areas had a lower risk of developing depressive symptoms, which was similar to the results of HE et al. [17],Kong XK et al. [18]. The improvement of the level of urbanization significantly reduced the rate of depression.
Results indicated that physical health status was related to depressive symptoms in middle-aged and elderly women, including poor self-rated health, physical disability, chronic disease,troubled with body pains, BADL disability, IADL disability, which werethe influencing factorsfor depression symptoms in those population, supported by relevant research findings [19-21].The results suggestedthat medical workers should attach great importance to the physical health of those population and minimize the negative impact of physical illness on mental health.
The results of the study showed that the number of children was associated withdepressive symptoms in middle-aged and elderly women, similar to previous literature [22]. The more the number of children, the higher the risk of depression symptoms they were.Thereasons why middle-aged and elderly females were more affected by depression symptomscould be attributed to two aspects: on the one hand, the large number of childrenincreases the cost of living, education and healthcare for middle-aged and elderly women, leading to a heavier economic burden.On the other hand, employment, marriage and other problems of multiple children may also increase the psychological burden of middle-aged and elderly women, and then lead to developing depression.In contrast, if living togetherwith a spouse or a partner, the symptoms of depression can be reduced. A spouse or a partner can take careeach other andsolve some tough problems together,especially when it comes to the children.Our study found that surfing the internet or not was also associated with depressive symptoms in middle-aged and elderly women, and the reason was that surfing the internet increased access to information and interpersonal communication. Our results also indicated thathaving care support or not when needed was related to depressive symptoms in middle-aged and older women. When they without care support from family members, relatives and friendswhen needed, depressive symptoms were evident, which may be related to the lack of relevant social support.This result pushed urgent requirements for us to concern overthe mental health status of the special groups and the government to improve the elder social security system.
The results of this study also showed that having debt was associated withdepressive symptoms in middle-aged and elderly women.Over-consumption and debt management have gradually become a common economic phenomenon, but the modern financial consciousness has
not followed up, resulting in middle-aged and elderly women have more psychological pressure for debt. Life satisfaction was correlated with depression symptoms in middle-aged and elderly women, which was confirmed in previous studies[23], indicating thatpeople who were less satisfied with their lives had more negative psychological feelings and were more prone to be depressed. Moreover,as confirmedin several studies [24-25],there was an imbalance in the regional distribution of depression symptoms in China, which may be closely related to the level of economic development among regions. Much can be doneby relevant institutions and departments to narrow the regional differences between middle-aged and elderly women, improve the level of social equity, and promote the healthy development of mental healthof this populationin different regions.
To sum up, according to the data analysis of this study, the prevalence of depressive symptoms among middle-aged and elderly Chinese women was high, accounting for more than 40%. To control and prevent depressive symptoms to improve mental health status in this population, it is suggested to improve their education level, physical health status, family relations and life satisfaction, reduce their debt-financed consumption, and to offer them more ways to access information, and more social support, as well as to promote urbanization.
Acknowledgments
The authors would like to thank the Institute of Social Science Survey of Peking University for their organizing of CHARLS, and all the participants, investigators and assistants of CHARLS.
Author Contributions
All authors have approved the final manuscript.
Declaration of Competing Interest
None.
References
[1] Yi S W. Depressive symptoms on the geriatric depression scale and suicide deaths in older middle-aged men: a prospective cohort study[J]. J Prev Med Public Health, 2016,49(3):176-82. DOI:10.3961/jpmph.16.012.
[2] FERRARI A J, SOMERVILLE A J, BAXTER A J, et al. Global variation in the prevalence and incidence of major depressive disorder: a systematic review of the epidemiological literature[J]. Psychol Med, 2013,43(3):471-81. DOI: 10.1017/S0033291712001511.
[3] HESER K, TEBARTH F, WIESE B, et al. Age of major depression onset, depressive symptoms, and risk for subsequent dementia: results of the German study on Ageing, Cognition, and Dementia in Primary Care Patients (AgeCoDe)[J]. Psychol Med, 2013,43(8):1597-610. DOI: 10.1017/S0033291712002449.
[4] LI Z, ZHEN W, MAO Z F. Depression and its influencing factors of the elderly in a community, Wuhan[J]. Modern Preventive Medicine, 2018,45(1):102-105.
[5] ZHANG W J, WEI M. The evaluation of the mortality and life expectancy of Chinese population[J]. Population Journal, 2016,38(3):18-28. DOI: 10.16405/j.cnki.1004-129X.2016.03.002.
[6] ZHOU Z, PAN L L,FANG Y. Associationbetween psychosocial factors and risk of depressive symptoms in the elderly [J]. Chinese Journal of Gerontology, 2019,39(16):4092-4094. DOI: 10.3969/j.issn.1005-9202.2019.16.071.
[7] WU S, ZHANG F Y, ZANG Z Y, et al. Analysis on depression and life satisfaction of the elderly in China and influencing factors[J]. J Zhengzhou Univ: MedSci, 2019,54(1):88-92. DOI: 10.13705/j.issn.1671-6825.2018.06.157.
[8] LIU S Y, QIAO Y N, WU Y, et al. The longitudinal relation between depressive symptoms and change in self-rated health: a nationwide cohort study[J]. J Psychiatr Res, 2021,136:217-223. DOI:10.1016/j.jpsychires. 2021.02.039.
[9] LI M M, FU Y N, WU M C, et al. Correlation between activities of daily living and depression in the elderly[J]. Modern Preventive Medicine, 2017,44(21): 3957-3961.
[10] ANDRESEN E M, MALMGREN J A, CARTER W B, et al. Screening for depression in well older adults: evaluation of a short form of the CES-D[J]. Am J Prev Med, 1994,10(2):77-84. DOI: 10.1016/S0749-3797(18)30622-6.
[11] CHEN J F, FANG M W,XIAO C H, et al. Activities of daily living and depressive symptoms in the Chinese elderly[J]. Chinese General Practice, 2020,23(22):2852-2855. DOI: 10.12114/j.issn.1007-9572. 2019.00.693.
[12] LI L, MA M Y,PENG H Y, et al. Prevalence and associated factors of depressive symptoms in China's rural elderly[J]. Chinese General Practice, 2021,24(27):3432-3438. DOI: 10.12114/j.issn.1007-9572.2021.00.577.
[13] LIU H, FAN X, LUO H, et al. Comparison of depressive symptoms and its influencing factors among the elderly in urban and rural areas: evidence from the China Health and Retirement Longitudinal Study (CHARLS)[J]. Int J Environ Res Public Health, 2021,18(8):3886. DOI: 10.3390/ijerph18083886.
[14] LI J S, MA W J. Prevalence and influencing factors of depression symptom among middle-aged and elderly people in China[J]. Chin J Public Heal, 2017,33(2):177-181. DOI: 10.11847/zgggws2017-33-02-02.
[15] LI Y, HE Y, ZHAO L, et al. Sociodemographic disparity of the depression prevalence in China: findings from the China Health and Retirement Longitudinal study[J].Int J Gerontol, 2019,13(1):33-37. DOI: 10.6890/IJGE.201903_13(1).0007.
[16] LUO F, GUO L, THAPA A, et al. Social isolation and depression onset among middle-aged and older adults in China: Moderating effects of education and gender differences[J]. J Affect Disord, 2021,283:71-76. DOI: 10.1016/j.jad.2021.01.022.
[17] HE S, SONG D, JIAN W Y. The association between urbanization and depression among the middle-aged and elderly: a longitudinal study in China[J].Inquiry, 2020,57:1-9. DOI: 10.1177/0046958020965470.
[18] KONG X K, XIAO Q L, LI J. Urban-rural comparison on the risk factors of geriatric depressive symptoms[J]. Chin Ment Health J, 2018,32(8):648-655. DOI: 10.3969/j.issn.1000-6729.2018.08.005.
[19] JIANG C H, ZHU F, QIN T T. Relationships between chronic diseases and depression among middle-aged and elderly people in China: a prospective study from CHARLS[J]. Curr Med
Sci, 2020,40(5):858-870. DOI:10.1007/s11596-020-2270-5.
[20] WANG X H, ZHAO S X, CAO J, et al. Correlation between depression and low back pain of elderly women in rural areas of China[J].Chinese Journal of Disease Control & Prevention, 2020,24(10):1215-1218. DOI:10.16462/j.cnki.zhjbkz.2020.10.020.
[21] HE M, MA J, REN Z, et al. Association between activities of daily living disability and depression symptoms of middle-aged and older Chinese adults and their spouses:a community based study[J]. J Affect Disord, 2019,242:135-142. DOI:10.1016/j.jad.2018.08.060.
[22] HU Z, WU Y Y, YANG H L, et al.Effects of fertility behaviors on depression among the elderly: empirical evidence from China[J]. Front Public Health, 2021,8:570832. DOI:10.3389/fpubh.2020.570832.
[23] LEE S W, CHOI J S, LEE M. Life satisfaction and depression in the oldest old: a longitudinal study[J]. Int J Aging Hum Dev, 2020,91(1): 37-59. DOI:10.1177/0091415019843448.
[24] TAO H W, ZHANG X, WANG Z. The eastern-middle-western depression and the determinants among Chinese rural elderly[J]. Chinese Journal of Disease Control & Prevention, 2018,22(7):696-699. DOI: 10.16462/j.cnki.zhjbkz.2018.07.010.
[25] WANG L, ZHANG X. The sexual disparity and determinants of depressive symptoms among the rural elderly in China[J]. Chinese Journal of Disease Control & Prevention, 2018,22(11):1148-1151. DOI: 10.16462/j.cnki.zhjbkz.2018.11.013.
(Received: 6 June 2021; Revised: 27 August 2021)

Key words: Depression, Middle aged, Aged, Femininity, China Health and Retirement Longitudinal Survey, Root cause analysis