Older Parents Benefit More in Health Outcome From Daughters’ Than Sons’ Emotional Care in China

Abstract

Objective: To examine whether older parents in China would benefit more from daughters’ care than from sons’ emotional care. Method: Analysis of the unique data sets of the Chinese Longitudinal Healthy Longevity Survey conducted in 2002, 2005, and 2008-2009 in 22 provinces. Results: As compared with having son(s), having daughter(s) is significantly more beneficial at older ages in China, with regard to maintaining higher cognitive capacity and reducing mortality risk. Such daughter advantages in providing emotional care to older parents are more profound among the oldest-old aged 80+ as compared with the young-old aged 65 to 79 and surprisingly more profound in rural areas as compared with urban areas, even though son preference is much more common among rural residents. Discussion: We describe how educational campaigns aimed at informing the public about the benefits of daughter(s) for older parents’ health outcome could help promote gender equality and reduce traditional son preference, especially in rural China.

Keywords

caregiving China cognitive function mortality healthy aging

Introduction

For thousands of years, both cultural tradition and economic benefits have led to the continued preference for sons in China, especial in rural areas. According to Confucian customs, only sons can perform ancestor worship and continue the family line (Ebenstein & Leung, 2010), which is one of the main reasons why most people in rural China strongly prefer to have at least one son. The country’s underdeveloped social welfare system for the elderly, particularly in rural areas, also contributes to son preference. The sentiment behind the old Chinese saying “Yang Er Fang Lao”—having a son prevents difficulties in old ages—is a belief held by many Chinese and a driving force behind son preference and discrimination against girls and women, especially in rural China.

Chinese culture is traditionally patriarchal, patrilineal, and patrilocal—adult sons are more likely to co-reside with their elderly parents and provide financial support, whereas daughters “marry out” and become members of their husband’s family (Bian, Logan, & Bian, 1998; Deutsch, 2006). Rural household labor supply also favors sons as they are better farm laborers and can also enhance family power in local conflicts (Ebenstein & Leung, 2010). Although daughters are less available to provide financial support to their own aged parents than sons, however, they can still provide instrumental and emotional care and support to natal parents. Such supports and their implications for older parents’ health outcomes are rarely documented using large-scale nationwide surveys from China, and this article aims to fill in this research gap. Based on data from three waves of the Chinese Longitudinal Healthy Longevity Survey (CLHLS), we explore whether elderly parents’ benefit from emotional support from adult children and how it may differ by the adult child’s gender. In the remaining part of this introductory section, we will briefly review previous studies related to our research questions.

Caregiving and Emotional Support to Elderly Parents From Adult Children

Given the patriarchal nature of Chinese families, caregiving from sons generally means that the actual hands-on care to aged parents in need is provided by sons’ wives—daughters-in-law of elderly parents, whereas daughters usually marry out (Liu & Kendig, 2000). In recent decades, the strengthening of the nuclear family (as opposed to the extended family), free-choice marriage, and an increase in women’s status may have led to stronger bonds between sons and their wives. Consequently, daughters-in-law are more involved with decision making and have more power within the family, with husbands and wives negotiating elderly parents care arrangements in contemporary China (Zhang & Wang, 2010). Such trends might increase conflicts in daily life between daughters-in-law and mothers-in-law, which are much more likely to occur, or even to be serious, as compared with daughter–mother relation, especially in rural areas (Meng, 2002; Pearson, 2002). As men are less likely to be caregivers than women, there is much lower potential for conflict between older parents and sons-in-law (Chappell & Kusch, 2007).

The general perception that sons provide more care to older parents in the context of Chinese tradition contrasts with the Western cultural and social practices, which have been increasingly influencing Chinese society since the late 1970s when China opened its door to the world. In Western countries, daughters are more likely than sons to provide daily care to elderly parents (Raley & Bianchi, 2006; Spitze & Logan, 1990), particularly personal care and household tasks (Abel, 1990; Miller & Cafasso, 1992). Some studies have shown that daughters (in other countries and China) are more likely to be actual caregivers than sons when the older parents are sick (Spitze & Logan, 1990; Whyte, 2003).

Survey data from urban China have found that sons are more likely to provide financial support to their older parents, whereas daughters are more likely to provide gifts and assistance in daily living to their older parents (Cooney & Di, 1999; Sun, 2002; Zhan, 2004). In a study of entry into long-term care in Shanghai, daughters’ availability to provide care mattered more than sons, a departure from traditional filial piety that a son should be the main caregiver (Chen & Ye, 2013). In urban China, patrilocality may be declining, with the importance of sons (for old-age care) diminishing. Urban parents may be more willing to receive support from daughters, but reliance on sons (and daughters-in-law) persists in rural areas, where daughters’ care is voluntary, and they are publically perceived as only a “back-up” option if no sons are available (Zhang & Wang, 2010). Xie and Zhu (2009) found that elderly parents who have daughters but no sons likely co-reside with a daughter and son-in-law, especially in rural China, and these only-daughters are emotionally close to parents. One recent study finds that the oldest-old in urban China who were disabled in activities of daily living (ADL) were 35% to 37% more likely to report an unmet need of help, if the primary caregiver are sons/daughters-in-law, compared with urban oldest-old whose primary caregiver is daughters/sons-in-law (Zhu, 2015). Yet, many studies under recognize the role of daughters in caring for impaired older parents (Zhan, 2004).

Although there are fewer studies of emotional support (as opposed to financial support), evidence from the United States found that daughters provide more emotional support than sons to both fathers and mothers (Chesley & Poppie, 2009). Research from the Netherlands found that daughters give greater emotional support to elderly mothers than do sons when the mothers experience a health decline (Broese Van Groenou &Knipscheer, 1999). Research found that emotional support was stronger from daughters than from sons in Hong Kong (Ng, Phillips, & Lee, 2002) and in rural Mainland China (Cong & Silverstein, 2012). This is consistent with the idea that daughters are kin-keepers of the family and more responsive to the needs of their older parents (Li & Seltzer, 2005; Rossi & Rossi, 1990). Research from rural northern China also provides evidence of elderly parents’ happier relationships with daughters than with sons (Zhang & Wang, 2010). A recent study, based on 4,479 older parent–adult child dyads interviewed in 2002 in China as part of the CLHLS, provided strong evidence that, controlling for various confounders of both adult children and elderly parents, adult daughters’ filial piety index is significantly higher than that of sons; in addition, adult daughters’ emotional relationships with older parents are significantly better than that of sons (Zeng, George, Brasher, Gu, & Vaupel, 2015).

Association Between Support From Adult Children (Sons vs. Daughters) and Older Parents’ Health Outcomes

A large literature demonstrates the importance of children’s support for elderly parents’ well-being (Silverstein & Bengtson, 1994; Zunzunegui, Beland, & Otero, 2001). In general, support from adult children has protective effects for elderly parents’ health and mortality. In a Spanish study, older parents who received emotional support from children were more likely to have good self-rated health (Zunzunegui et al., 2001). Several studies from diverse populations (Europe, United States, Mainland China, and Taiwan) have found that support from children reduces older parents’ depressive symptoms (Buber & Engelhardt, 2008; Byers, Levy, Allore, Bruce, & Kasl, 2008; Li, Liang, Toler, & Gu, 2005). A study from Taiwan found that older adults with daily activity limitations who shared their worries with adult children had fewer depressive symptoms (Weinstein, Glei, Yamazaki, & Chang, 2004). A U.S.-based study found that parents had reduced mortality risk when they had more affectionate relationships with adult children (Silverstein & Bengtson, 1991).

Previous research demonstrated that the amount and type of support from children may vary by adult child’s sex and cultural context (Silverstein & Bengtson, 1994; Zunzunegui et al., 2001). In recent years, a number of studies have examined associations of sex composition of children and older parent’s mortality and health (Cesarini, Lindqvist, & Wallace, 2009; Hurt, Ronsmans, & Quigley, 2006). One recent study found no protective effect of sons on elderly parents’ survival in either Mainland China or Taiwan, and it also found that, in Taiwan, daughters may have been more beneficial than sons in reducing older parents’ mortality in recent years (Pham-Kanter &Goldman, 2011). Similarly, research from rural Poland found that a higher number of daughters increased fathers’ life span. This same study found that greater numbers of children (of either sex) reduce maternal longevity (Jasienska, Nenko, & Jasienski, 2006). A study from rural Bangladesh found no effect of number of daughters on either maternal or paternal mortality, but greater number of sons increased mortality for parents (Hurt et al., 2006).

Research Questions to be Addressed

Based on the literature reviewed above and the relevant unique data we have in hand, we intend to empirically address the following specific research question:

Research Question: In regards to health outcomes, do older parents benefit more from emotional care provided by daughters (and sons-in-law) than that provided by sons (and daughters-in-law)?

To address this research question, we will explore how caregiving received from adult daughters (and sons-in-law) versus sons (and daughters-in-law) and gender composition of children may be associated with older parents’ health indicators and mortality risk. The health indicators of older parents to be examined in this study include changes in cognitive function, ADL, self-rated health, and deficit index (see next section for details).

The oldest-old most likely need care and have been growing at a much faster rate than the young-old, and the rural–urban gap of old-age care is dramatically large in China (Zeng & George, 2010). Yet, no previous studies have compared the oldest-old and young-old parents, or differences in rural and urban residents, with respect to how sex composition of children and care from daughters versus sons affects older parents’ health outcomes. We will try to fill this research gap in this article. In addition, based on presentation and discussion of the results of the empirical analysis, we will discuss related policy recommendations.

Data Source, Measures, and Method

Data Source

The data used in this article are from the third, fourth, and fifth waves of the CLHLS, conducted in 2002, 2005, and 2008-2009, with a total sample size of 19,697 respondents aged 65+ from these three waves of CLHLS (see Table 1 for more details). The CLHLS baseline survey was conducted in 1998, and all of its follow-up surveys include re-interviews with survival participants. In the first four follow-up surveys conducted in 2000, 2002, 2005, and 2008-2009, CLHLS also included replacement for deceased and lost-to-follow-up elders. The CLHLS has been conducted in randomly selected, about half of the counties and cities, in 22 of the 31 provinces and municipalities of China (Liaoning, Jilin, Heilongjiang, Hebei, Beijing, Tianjin, Shanxi, Shaanxi, Shanghai, Jiangsu, Zhejiang, Anhui, Fujian, Jiangxi, Shandong, Henan, Hubei, Hunan, Guangdong, Guangxi, Sichuan, and Chongqing).¹ The survey areas covered 985 million persons in the 1998 baseline year, which was about 85% of the total population of China. The 1998 baseline and 2000 follow-up surveys included oldest-old interviewees aged 80+ and did not include younger elders aged 65 to 79. The CLHLS has been expanded in the 2002 and later waves to cover ages 65 and above through the addition of a new sub-sample of interviewees aged 65 to 79. The CLHLS adopted a targeted random-sample design to insure the representativeness, even distribution and large enough sub-sample size of the oldest-old, aged 80 to 110, with appropriately compatible young-old, aged 65 to 79. More specifically, the CLHLS baseline survey tried to interview all centenarians who voluntarily agreed to participate in the study in the sampled counties and cities; for each centenarian interviewee, the CLHLS interviewed one randomly selected nearby octogenarian and one randomly selected nearby nonagenarian of predefined age and sex. In the 2002, 2005, and 2008-2009 waves, the CLHLS interviewed approximately three randomly selected nearby elders aged 65 to 79 of predefined age and sex in conjunction with every two centenarians. The face-to-face questionnaire data were collected in the CLHLS using internationally standardized questionnaires adapted to the Chinese cultural and social context. Information about the date of death and health status of the elderly who were interviewed in the previous wave but died before the subsequent survey were collected by interviewing a close family member (Zeng & Gu, 2008).

Table 1.

Descriptive Statistics.

	Ages 65+	Ages 65-79	Ages 80+	Urban	Rural
Total sample size of the participants	n = 19,697	n = 5,514	n = 14,183	n = 8,432	n = 11,265
Sample size of cognitively unimpaired participants	n = 11,475	n = 6,354	n = 5,121	n = 5,136	n = 6,339
Composition of living children (%)
No living children	10.3	3.8	12.9	11.9	9.2
Having 1+ daughter(s) only	11.8	5.8	14.1	12.8	11.0
Having 1+ son(s) only	14.6	12.3	15.4	14.0	15.0
Having both son(s) and daughter(s)	63.3	78.1	57.6	61.3	64.8
% of deceased respondents 2002-2008	55.9	19.6	70.0	54.8	56.8
Shares feelings with most frequently (%)
Spouse	23.8	50.1	12.4	25.4	22.5
Son and daughter-in-law	36.1	21.8	42.3	29.3	41.8
Daughter and son-in-law	8.8	5.9	10.1	12.0	6.2
Son and daughter	6.2	5.8	6.5	7.2	5.4
Other family member	6.9	2.0	9.0	6.1	7.6
Other	10.91	9.5	11.5	12.8	9.3
Control variables
M age	86.6	72.0	92.7	86.5	86.5
% of Han ethnicity	93.8	94.1	93.7	95.6	92.5
% currently married	28.3	62.2	15.1	29.9	27.1
% having 1+ years of schooling	27.2	37.9	23.1	30.0	25.1
% with economic independence	19.2	32.5	14.1	35.8	6.8
M number of living children	3.2	4.0	2.9	3.1	3.4
% of close proximity to children	66.5	75.9	62.9	51.4	77.9
% of currently smokes	18.5	28.4	14.6	16.7	19.8
% of currently drinks alcohol	20.6	24.6	19.1	19.0	21.9
% of currently does regular exercise	29.7	39.2	25.9	39.4	22.4
M value of Deficit Index	0.26	0.11	0.30	0.27	0.26

The systematic assessments on data quality concerning accuracy of age reporting, reliability, validity, consistency of the main measures, and randomness of attrition show reasonably good quality in the CLHLS data sets (Goodkind, 2009; Gu & Dupre, 2008; Zeng, 2012). The unique CLHLS data sets have been publicly available and widely applied in healthy aging studies by scholars around the world. For more information of the technical documents and the list of publications using the CLHLS data sets, refer to http://centerforaging.duke.edu/chinese-longitudinal-healthy-longevity-survey

Measures

The data set used in this study include 19,696 older parents aged 65+ who were interviewed in 2002 or 2005 and their follow-up data collected in 2005 or 2008-2009 (see Table 1 for age and rural/urban sample compositions).

Dependent variables

Changes in cognitive function of older parents. The cognitive function of the elderly interviewees in the CLHLS was screened by the Chinese version of the Mini-Mental State Examination (MMSE), which has the same contents as the international standard of the MMSE Questionnaire (Folstein, Folstein, & McHugh, 1975), but the CLHLS team culturally translated and adapted it into the Chinese language. The CLHLS team did careful pilot interview tests to validate the Chinese version of MMSE before the surveys were conducted. The total possible score on the MMSE is 30, with lower scores indicating poorer cognitive ability. We defined those who had a score of less than 24 as cognitively impaired as suggested by Folstein et al. (1975) and widely used internationally and in the Chinese studies. Previous publications concerning cognitive function using the CLHLS data confirmed that such a cutoff is valid (Shen & Zeng, 2010; Yin et al., 2012). Cognitive capacity decline was defined as when a person whose score was greater than 24 in 2002 or 2005 but had a score less than 24 in 2005 or 2008-2009.

ADL. If an older parent needed help to perform any of the six daily tasks of bathing, dressing, indoor transferring, toileting, eating, and incontinence, he or she was considered ADL disabled and needed daily care; otherwise, he or she was not disabled and did not need daily care.

Self-rated health. The CLHLS asked the question “How do you rate your health at present?” to each interviewee, and no proxy answers were allowed. The possible answers to this question were grouped into two categories of “good” and “not good.”

The deficits index (DI; previously called the frailty index), which has been widely applied in healthy aging studies, is a summation of deficits in variables reflecting different dimensions of health conditions. Following the general international practice (Goggins, Woo, Sham, & Ho, 2005; Kulminski et al., 2008), we estimated the DI for each respondent from 39 variables, with a score of 1 if the condition was present and a score of 2 if a serious illness caused the person to be bedridden or hospitalized. These variables included cognitive function, ADL, instrumental ADL, physical performance, self-rated health, interviewer-reported health, hearing and vision loss, heart rhythm, psychological distress, serious illness in the past 2 years, and specific chronic diseases. We then constructed a DI by summarizing all deficits and dividing by the total number of possible deficits. We treat DI scores as a continuous variable.

Mortality risk. As described in the “Data Source” section above, information on date of death were collected for the interviewees who were interviewed in 2002 or 2005, but died in the inter-wave period 2002-2005 or 2005-2008/2009. Survival time was entered as days counted from the date of the interview in 2002 or 2005 to the date of the interview in 2005 or 2008-2009 for survivors and to the date at death for the deceased, while we controlled for respondent’s initial age in 2002 or 2005.

The key independent variables of interests

The key independent variables of interest are gender composition of surviving children (having daughter(s) only vs. having son(s) only and having daughter(s) only vs. having both son(s) and daughter(s)), as well as with whom (daughter and son-in-law vs. son and daughter-in-law) the parent shares feelings with most frequently.

The variable “with whom the parent shares feelings most frequently” is based on the question “To whom do you talk first when you need to share something of your thoughts?” The possible answers (choose one) are spouse, son, daughter, daughter-in-law, son-in-law, grandchildren and their spouses, other relatives, friends/neighbors, social workers, housekeeper, and nobody. Those who answered daughter or son-in-law were combined into one category, and those who answered son or daughter-in-law were combined into one category. This is because daughter and son-in-law are a married couple and son and daughter-in-law are a married couple, and a married couple sharefamilial responsibility, financial interests, and decision making. We do so also to avoid unnecessary complication with too many categories. We focus on comparing the odds ratios (ORs) of daughter and son-in-law versus son and daughter-in-law in this article. In Chinese society, an older parent usually talks first with the adult child he/she trusts the most, or who is the most available when the parent needs to share something of his or her thoughts. Therefore, we regard “sharing feeling with daughter/son-in-law or son/daughter-in-law most frequently” as one of the indicators of emotional support from adult children.

Potentially confounding variables

Potentially confounding variables are the elderly respondents’ gender, single year of age, urban/rural residence (defined by the respondents’ actual long-term residential locations rather than the strict Chinese household registration [hukou] system), number of living children, proximity to children (if a respondent had a co-resident child or had a child living in the same village/district and received frequent visits at the time of an interview, he or she was coded as “close proximity to children”), ethnicity, education, economic status, marital status, and health practices at their interviews in 2002 or 2005. For the survival analysis using mortality risk as the dependent variable, we also controlled for respondents’ health status measured by the deficit index at their interviews in 2002 or 2005.

The frequency distributions of the sample by the dependent variables, key independent variables, and the potentially confounding variables are listed in Table 1.

Statistical Methods

In seeking answers to our research questions based on the data sets from the CLHLS 2002, 2005, and 2008-2009 waves, the statistical methods of logistic regression and parametric Weibull hazards survival analysis were used. More specifically, we used binary logistic regression models to predict follow-up declining capacities in cognitive function, ADL disability, and self-rated health, with key independent variables of gender composition of surviving children and sharing feelings with daughter (and son-in-law) versus son (and daughter-in-law), controlling for other potential confounding variables. We used binary and ordered logistic regressions as well as linear regressions to explore the effects of gender composition of surviving children and sharing feeling with children on older parents’ deficit index, adjusting for other potential confounding variables. Because the proportionality assumption in Cox proportional hazards regressions was violated in some covariates (e.g., education and economic status) and Weibull survival analysis yielded the lowest Akaike Information Criterion (AIC) index among other parametric survival regressions, we use Weibull hazards survival analysis model to address how gender composition of living children and sharing feelings most frequently with daughter (and son-in-law) versus son (and daughter-in-law) may affect older parents’ subsequent mortality. Following a similar strategy adopted in other studies (Crimmins, Hayward, & Saito, 1994), those first interviewed in 2002 or 2005 as well as their follow-up data are pooled in the present study to increase statistical precision and produce more robust results. Note that each person has only one observation in the analysis and thus there are no intrapersonal cluster effects. We tried the analysis controlling for survey wave, and it did not substantively change the results, as the 2002 and 2005 waves were close to each other and thus there was no need to include survey wave as a covariate. As the proportion of missing value of each of the variables in this study is less than 2%, we imputed them with the mode if the variables were categorical or binary and with the means if they were continuous. Other alternative strategies were also tried, and the results were almost identical.

Results

The Daughters’ Better Emotional Care May Help Older Parents to Have a Lower Risk of Follow-Up Decline in Cognitive Capacity

Table 2 presents the estimates of the ORs of follow-up decline in cognitive capacity comparing those elderly who most frequently share feelings with a daughter (and son-in-law) in daily life versus those who most frequently share feelings with a son (and daughter-in-law), as well as comparing the parents with different gender composition of surviving children. The estimates in row A of Table 2 show that as compared with those elderly who most frequently share feelings with a son (and daughter-in-law) in daily life, the elderly aged 65+ who most frequently share feelings with a daughter (and son-in-law) had significantly lower risk (p< .05; OR = 0.84) of declining cognitive capacity in the 3-year period. Such daughter advantage of providing better emotional care in preventing older parents’ decline in cognitive capacity is much stronger in rural areas (p< .05; OR = 0.68) than in urban areas (not statistically significant; OR = 1.0). The estimates of ORs indicate that the risk of follow-up decline in cognitive capacity of older parents who have daughter(s) only was smaller than those who have sons only (see row B of Table 2) and smaller than those who have both sons and daughters (see row C of Table 2). However, none of these estimates are statistically significant.

Table 2.

Odds Ratios of Follow-Up Decline in Cognitive Capacity by With Whom Elderly Parent Shares Feelings in Daily Life and Gender Composition of Living Adult Children, Based on Data of 2002-2008 CLHLS.

	Ages 65+ (n = 11,475) [95% CI]	Ages 65-79 (n = 6,354) [95% CI]	Ages 80+ (n = 5,121) [95% CI]	Urban (n = 5,136) [95% CI]	Rural (n = 6,339) [95% CI]
A. Shares feelings most frequently with daughter (and son-in-law) vs. son (and daughter-in-law)	0.84*	0.67*	0.91	1.00	0.68*
	[0.71, 0.98]	[0.47, 0.94]	[0.76, 1.10]	[0.80, 1.25]	[0.54, 0.87]
B. Having daughter(s) only vs. having son(s) only	0.86	0.80	0.89	0.89	0.84
B. Having daughter(s) only vs. having son(s) only	[0.70, 1.06]	[0.53, 1.22]	[0.70, 1.13]	[0.66, 1.21]	[0.64, 1.12]
C. Having daughter(s) only vs. having both son(s) and daughter(s)	0.96	0.94	0.95	0.94	0.98
	[0.80, 1.15]	[0.64, 1.36]	[0.77, 1.17]	[0.72, 1.23]	[0.76. 1.26]

Note. All estimates are obtained from binary logistic regressions controlling for gender, single year of age, number of living children, proximity to children, urban/rural residence, ethnicity, education, economic status, marital status, and health practices (currently smoking, drinking alcohol, and doing exercise). The chi-square test results indicated that the model fits of the statistical models presented in this table are significant. CLHLS = Chinese Longitudinal Healthy Longevity Survey; CI = confidence interval.

†

p< .1. *p< .05. **p< .01. ***p< .001.

Results of the Regressions Using ADL, Self-Rated Health, and Deficit Index as Dependent Variables

Although our empirical analysis has demonstrated that the daughter advantage of providing better emotional care significantly helped older parents to have a lower risk of follow-up decline in cognitive capacity, additional analysis (data not presented here but available on request) indicated that gender composition of children had no statistically significant association with older parents’ subsequent decline in functional capacity of ADL, self-rated health, or deficit index. These results indicate that the benefit of having daughter(s) for the elderly parents may be more evident with respect to mental health, which is closely related to emotional care, rather than physical health and subjective well-being, which often demands good financial support. It has been shown that financial support could help to provide necessary facilities and a favorable living environment so that ADL disability and poor self-rated health of elders could be avoided even in the presence of a decline in physical health (Feng, Hoenig, Gu, Zeng, & Purser, 2010; Feng, Son, & Zeng, 2015; Purser et al., 2012). It is thus understandable that having daughter(s) did not result in significant benefits for older parents in their ADL capacity, subjective well-being, or deficit index, because, as compared with son(s), Chinese daughter(s) are not as advantageous in providing financial support to their elderly parents.

Daughter Advantage of Providing Better Emotional Care May Help Older Parents to Have Lower Mortality Risks

Table 3 presents the estimates of the relative risk of mortality comparing elderly parents who shared feelings most frequently with a daughter (and son-in-law) versus a son (and daughter-in-law) and also associations with the gender composition of their living children. Statistically significant reduction in the relative risk of mortality are found for those elderly aged 65+ (rural and urban combined) who shared feelings most frequently with daughter (and son-in-law; p< .05; OR = 0.93; Panel I-(A) of Table 3), as compared with the older parents who shared feelings most frequently with son (and daughter-in-law). This effect is statistically significant among the oldest-old parents of rural–urban combined (p< .01; OR = 0.92; Panel III-(A) of Table 3), but there is no significant effect among young-old parents of rural–urban combined (Panel II-(A) of Table 3).

Table 3.

Relative Mortality Risks of Elderly Parents by With Whom Elderly Parent Shares Feelings in Daily Life and Gender Composition of Living Adult Children, Based on Data of 2002-2008 CLHLS.

	Both sex combined		Rural and urban combined
	Rural [95% CI]	Urban [95% CI]	Males [95% CI]	Females [95% CI]	Two sexes [95% CI]
I. All elderly parents aged 65+	n = 11,264	n = 8,432	n = 8,275	n = 11,421	n = 19,697
I-(A) Shares feelings most frequently with daughter (and son-in-law) vs. son (and daughter-in-law)	0.87**	0.99	0.92*	0.93	0.93*
	[0.79, 0.95]	[0.91, 1.07]	[0.86, 0.99]	[0.84, 1.03]	[0.87, 0.98]
I-(B) Having daughter(s) only vs. having son(s) only	0.96	0.96	1.07	0.90**	0.96
I-(B) Having daughter(s) only vs. having son(s) only	[0.88, 1.04]	[0.88, 1.06]	[0.96, 1.20]	[0.84, 0.97]	[0.90, 1.02]
I-(C) Having daughter(s) only vs. having both son(s) and daughter(s)	0.98	0.99	1.03	0.97	0.98
	[0.92, 1.10]	[0.91, 1.07]	[0.94, 1.12]	[0.91, 1.03]	[0.93, 1.04]
II. Young-old parents aged 65-79	n = 3,233	n = 2,281	n = 2,834	n = 2,680	n = 5,514
II-(A) Shares feelings most frequently with daughter (and son-in-law) vs. son (and daughter-in-law)	1.07	1.11	1.25	1.02	1.07
	[0.76, 1.51]	[0.76, 1.60]	[0.86, 1.82]	[0.74, 1.41	[0.83, 1.37]
II-(B) Having daughter(s) only vs. having son(s) only	1.00	1.11	1.01	1.07	1.02
II-(B) Having daughter(s) only vs. having son(s) only	[0.68, 1.48]	[0.67, 1.84]	[0.68, 1.52]	[0.67, 1.70]	[0.75, 1.39]
II-(C) Having daughter(s) only vs. having both son(s) and daughter(s)	1.13	0.92	1.14	1.21	1.18
	[0.80, 1.60]	[0.61, 1.41]	[0.75, 1.73]	[0.73, 2.00]	[0.86, 1.62]
III. Oldest-old parents aged 80+	n = 8,031	n = 6,151	n = 5,441	n = 8,741	n = 14,182
III-(A) Shares feelings most frequently with daughter (and son-in-law) vs son (and daughter-in-law)	0.86**	0.98	0.91+	0.95	0.92**
	[0.78, 0.94]	[0.90, 1.06]	[0.80, 1.01]	[0.88, 1.03]	[0.87, 0.98]
III-(B) Having daughter(s) only vs. having son(s) only	0.95	0.94	1.06	0.90**	0.94+
III-(B) Having daughter(s) only vs. having son(s) only	[0.88, 1.03]	[0.85, 1.04]	[0.95, 1.18]	[0.83, 0.97]	[0.89, 1.01]
III-(C) Having daughter(s) only vs. having both son(s) and daughter(s)	1.02	0.99	1.03	0.95	0.98
	[0.96, 1.08]	[0.91, 1.07]	[0.94, 1.12]	[0.89, 1.02]	[0.93, 1.03]

Note. All estimates are obtained from Weibull hazards survival analysis model controlling for gender, single year of age, number of living children, proximity to children, urban/rural residence, ethnicity, education, economic status, marital status, health practices (currently smoking, drinking alcohol, and doing exercise), and deficit index at previous interviews. The chi-square test results indicated that the model fits of the statistical models presented in this table are significant. CLHLS = Chinese Longitudinal Healthy Longevity Survey; CI = confidence intervals.

†

p< .1. *p< .05. **p< .01. ***p< .001.

As compared with those who shared feelings most frequently with a son (and daughter-in-law), those who share feelings most frequently with a daughter (and son-in-law) had significantly lower mortality risk among the rural oldest-old aged 80+ (p< .01; OR = 0.86; III-(A) of Table 3) and for all rural elders aged 65+ (p< .01; OR = 0.87; I-(A) of Table 3), respectively, but there was no statistically significant effect among urban oldest-old and all elderly parents (Panels III-(A) and I-(A) of Table 3).

As compared with those female elderly aged 65+ who have son(s) only, those elderly mothers 65+ with daughter(s) only had significantly lower mortality risk in the follow-up periods (p< .01; OR = 0.90; I-(B) of Table 3). However, the corresponding estimates for male elderly are not statistically significant (Panel I-(B) of Table 3). It is very interesting to note that the mortality reduction associated with having daughter(s) only (vs. having son(s) only) is substantial and statistically significant among oldest-old mothers (p< .01; OR = 0.90; III-(B) of Table 3), but there are no significant effects for young-old mothers (Panel II-(B) of Table 3). As shown in Panels I-(C), II-(C), and III-(C) of Table 3, elderly (either young-old or oldest-old) who have daughter(s) only and elderly who have both son(s) and daughter(s) do not demonstrate any statistically significant differences in the estimates of the relative risk of mortality.

Discussion

Based on a unique and large data set, our analysis demonstrates that as compared with having son(s), having daughter(s) is beneficial for older parents in China, with respect to maintaining a better cognitive capacity and reducing mortality risk, while controlling for various confounding factors. We discovered that such daughter advantages of providing better emotional care are generally more profound among the oldest-old parents as compared with young-old parents and surprisingly more profound in rural areas as compared with urban areas,² even though son preference is much more prevalent among rural residents.

How to explain these findings, which contradict the commonly held son preference in China, especially in rural areas? We believe that it may be at least partly explained by gender differences in adult children’s roles in support for older parents and changes in these roles in recent years. Although adult sons usually provide more financial support to their older parents in China, especially in rural areas, adult daughters who are more filial than adult sons (Zeng et al., 2015) are likely provide better emotional care to their elderly parents who frequently need it, as shown here and in other studies (Chesley & Poppie, 2009; Cong & Silverstein, 2012; Spitze & Logan, 1990; Sun, 2002; Whyte, 2003; Zhu, 2015). For example, as compared with adult sons who are usually busy and may not like to talk with their older parents in depth, adult daughters who have better emotional relationship with elderly parents (Zeng et al., 2015) may listen more carefully and patiently. In addition, they are more likely to respond to what older parents say and to provide emotional and daily care to parents if they need help (Broese Van Groenou & Knipscheer, 1999; Spitze & Logan, 1990). Indeed, a daughter’s unique advantage in providing emotional care to their older parents in China is consistent with findings from Western societies, where daughters also provide more emotional support to parents than do sons (Chesley & Poppie, 2009; Kahn, McGill & Bianchi, 2011). This is largely because, as compared with men, women in general provide greater emotional supports to their family members and friends in the social network. This can be in large part attributed to gender socialization. Women and girls are socialized to be nurturing, family oriented, and kin-keepers in the family (Rossi & Rossi, 1990), and on a globalized planet, these gender roles may become more similar across Eastern and Western societies.

Another explanation is that the tension between daughters-in-law and mothers-in-law is the most frequent dispute in intergenerational relationships in China (Yang & Chandler, 1992), and thus it is interpretable that our empirical analyses demonstrating older parents who share feelings most frequently with daughters (and sons-in-law), compared with sons (and daughters-in-law), have significantly lower risk of cognitive capacity decline and lower mortality risk (Tables 2 and 3).

Given the fact that pension supports are generally underdeveloped, especially in rural areas, those who have no sons are very likely to live with a married daughter, and they may be better-off in receiving daily care compared with those who have son(s) only and may face the challenge of conflict between daughter-in-law and mother-in-law. These factors may lead to the elderly who have daughter(s) only being substantially better-off in terms of cognitive capacity and survival probability, compared with those who have son(s) only (Tables 2 and 3), or, at a minimum, not being worse off because of a lack of male children.

We found that, comparing the elderly who have daughter(s) only with those elderly who have both son(s) and daughter(s), there is no statistically significant difference in all of the estimates of older parents’ follow-up cognitive capacity decline and the relative risk of mortality (Panel C of Table 2 and Panels I-(C), II-(C), and III-(C) of Table 3). This is consistent with findings from the study by Pham-Kanter and Goldman (2011) using data from Mainland China and Taiwan (as reviewed earlier). Furthermore, another study using the CLHLS data demonstrated that daughters (and sons-in-law) provided more satisfactory care to their disabled older parents than sons (and daughters-in-law) did, even when both sons and daughters were available in the family (Zeng et al., 2015). One possible explanation for our and others’ findings is that the conflicts in daily life between daughter-in-law and mother-in-law are much more likely to occur, or even to be serious, as compared with daughter–mother relations (Meng, 2002; Pearson, 2002). The potential risk of conflict and poorer relationships between daughter-in-law and mother-in-law in China may counteract the effects of extra resources for those older parents who have both son(s) and daughter(s).

Compared with urban areas, the conflicts between daughters-in-law and mothers-in-law may be more likely to occur in rural areas today where the sons are more likely to be absent due to rural-to-urban labor migration, which has increased dramatically in recent years. Furthermore, the more complicated household work in rural areas may result in more frequent direct interactions and higher likelihood of conflict in daily life between co-residing daughters-in-law and mothers-in-law than that in urban areas. The generally lower education level of rural daughters-in-law and mothers-in-law may also result in poorer capability of resolving their daily life conflicts as compared with their urban counterparts. Previous studies found that the tension between daughters-in-law and mothers-in-law is the most frequent dispute in intergenerational relationships in rural China (Yang & Chhandler, 1992).

After more than 30 years of rapid economic and social development, rural daughters and their husbands can acquire more education, technical skills, and economic capacity, which may reduce their disadvantages in familial affairs and power, and reduce the gap between daughters and sons (Xu, 2001). Therefore, although peasants’ pre-dominant perception of son preference persists, the traditional advantages of sons may be relatively weaker today. Thus, it is interpretable that daughter advantages in providing better emotional care to older parents are more profound in rural areas as compared with urban areas, although son preference is still much more prevalent among rural residents, mainly because of Chinese cultural traditions and underdevelopment of the pension system in rural areas.

The oldest-old are more frail and vulnerable in cognitive capacity decline and mortality risk as age advances, and thus daughters’ better and more careful emotional care would be more beneficial and effective for the oldest-old parents than for the young-old parents. Another possible explanation is that the children of the oldest-old—who are elderly themselves or at least middle-aged—may have higher filial piety than adult children of the young-old, due to more connections to, and influence from, traditional Chinese family values of the period that they grew up in. This may enable the oldest-old to receive and benefit more and better emotional support from older children than the young-old receive from relatively younger children.

Our empirical findings that older parents may benefit more in cognition and survival from daughters’ emotional support than from sons in China, especially in rural areas and among the oldest-old, are plausible and explainable by the Chinese social context and reality today. However, such facts may still be covered up or diluted by the strong traditional son preference. Thus, we recommend that the Chinese government and society may add daughters’ providing better emotional care to elderly parents into existing educational propaganda programs (such as media reports, movies, and TV shows) aimed to reduce gender disparities. Such actions may be helpful in efforts of reducing the traditional idea of “Yang Er Fang Lao” (having a son prevents difficulties in old age), which is one core idea behind son preference, especially in rural areas. This is particularly relevant today, because emotional support may become more important than financial support in China where the living conditions of a large majority of the population, including the elderly, has been substantially improved. Public media propaganda, based on data from a nationwide survey, which explains that daughters provide better emotional care to older parents may be particularly informative for young couples who do not currently face the challenges of care needs and may not anticipate being disabled in old age. However, these empirical findings and policy recommendations are far from sufficient to eliminate the traditional son preference, because other important factors are operating and must be taken into consideration.

As mentioned in the “Introduction” section, one of the major problems behind the persistence of son preference in China is the underdevelopment of the pension system in rural China, where the elderly rely heavily on monetary transfers and other kinds of economic support from adult children, mainly sons (Chu, Xie, & Yu, 2011). In contrast, the elderly in urban areas in China (and more developed countries) get most of their financial support from pensions and savings. Actions are being taken with the recent and promising policy guidance, and the New Rural Pension Program in China is expected to be quickly developed. Its effectiveness, however, remains to be seen, and more action is necessary to increase its efficiency and to effectively implement the pension program in rural areas (Hu, Yang, & Yan, 2015; Ye, 2009).

In sum, we firmly believe that China needs to strengthen the integrative policy-related actions: (a) effectively inform the public that daughters’ better care to older parents is beneficial for some important aspects of parents’ health, including maintaining higher cognitive capacity and reducing mortality risk, especially in rural areas and at oldest-old ages; (b) further develop the pension system to fully cover all rural and urban residents; and (c) continuously and effectively promote equality between males and females. We expect that strengthening these integrated policy-related actions would substantially reduce and eventually eliminate traditional son preference and enable Chinese elderly to enjoy a better life.

Footnotes

Authors’ Note

The views expressed in this article are solely those of the authors and do not necessarily reflect those of Duke University, Peking University, the University of Rhode Island, the United Nations, or the Max Planck Institute for Demographic Research. Dr. Danan Gu’s work was mainly completed when he was at Duke University. He now works at the United Nations Population Division.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study is part of the project funded by National Institute on Aging/National Institutes of Health (R01AG023627-10), National Natural Science Foundation of China (71110107025, 71233001, 71490732), and United Nations Funds for Population Activities.

Notes

References

Abel

E. K.

(1990). Informal care for the disabled elderly. Research on Aging, 12, 139-157.

Bian

Logan

J. R.

Bian

(1998). Intergenerational relations in urban China: Proximity, contact, and help to parents. Demography, 35, 115-124.

Broese Van Groenou

M. I.

Knipscheer

C. P.

(1999). Onset of physical impairment of independently living older adults and the support received from sons and daughters in the Netherlands. The International Journal of Aging and Human Development, 48, 263-278.

Buber

Engelhardt

(2008). Children’s impact on the mental health of their older mothers and fathers: Findings from the Survey of Health, Ageing and Retirement in Europe. European Journal of Ageing, 5, 31-45. doi:10.1007/s10433-008-0074-8

Byers

A. L.

Levy

B. R.

Allore

H. G.

Bruce

M. L.

Kasl

S. V.

(2008). When parents matter to their adult children: Filial reliance associated with parents’ depressive symptoms. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 63, P33-P40.

Cesarini

Lindqvist

Wallace

(2009). Is there an adverse effect of sons on maternal longevity? Proceedings of the Royal Society B: Biological Sciences, 276, 2081-2084.

Chappell

N. L.

Kusch

(2007). The gendered nature of filial piety—A study among Chinese Canadians. Journal of Cross-Cultural Gerontology, 22, 29-45.

Chen

(2013). The role of children’s support in elders’ decisions to live in a Yanglaoyuan (Residential Long-Term Care). Journal of Cross-Cultural Gerontology, 28, 75-87.

Chesley

Poppie

(2009). Assisting parents and in-laws: Gender, type of assistance, and couples’ employment. Journal of Marriage and Family, 71, 247-262.

10.

Chu

Xie

R. R.

(2011). Coresidence with elderly parents: A comparative study of Southeast China and Taiwan. Journal of Marriage and Family, 73, 120-135.

11.

Coale

A. J.

(1991). The effect of age misreporting in China on the calculation of mortality rates at very high ages. Demography, 28, 293-301.

12.

Cong

Silverstein

(2012). Caring for grandchildren and intergenerational support in rural China: A gendered extended family perspective. Ageing and Society, 32, 425-450.

13.

Cooney

R. S.

(1999). Primary family caregivers of impaired elderly in Shanghai-China: Kin relationship and caregiver burden. Research on Aging, 21, 739-761.

14.

Crimmins

E. M.

Hayward

M. D.

Saito

(1994). Changing mortality and mobility rates and health status and life expectancy of older population. Demography, 31, 159-175.

15.

Deutsch

F. M.

(2006). Filial piety, patrilineality, and China’s one-child policy. Journal of Family Issues, 27, 366-389.

16.

Ebenstein

Leung

(2010). Son preference and access to social insurance: Evidence from China’s rural pension program. Population and Development Review, 36, 47-70.

17.

Feng

Hoenig

H. M.

Zeng

Purser

J. L.

(2010). Impact of new disability subtypes on 3-year mortality in Chinese older adults. Journal of the American Geriatrics Society, 58, 1952-1958.

18.

Feng

Son

Zeng

(2015). Prevalence and correlates of successful ageing: A comparison between China and South Korea. European Journal of Ageing, 12(2), 83-94.

19.

Folstein

M. F. Y.

Folstein

S. E.

McHugh

P. R.

(1975). “Mini-mental state.” A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12(3), 189-198.

20.

Goggins

W. B.

Woo

Sham

S. C.

(2005). Frailty index as a measure of biological age in a Chinese population. The Journals of Gerontology. Series A, Biological Sciences and Medical Sciences, 60, 1046-1051.

21.

Goodkind

(2009). [Review of the book Healthy Longevity in China: Demographic, Socioeconomic, and Psychological Dimensions]. Population Studies, 63(3), 1-7.

22.

Dupre

M. E.

(2008). Assessment of reliability of mortality and morbidity in the 1998–2002 CLHLS waves. In Zeng

Poston

Vlosky

D. A.

(Eds.), Healthy longevity in China: Demographic, socioeconomic, and psychological dimensions (pp. 99-116). Dordrecht, The Netherlands: Springer.

23.

Yang

Yan

(2015). Trends and tasks of social security in China in the 13th five-year national development plan period. Journal of the Party School of the Central Committee of the CPC, 1, 85-90. [In Chinese]

24.

Hurt

Ronsmans

Quigley

(2006). Does the number of sons born affect long-term mortality of parents? A cohort study in rural Bangladesh. Proceedings of the Royal Society B: Biological Sciences, 273, 149-155.

25.

Jasienska

Nenko

Jasienski

(2006). Daughters increase longevity of fathers, but daughters and sons equally reduce longevity of mothers. American Journal of Human Biology, 18, 422-425.

26.

Kahn

J. R.

McGill

B. S.

Bianchi

S. M.

(2011). Help to family and friends: Are there gender differences at older ages? Journal of Marriage and Family, 73(1), 77-92.

27.

Kulminski

A. M.

Ukraintseva

S. V.

Kulminskaya

I. V.

Arbeev

K. G.

Land

K. C.

Yashin

A. Y.

(2008). Cumulative deficits better characterize susceptibility to death in the elderly than phenotypic frailty: Lessons from the Cardiovascular Health Study. Journal of the American Geriatrics Society, 56, 898-903.

28.

L. W.

Liang

Toler

(2005). Widowhood and depressive symptoms among older Chinese: Do gender and source of support make a difference? Social Science & Medicine, 60, 637-647. doi:10.1016/j.socscimed.2004.06.014

29.

L. W.

Seltzer

M. M.

(2005). Relationship quality with parent, daughter role salience, and self-esteem of daughter caregivers. Marriage & Family Review, 37, 63-82.

30.

Liu

W. T.

Kendig

(2000). Critical issues of caregiving: East-West dialogue. In Liu

W. T.

Kendig

(Eds.), Who should care for the elderly (pp. 1-23). Singapore: Singapore University Press.

31.

Meng

(2002). Rebellion and revenge: The meaning of suicide of women in rural China. International Journal of Social Welfare, 11, 300-309. doi:10.1111/1468-2397.00239

32.

Miller

Cafasso

(1992). Gender differences in caregiving: Fact or artifact? The Gerontologist, 32, 498-507.

33.

A. C. Y.

Phillips

D. R.

Lee

W. K.

(2002). Persistence and challenges to filial piety and informal support of older persons in a modern Chinese society: A case study in Tuen Mun, Hong Kong. Journal of Aging Studies, 16, 135-153.

34.

Pearson

(2002). Ling’s death: An ethnography of a Chinese woman’s suicide. Suicide and Life-Threatening Behavior, 32, 347-358.

35.

Pham-Kanter

Goldman

(2011). Do sons reduce parental mortality? Journal of Epidemiology and Community Health, 66, 710-715.

36.

Poston

D. L.

Jr. Luo

(2004). Zhongguo 2000 nian shaoshu minzu de nian ling dui ji he shu zi pian hao [Age structure and composition of the Chinese minorities in 2000]. Zhongguo Shaoshu Minzu Renkou (Chinese Minority Populations), 19(3), 9-15.

37.

Purser

J. L.

Golightly

Y. M.

Feng

Helmick

C. G.

Renner

J. B.

Jordan

J. M.

(2012). Association of slower walking speed with incident knee osteoarthritis-related outcomes. Arthritis Care & Research, 64, 1028-1035.

38.

Raley

S. J.

Bianchi

(2006). Sons, daughters, and family processes: Does gender of children matter? Annual Review of Sociology, 32, 401-421.

39.

Rossi

A. S.

Rossi

P. H.

(1990). Of human bonding: Parent-child relations across the life course. Hawthorne, NY: Aldine.

40.

Shen

Zeng

(2010). The association between resilience and survival among Chinese elderly. Demographic Research, 23(5), 105-116.

41.

Silverstein

Bengtson

V. L.

(1991). Do close parent-child relations reduce the mortality risk of older parents? Journal of Health and Social Behavior, 32, 382-395.

42.

Silverstein

Bengtson

V. L.

(1994). Does intergenerational social support influence the psychological well-being of older parents? The contingencies of declining health and widowhood. Social Science & Medicine, 38, 943-957.

43.

Spitze

Logan

(1990). Sons, daughters, and intergenerational social support. Journal of Marriage and Family, 52, 420-430.

44.

Sun

(2002). Old age support in contemporary urban China from both parents’ and children’s perspectives. Research on Aging, 24, 337-359.

45.

Wang

Zeng

Jeune

Vaupel

J. W.

(1998). Age validation of Han Chinese centenarians. Genus, 54, 123-141.

46.

Weinstein

Glei

D. A.

Yamazaki

Chang

M.-C.

(2004). The role of intergenerational relations in the association between life stressors and depressive symptoms. Research on Aging, 26, 511-530. doi:10.1177/0164027504266463

47.

Whyte

M. K.

(Ed.). (2003). China’s revolutions and intergenerational relations. Ann Arbor: The University of Michigan Press.

48.

Xie

Zhu

(2009). Do sons or daughters give more money to parents in urban China? Journal of Marriage and Family, 71, 174-186.

49.

(2001). Family support for old people in rural China. Social Policy & Administration, 35, 307-320.

50.

Yang

Chandler

(1992). Intergenerational relations: Grievances of the elderly in rural China. Journal of Comparative Family Studies, 23, 431-453.

51.

(2009, September 29). Establish a green cover for people—The 60-years of the social security program in China. People’s Daily (overseas edition), p. 2 [in Chinese]. Retrieved from http://paper.people.com.cn/rmrbhwb/html/2009-09/29/content_352961.htm

52.

Yin

Z.-X.

Shi

X.-M.

Kraus

V. B.

Fitzgerald

S. M.

Qian

H.-z

J.-w.

Zeng

(2012). High normal plasma triglycerides are associated with preserved cognitive function in Chinese oldest-old. Age and Ageing, 41, 600-606.

53.

Zeng

(2012). Towards deeper research and better policy for healthy aging—Using the unique data of Chinese Longitudinal Healthy Longevity Survey. China Economic Journal, 5, 131-149.

54.

Zeng

George

L. K.

(2010). Population aging and old-age care in China. In Dannefer

Phillipson

(Eds.), SAGE handbook of social gerontology (pp. 420-429). Thousand Oaks, CA: SAGE.

55.

Zeng

George

L. K.

Brasher

Vaupel

J. W.

(2015). Older parents enjoy better filial piety and care from daughters than sons in China–and policy discussions to reduce son-preference and high sex ratio at birth. Working Paper of Max Planck Institute for Demographic Research, Rostock, Germany.

56.

Zeng

(2008). Reliability of age reporting among the Chinese oldest-old in the CLHLS data sets. In Zeng

Poston

D. J.

Vlosky

D. A.

(Eds.), Healthy longevity in China: Demographic, socioeconomic, and psychological dimensions (pp. 61-78). Dordrecht, The Netherlands: Springer.

57.

Zhan

H. J.

(2004). Willingness and expectations: Intergenerational differences in attitudes toward filial responsibility in China. Marriage & Family Review, 36, 175-200.

58.

Zhang

Wang

(2010). Meal and residence rotation of elderly parents in contemporary rural Northern China. Journal of Cross-Cultural Gerontology, 25, 217-237.

59.

Zhu

(2015). Unmet needs in long-term care and their associated factors among the oldest old in China. BMC Geriatrics, 15, Article 46. doi:10.1186/s12877-015-0045-9

60.

Zunzunegui

M. V.

Beland

Otero

(2001). Support from children, living arrangements, self-rated health and depressive symptoms of older people in Spain. International Journal of Epidemiology, 30, 1090-1099.