
==== Front
J Glob Health
J Glob Health
JGH
Journal of Global Health
2047-2978
2047-2986
International Society of Global Health

jogh-14-04193
10.7189/jogh.14.04193
Articles
The influence of widowhood and social engagement on cognitive impairment among Chinese older adults and factors mediating their association
Sheng et al. Widowhood and social engagement's impact on cognitive impairment
https://orcid.org/0000-0001-5899-048X
Sheng Mingyuan 2
https://orcid.org/0009-0003-8302-6563
Young Kathleen 3
https://orcid.org/0000-0001-9710-8648
Li Ying 4
https://orcid.org/0000-0002-0774-7225
Zhang Yeyuan 2
https://orcid.org/0009-0003-9964-6521
Wang Jiale 5
https://orcid.org/0000-0003-3370-3565
Jiang Shuhan 1
1 School of Humanities and Management, Zhejiang Chinese Medical University, Hangzhou, China
2 School of Public Health, Zhejiang Chinese Medical University, Hangzhou, China
3 Department of Health Sciences, MPH and Public Health Education Programs, California State University, Northridge, California, USA
4 School of Public Health, Xi'an Medical University, Xi’an, China
5 School of Nursing, Shaoxing University Yuanpei College, Shaoxing, China
Correspondence to:
Shuhan Jiang
School of Humanities and Management, Zhejiang Chinese Medical University
Binwen Rd. 548. Hangzhou, China, 310000
China
Shuhan_jiang@163.com
20 9 2024
2024
14 04193Copyright © 2024 by the Journal of Global Health. All rights reserved.
2024
https://creativecommons.org/licenses/by/4.0/ This work is licensed under a Creative Commons Attribution 4.0 International License.
Background

Prior studies exploring the impact of widowhood on cognitive impairment in later life have been focussed on the USA and Europe. We aimed to explore the mediating role of social engagement, health behaviours, and subjective well-being in the association between widowhood and cognitive impairment in the Chinese population.

Methods

We conducted a study on 7796 older individuals enrolled in the 2018 wave of the Chinese Longitudinal Health Longevity Study. We used logistic regression models to analyse the impact of widowhood on cognitive health among older adults and performed mediation analysis to determine possible mediating factors in this relationship.

Results

Widows and widowers had a higher risk of having cognitive impairment than married older adults (95% confidence interval (CI) = 1.312, 2.279). The results from structural equation modelling (SEM) provided a good fit to the observed data (χ2 = 24.909; P = 0.00) and indicated that the effect of widowhood on cognitive impairment was partially mediated by social engagement, lifestyle behaviours, and subjective well-being (β = 0.075; P < 0.01).

Conclusions

Our findings contribute to existing research on the mechanisms underlying the association between widowhood and cognitive impairment among older individuals, suggesting a need for policies targeted at the specific needs of this vulnerable population, such as the maintenance of social interactions, adoption of a healthy lifestyle, improvement of subjective well-being, and provision of necessary support systems.
==== Body
pmcPopulation ageing has been a growing concern for public health globally. This is especially true for China, whose population is ageing much faster than previously reported, with projections suggesting it will have more than 400 million residents who are ≥60 years of age by 2040 [1]. This would also lead to the prevalence, morbidity, and mortality of several diseases increasing sooner than expected, particularly for chronic non-infectious diseases, placing a burden on society and families to allocate resources for medical treatment, therapy, and health care for seniors.

To exacerbate these burdens further, approximately one-quarter (18.95%) of Chinese older adults had previously reported suffering from cognitive impairment [2]. As people with cognitive impairment frequently struggle with daily chores and require extensive medical and personal care, maintaining cognitive function remains a key part of health and well-being in later life [3–5]. Prior research has found a variety of diseases and bad lifestyle habits to be closely associated with the occurrence and development of cognitive impairment, with hypertension and diabetes, poor social relationships, alcohol consumption, a lack of formal education, a lack of physical exercise, and sleep disorders emerging as important and preventable risk factors [6,7].

The number of widowed older adults is predicted to increase to 118.4 million by 2050 due to gender-based disparities in life expectancy and China's ageing population [8]. Although there are several studies on the impact of widowhood and cognitive impairment on this population, most have been conducted in the USA and Europe and have treated marital status exclusively as a control variable. Existing studies on the Chinese population, in turn, have not yet focussed on how lifestyle changes after widowhood affect this relationship [9]. Several factors can influence the development of cognitive impairment [10]; however, some can be avoided through changes in an individual’s lifestyle, even in later life [11].

Several studies have indicated the health benefits of marriage, as it ensures social, psychological, and financial resources while also encouraging people to maintain healthy lifestyles [12]. For example, spouses can motivate and support each other to quit smoking, engage in physical activities, or maintain a healthy diet, thereby potentially reducing the risk of death and dementia health [6,13]. However, for older adults, widowhood following the death of a partner emerges as an especially traumatic and challenging life experience [14–16]. Research has shown that the lifestyles of widowed older adults significantly differ from those of non-widowed adults. For example, widows and widowers were found to be at higher risk of poor nutrient intake, declined sleep quality, increased alcohol consumption, and initiating or restarting smoking following their partner’s death [15,16]. Older adults simultaneously experience a decline in social engagement, as family members and friends who had been reliant on the recently deceased spouse to maintain close relationships gradually move away [17]. They are also known to develop negative emotions such as depression, anxiety, and loneliness following a period of grieving [18], with evidence suggesting an association between widowhood and poor mental health [19,20]. The mental health of older widows and widowers directly affects their attitude toward life satisfaction and well-being [21]. Yet despite a consensus that mental health is essential for active healthy ageing, few studies have explored the role of subjective well-being in the relationship between widowhood and cognitive impairment.

Research comparing widowed and non-widowed older individuals found that widowhood was associated with a higher risk of cognitive impairment. Stress and grief following the death of a spouse were also found to increase a widowed individual’s risk of depression and other chronic health problems, as well as dementia and cognitive impairment [22,23]. A cohort study with a 17-year follow-up involving 6766 older adults in the USA discovered a linear connection between widowhood duration and cognitive decline [24], while a longitudinal study with an eight-year follow-up in Korea showed that widowed older adults experienced considerable cognitive deterioration compared to non-widowed individuals [25]. Although the link between widowhood and cognitive function in Chinese older adults has received less attention, a study in four cities reported similar results [26].

The mechanism between widowhood and cognitive function is frequently discussed through the stress and marital resource models [27]. According to the stress model, hippocampal atrophy may result from emotional trauma-fueled stress-induced glucocorticoid secretion [28]. The marital resource model, in turn, proposes that marriage helps one establish or obtain social interaction, social support, and financial resources, which are associated with better cognitive function in ageing [28]. Marital interaction can also provide daily cognitive stimulation for the older adult, which is crucial for maintaining cognitive function [29]. Furthermore, emerging biomarker research discovered independent and interactive relationships among widowhood, β-amyloid levels (an abnormal brain protein associated with Alzheimer disease), and cognitive decline [30]. These mechanisms suggest that widowed individuals are at a higher risk of cognitive impairment than married ones.

Aside from the loss of economic or social support, widowed individuals may also experience spiritual loss, loneliness, and other inconveniences in daily life, resulting in further serious health consequences. A study among community-dwelling older Chinese adults in Singapore found that widowhood was associated with a higher risk of cognitive impairment than being married, but the link was statistically significant for men only [31]. However, in their population-based research on 4370 older adults, Shahar et al. [32] found no correlation between widowhood and dementia. Another study conducted in three Chinese communities found no significant correlation between being widowed and cognitive scores after controlling for covariates [27]. This may be partially attributed to the fact that widowhood has variable effects depending on social and cultural contexts and that the importance of various pathways relating widowhood to health may vary by gender and regional norms [14]. The relationship, therefore, requires further studies.

Previous research in developed countries has established a positive association between social engagement and improved health outcomes across multiple domains, including mortality, morbidities, well-being, cognitive function, and life quality [33,34]. For example, individuals with higher levels of social activity were found to typically have greater cognitive functions than those who are less socially engaged [35]. This might be due to the emotional, behavioural, and physiological benefits of social interaction on the functional abilities of older individuals [36], as older adults who are highly socially engaged are less likely to experience functional disability due to the maintenance of their cognitive ability [37]. Additionally, engaging in protective health behaviours such as an active lifestyle can improve the underlying relationship between social engagement and health [38].

With these research gaps in mind, we designed a study based on a nationwide sample of older Chinese individuals to explore the potential mediating roles of social engagement, health behaviours, and subjective well-being in the association between cognitive impairment and widowhood. We hypothesised that widowhood directly impacts both cognitive health and social engagement, and that it indirectly influences cognitive health through its effects on post-widowhood social engagement, subjective well-being, and various health-related behaviours, including sleep duration, physical activity, alcohol consumption, smoking, and dietary habits.

METHODS

Data source and study population

We retrieved data from the 2018 wave of the Chinese Longitudinal Healthy Longevity Survey (CLHLS). The CLHLS is the largest collection of survey data on the older adult population in China, encompassing 23 provinces inhabited by approximately 85% of the Chinese population. Following a baseline survey using a self-administered questionnaire in 1998, additional waves were conducted in 2000, 2002, 2005, 2008, 2011, 2014, and 2018 to investigate the health status and associated social, behavioural, and biological aspects among Chinese older adults recruited on a volunteer participatory basis. After we screened for invalid values and outliers, we selected 3605 men and 4191 women within the 2018 survey, comprising 1367 widowed men and 2767 widowed women (Table 1).

Table1 Table of variable assignments

	Measurement	
Dependent variable		
Social engagement	1 = impaired social engagement; 0 = healthy social engagement	
Cognitive health	1 = impaired cognitive; 0 = no cognitive impairment	
Independent variables		
Widowhood situation	1 = widowed; 0 = married and living with a spouse	
Intervening variable		
State of mind		
Subject well-being	Scores ranging from 5 to 25	
Health behaviour		
Dietary behaviours	Scores ranging from 2 to 7	
Sleep	1 = less/more; 0 = 7–8 h	
Drinking	1 = yes; 0 = no	
Smoking	1 = yes; 0 = no	
Exercise	1 = no; 0 = yes	

Measure

Widowhood status was measured through the question ‘What is your current marital status?’ with response options being ‘married’, ‘divorced’, ‘widowed’, and ‘never married’. We only included study respondents who had either been married or had been widowed to decrease the impact of divorce, re-marriage, or never having been married on the outcome variables of interest. We divided the respondents into two groups based on marital status collected in the CLHLS: those who are currently married and had not gone through a divorce and widowers who had not remarried (reference group) [27].

Cognitive health was assessed using the Chinese version of the Mini-Mental State Examination (MMSE) from 1995. In our study, we used education status-specific categories to define cognitive impairment based on the latest normative and validation study of the MMSE (2013) in the Chinese population, with possible scores ranging from 0 to 30 [39]. Participants with an MMSE score of ≤16 and no formal education, MMSE scores of 17–19 with 1–6 years of education, and MMSE scores of 20–23 with more than 6 years of education were categorised as having impaired cognitive function and coded as 0. Those with MMSE scores above these thresholds were classified as cognitively healthy and coded as 1.

We further used three risk factor categories as mediating variables in our model based on prior research on widowhood in older adult populations (Table 1): social engagement; mental state, including subjective well-being; and health behaviours, including drinking, smoking, daily exercise habits, sleep duration, and dietary behaviours.

We calculated the level of social engagement based on social activities in the last month, including visiting neighbours/friends, attending a dance class, exercising in a park or other places such as a sports or social club, etc. As most of the respondents did not engage in any social activity, we dichotomised the variable as participating in more than one activity (coded as 1) and not having any social activities (coded as 0). Older adults who had never participated in any social or outdoor activities were reported or documented as socially unhealthy.

Based on the previous studies of dietary behaviours among Chinese older adults, we selected four items from the CLHLS questionnaire [11] and recorded each its frequency by rank. The ranking levels for fruits and vegetables were ‘4 = almost every day’, ‘3 = quite often’, ‘2 = occasionally’, and ‘1 = rarely or never’, while the levels of consumption of meat and eggs were ‘3 = almost every day’, ‘2 = occasionally’, or ‘1 = rarely or never’, with combined scores ranging from 2 to 7.

Information concerning sleep duration obtained through the question: ‘How many hours on average do you sleep every day, including napping?’, whereby seven to eight hours of sleep indicated sufficient sleep duration. This classification was based on the profile study by Hirshkowitz and colleagues [40] which suggested it as the recommended amount of sleep for older adults. This recommendation was later adopted as the sleep duration guideline by the Centers for Disease Control and Prevention. We coded all sleep-related variables as binary variables.

The participants were asked whether they currently drink alcohol and/or smoke, with possible response options being ‘yes’ or ‘no’. Information on exercise habit was measured in the CLHLS using two questions: ‘Do you exercise regularly now?’ (referring to intentional fitness activities like Qigong, walking, playing ball, running, etc.) and ‘Did you previously engage in regular exercise?’ Participants who answered ‘yes’ to both questions were classified as having an exercise habit and coded as 1, and defined as ‘exercise’.

Subjective well-being in the CLHLS was measured using five items identified from prior research [41]. The questions used to measure experienced well-being were ‘Do you always look on the bright side of things?’, ‘Are you happy now as when you were younger?’, ‘Do you often feel fearful or anxious?’, ‘Do you often feel lonely and isolated?’, and ‘Do you feel the older you get the more useless you are?’. The responses ranged from 1 to 5., with ‘1’ representing ‘always or very good’ and ‘5′ representing ‘never or very bad’. We recoded these variable as 1 = the weakest feeling and 5 = the strongest feeling. This generated two indices, whereby summing items 3–5 produced a positive well-being index and summing items 1–2 gave a produced well-being index. The overall measure of well-being ranged from 5 to 25, with higher numbers indicating better well-being.

The CLHLS had ‘unable to answer’ as a response option for in many questions, which we then treated as a missing value in our analyses.

Statistical analysis

We used descriptive statistics to summarise the general characteristics of our sample, presenting continuous variables through means/standard deviations and categorical variables through numbers/frequencies.

We first evaluated the effect of an independent variable on a dependent variable while excluding any mediating variables by using logistic regression models to analyse the effect of widowhood on cognitive health among older adults. We then conducted further analyses by including those variables that were significantly associated with the dependent variable (P < 0.05) and those of interest to this study in a multivariate logistic regression model. Specifically, we added social engagement, health behaviours, and subjective well-being to the model as control variables respectively to analyse various influencing factors. We reported our findings through regression coefficients and 95% confidence intervals (CIs).

We then designed a mediation analysis to determine if health behaviour and subjective well-being had a mediating role in the association between widowhood and cognitive and social health status. We evaluated the model per the following indicators: χ2 means and degrees of freedom (df), with χ2 means/df <3 indicating an acceptable fit; the comparative fit index (CFI): the Tucker-Lewis index (TLI); the standardised root means square residual (SRMR), and the root mean square error of approximation (RMSEA), with an RMSEA value ≤0.05 indicating a good model fit. The acceptable value for the non-formed fit index (NFI), goodness-of-fit index (GFI), and adjusted goodness-of-fit index (AGFI) is 0.8–0.99.

We used IBM SPSS Statistics, version 26 (IBM, SPSS Inc., NY, USA) and SPSS AMOS, version 24.0 (IBM, SPSS Inc., NY, USA) for all data analyses.

RESULTS

Descriptive statistics

In total, 53.8% of study participants were female; 8.8% of older adults reported cognitive impairment, while 53% reported losing a spouse (Table 2).

Table 2 Characteristics of the study sample and widowhood prevalence among Chinese older adults (n = 7796)*

Variable	n (%)	Widowhood, n (%)	χ2/t	P-value	
Hukou			23.474	<0.001	
Urban	2473 (31.7)	1212 (49.0)			
Rural	5323 (68.3)	2922 (54.9)			
Sex			614.479	<0.001	
Male	3605 (46.2)	1367 (37.9)			
Female	4191 (53.8)	2767 (66.0)			
Age			2092.501	<0.001	
65–79	3216 (41.3)	789 (24.5)			
80–94	3099 (39.8)	1986 (64.1)			
>95	1481 (19.0)	1359 (91.8)			
Ethnicity			0.931	0.335	
Han	7347 (94.2)	3886 (55.2)			
Minority	449 (5.8)	248 (52.9)			
Year of schooling			799.600	<0.001	
0	3395 (43.5)	2374 (69.9)			
1–6	2737 (35.1)	1263 (46.1)			
>7	1664 (21.3)	497 (29.9)			
Self-report economic status			2.045	0.727	
Very rich	209 (2.7)	113 (54.1)			
Rich	1306 (16.8)	675 (51.7)			
Average	5527 (70.9)	2934 (53.1))			
Poorer	668 (8.6)	363 (54.3)			
Very poor	86 (1.1)	49 (57.0)			
Living status			1091.499	<0.001	
Co-habitation	6364 (81.6)	2812 (44.2)			
Living alone	1221 (15.7)	1141 (93.4)			
In nursing home	211 (2.7)	181 (85.8)			
Smoking or not (no = 0)			135.905	<0.001	
Yes	1245 (16.0)	472 (37.9)			
No	6551 (84.0)	3662 (55.9)			
Drinking or no (no = 0)			104.532	<0.001	
Yes	1230 (15.8)	488 (39.7)			
No	6566 (84.2)	3646 (55.5)			
Recommended sleep duration (no = 0)			48.669	<0.001	
Yes	2915 (37.4)	1397 (47.9)			
No	4881 (62.6)	2737 (56.1)			
Daily exercise or not			128.019	<0.001	
Yes	2877 (36.9)	1285 (44.7)			
No	4919 (63.1)	2849 (57.9)			
Cognitive health			291.807	<0.001	
No cognitive impairment	7113 (91.2)	3559 (50.0)			
Impaired cognitive	683 (8.8)	575 (84.2)			
Social health			266.076	<0.001	
High social engagement	5624 (72.1)	2660 (47.3)			
Low social engagement	2172 (27.9)	1474 (67.9)			
Dietary behaviour, x̄ (SD)	4.530 (1.423)		10.399	<0.001	
Subjective well-being, x̄ (SD)	16.64 (3.552)		10.893	<0.001	
SD – standard deviation, x̄ – mean

*Presented as n (%) unless specified otherwise.

Multi-level logistic regression results

Widows and widowers (84.2%) had a higher risk of having cognitive impairment than married older adults (37.9%). The results for the full model showed that being widowed was significantly associated with cognitive impairment, with the odds ratio of widows and widowers for having impaired cognition being 1.729 (95% CI = 1.312, 2.279) times higher than those who are married. Conversely, lower subjective well-being (odds ratio (OR) = 0.835; 95% CI = 0.818, 0.853), impaired social engagement (OR = 2.431; 95% CI = 1.993, 2.966), lack of daily exercise (OR = 1.585; 95% CI = 1.223, 2.053), and unhealthy dietary behaviours (OR = 0.812; 95% CI = 0.737, 0.895) were significantly associated with cognitive impairment (Table 3).

Table 3 Logistic regression analyses on cognitive impairment of older adult (n = 7796)*

Variables	Model 1	P-value	Model 2	P-value	Model 3	P-value	Model 4	P-value	Model 5	P-value	
Hukou (ref: rural)	0.774 (0.628, 0.954)	<0.05	0.962 (0.774, 1.195)	0.726	0.811 (0.648, 1.016)	0.069	0.736 (0.594, 0.912)	<0.01	0.877 (0.693, 1.109)	0.272	
Sex (ref: women)	0.901 (0.738, 1.100)	0.307	0.950 (0.758, 1.192)	0.659	0.954 (0.772, 1.179)	0.662	0.940 (0.767, 1.153)	0.553	1.007 (0.789, 1.284)	0.956	
Age 80–94 (ref: 65–79)	3.987 (2.842, 5.592)	<0.001	3.733 (1.658, 5.242)	<0.001	3.727 (2.642, 5.259)	<0.001	3.373 (2.397, 4.746)	<0.001	3.162 (2.230, 4.484)	<0.001	
Age >95 (ref: 65–79)	16.704 (11.765, 23.715)	<0.001	14.224 (9.992, 20.249)	<0.001	13.650 (9.519, 19.573)	<0.001	11.363 (7.944, 16.255)	<0.001	9.465 (6.534, 13.710)	<0.001	
1–6 y of schooling (ref: 0 y)	0.700 (0.559, 0.877)	<0.01	0.776 (0.609, 0.964)	<0.05.	0.816 (0.641, 1.037)	0.097	0.728 (0.579, 0.916)	<0.01	0.855 (0.668, 1.094)	0.212	
>7 y of schooling (ref: 0 y)	1.031 (0.761, 1.397)	0.845	1.284 (0.939, 1.755)	0.118	1.400 (1.017, 1.928)	<0.05	1.152 (0.844, 1.571)	0.372	1.664 (1.196, 2.314)	<0.01	
Living alone (ref: cohabitation)	0.594 (0.459, 0.768)	<0.001	0.549 (0.422, 0.713)	<0.001	0.627 (0.478, 0.823)	<0.01	0.680 (0.523, 0.884)	<0.01	0.653 (0.494 0.863)	<0.01	
In nursing home (ref: cohabitation)	2.083 (1.404, 3.091)	<0.001	2.388 (1.592, 3.581)	<0.001	2.195 (1.460, 3.300)	<0.001	2.203 (1.473, 3.295)	<0.001	2.424 (1.590, 3.697)	<0.001	
Widowed or not (ref: no)	1.772 (1.369, 2.293)	<0.001	1.772 (1.366, 2.299)	<0.001	1.781 (1.358, 2.334)	<0.001	1.689 (1.299, 2.195)	<0.001	1.729 (1.312, 2.279)	<0.001	
Smoking or not (ref: no)			1.116 (0.884, 1.409)	0.356					1.073 (0.836, 1.375)	0.581	
Drinking or not (ref: no)			0.855 (0.678, 1.076)	0.182					0.882 (0.688, 1.131)	0.322	
Daily exercise or not (ref: yes)			2.403 (1.893, 3.051)	<0.001					1.585 (1.223, 2.053)	<0.001	
Recommended sleep duration (ref: yes)			1.183 (0.979, 1.429)	0.082					0.986 (0.806, 1.206)	0.887	
Dietary behaviours			0.717 (0.655, 0.784)	<0.001					0.812 (0.737, 0.895)	<0.001	
Subjective well-being					0.819 (0.803, 0.837)	<0.001			0.835 (0.818, 0.853)	<0.001	
Social engagement (ref: healthy)							3.163 (2.639, 3.791)	<0.001	2.431 (1.993, 2.966)	<0.001	
Fixed parameter	0.017	<0.001	0.039	<0.001	0.358	<0.001	0.012	<0.001	0.413	<0.05	
CI – confidence interval, OR – odds ratio, ref – reference

*Presented as OR (95% CI) unless specified otherwise.

Widowed individuals who reported low social engagement were 1.182 (95% CI = 1.024, 1.365) times higher than those who are married to those who have cognitive impairment in the full model. Subjective well-being (OR = 0.961; 95% CI = 0.947, 0.976), daily exercise OR = 3.524 (95% CI = 3.074, 4.040), smoking (OR = 1.196; 95% CI = 1.031, 1.148), dietary behaviours (OR = 0.831; 95% CI = 0.782, 0.883) all showed a statistically significant association with social engagement (Table 4).

Table 4 Logistic regression analyses on social engagement of older adult (n = 7796)

Variables	Model 1	P-value	Model 2	P-value	Model 3	P-value	Model 4	P-value	
Hukou (ref: rural)	1.032 (0.908, 1.173)	0.627	1.397 (1.217, 1.603)	<0.001	1.059 (0.931, 1.204)	0.384	1.399 (1.218, 1.607)	<0.001	
Sex (ref: women)	0.965 (0.855, 1.089)	0.566	0.935 (0.809, 1.081)	0.366	0.982 (0.869, 1.109)	0.767	0.941 (0.814, 1.089)	0.417	
Age 80–94 (ref: 65–79)	2.147 (1.865, 2.473)	<0.001	2.070 (1.790, 2.393)	<0.001	2.127 (1.865, 2.450)	<0.001	2.061 (1.782, 2.384)	<0.001	
Age >95 (ref: 65–79)	5.876 (4.945, 6.981)	<0.001	5.017 (4.196, 5.997)	<0.001	5.486 (4.611, 6.527)	<0.001	4.825 (4.033, 5.774)	<0.001	
1–6 y of schooling (ref: 0 y)	0.794 (0.697, 0.905)	<0.001	0.895 (0.782, 1.026)	0.111	0.823 (0.721, 0.939)	<0.01	0.911 (0.795, 1.044)	0.182	
>7 y of schooling (ref: 0 y)	0.585 (0.487, 0.703)	<0.001	0.734 (0.606, 0.891)	<0.01	0.621 (0.516, 0.747)	<0.001	0.751 (0.619, 0.911)	<0.01	
Living alone (ref: co-habitation)	0.619 (0.528, 0.725)	<0.001	0.590 (0.501, 0.696)	<0.001	0.627 (0.535, 0.736)	<0.001	0.598 (0.508, 0.706)	<0.001	
In nursing home (ref: co-habitation)	1.201 (0.884, 1.632)	0.241	1.336 (0.967, 1.846)	0.079	1.180 (0.867, 1.604)	0.292	1.312 (0.950, 1.814)	0.100	
Widowed or not (ref: no)	1.189 (1.035, 1.366)	<0.05	1.188 (1.029, 1.371)	<0.05	1.180 (1.027, 1.357)	<0.05	1.182 (1.024, 1.365)	<0.05	
Smoking or not (ref: 0)			1.196 (1.032, 1.386)	<0.05			1.196 (1.031, 1.386)	<0.05	
Drinking or not (ref: 0)			0.988 (0.857, 1.138)	0.863			0.996 (0.864, 1.148)	0.957	
Daily exercise or not (ref: yes)			3.633 (3.171, 4.163)	<0.001			3.524 (3.074, 4.040)	<0.001	
Recommended sleep duration (ref: yes)			1.207 (1.075, 1.357)	<0.01			1.181 (1.051, 1.328)	<0.01	
Dietary behaviours			0.815 (0.767, 0.866)	<0.001			0.831 (0.782, 0.883)	<0.001	
Subjective well-being					0.940 (0.926, 0.954)	<0.001	0.961 (0.947, 0.976)	<0.001	
Fixed parameter	0.215	<0.001	0.184	<0.001	0.591	<0.001	0.329	<0.001	
CI – confidence interval, OR – odds ratio, ref – reference

*Presented as OR (95% CI) unless specified otherwise.

Mediating effect

The test of the hypothesised path model of cognitive and social engagement based on the total sample showed that it provided a good fit to the observed data (χ2 = 24.909; P = 0.003; χ2 mean/df = 2.768; GFI = 0.999; NFI = 0.995; incremental fit index (IFI) = 0.997; TLI = 0.986; CFI = 0.996; RMSEA = 0.015; root means square residual (RMR = 0.007), SRMR = 0.010.)

Widowhood had a significant direct effect on the cognitive impairment of older adults (β = 0.118; P < 0.01), with the effect being partially mediated by social engagement, lifestyle behaviours, and subjective well-being (β = 0.075; P < 0.010). Compared to their married counterparts, widowed participants were less likely to engage in daily exercise (β = 0.028; P < 0.01) and social activities (β = 0.180; P < 0.01). Widowhood also had a significant direct effect on the social engagement of older adults (β = 0.132; P < 0.01), which was partially mediated by lifestyle behaviours and subjective well-being (β = 0.053; P < 0.01). Overall, the modified model explained 17.8% of the variance in cognitive health and 11.4% of the variance in social engagement (Figure 1, Table 5).

Figure 1 Standardised solutions for the structural model of widowhood effects on the cognitive impairment of Chinese older adults. (n = 7796). The control variables were included in the analysis but are not presented for clarity. Model fit indexes: χ2 = 24.909; P = 0.003; CMIN/DF = 2.768; CFI = 0.996; GFI = 0.999; NFI = 0.995; IFI = 0.997; TLI = 0.986; CFI = 0.996; TLI = 0.986; RMSEA = 0.015; RMR = 0.007; SRMR = 0.010. *P < 0.05. †P < 0.01. ‡P < 0.001.

Table 5 Path analysis findings: total, direct, and indirect effects (n = 7796)*

Variables	Direct effect	P-value	Indirect effect	P-value	Total effect	P-value	SMC	
Cognitive health								
Daily exercise	0.028	<0.01	0.041	<0.01	0.069	<0.01		
Dietary behaviours	−0.021		−0.009	<0.05	−0.030	<0.05		
Subjective well-being	−0.294	<0.05	−0.017	<0.05	−0.311	<0.05		
Recommended sleep duration			0.008	<0.01	0.008	<0.01		
Smoking			0.005	<0.05	0.005	<0.05		
Drinking			−0.001		−0.001			
Social engagement	0.180	<0.01			0.180	<0.01	0.178	
Widowhood	0.118	<0.01	0.075	<0.01	0.193	<0.01		
Social engagement								
Daily exercise	0.229	<0.01			0.229	<0.01		
Dietary behaviours	−0.048	<0.05			−0.048	<0.05		
Subjective well-being	−0.095	<0.01			−0.095	<0.01		
Recommended sleep duration	0.043	<0.01			0.043	<0.01		
Smoking	0.026	<0.05			0.026	<0.05		
Drinking	−0.003				−0.003		0.114	
Widowhood	0.132	<0.01	0.053	<0.01	0.185	<0.01		
SMC – squared multiple correlation

*Presented as β unless specified otherwise.

DISCUSSION

Based on data from the 2018 wave of the CLHLS, we observed that widowed older adult individuals are at a higher risk of experiencing cognitive impairment than non-widowed participants. This finding is in line with the broader existing literature [42,43]. To explore the mechanisms underlying the relationship between widowhood and cognitive function, we used the stress and marital resource models, which had been well-established through prior research [12,44]. Widows and widowers often experience multiple stressors, including relocation (e.g. many of them are asked to move from home and live with their children or relatives in post-widowhood), disruptions to daily routines, increased financial strain, and the loss of a confidant [45,46]. These stressors, along with the loss of marital resources, frequently increase the risk of depression and other chronic health issues, which can detrimentally impact cognitive function [22]. However, a recent cross-sectional study of 29 242 older adult women showed that the negative relationship of widowhood to cognitive health may gradually stabilise over a period of up to 20 years [47]. Future studies could explore how the effects of widowhood on cognitive function change over time and through the widowed individual’s lifespan.

We also further explored how lifestyle behaviours alter cognitive impairment during widowhood in the context of China, a country whose traditional ethical values and cultural norms differ from those of Western societies. We specifically focussed on the relationship between social behaviour change and cognitive impairment in the widowed group. The results of our mediation model suggest that social engagement, health behaviours, and subjective well-being may play an important role in the relationship between widowhood and cognitive impairment.

We further found that social engagement is a key mediator in how widowhood affects the cognitive health of older persons. Although prior studies did not find that social engagement attributed much to the negative impacts of widowhood on cognitive outcomes [48,49], others have suggested that widowed individuals are less likely to participate in social activities [48,50]. Moreover, social disengagement in later life may result in cognitive decline [51]. The theory of socioemotional selectivity, for example, suggests may explain how spousal loss influences social engagement. Older adults focus on fewer, but more intimate relationships to satisfy their emotional needs, which helps them retain social network stability and support [52]. However, mental stress and depression after widowhood may make widowed individuals less inclined to engage in social activities [48], which means less access to social support, social integration, and social control affecting the individual’s active participation in a specific health behaviour. These factors may damage health and well-being. For instance, lower social participation not only makes the older adult lose resources in maintaining cognitive ability [36], but also leads to a high risk of cognitive impairment. This can also create a loss of protective health behaviours stemming from social participation [38]. Therefore, families should help older adults participate in activities and support them in their integration into society, either consistently or on a regular/daily basis. Society and governments can also identify the heterogeneous personal characteristics of older adults, such as social engagement traits and marital situations, and use them to formulate effective, targeted interventions to minimise the loss of social resources.

Our results suggest that a portion of the connection between widowhood and cognitive impairment could be attributed to health behaviours. Earlier studies have shown that widowed individuals were more likely to smoke than their non-widowed counterparts [16]. Reduced sleep duration and irregular healthy food habits likewise emerged as negative effects of widowhood [15,16]. This is also consistent with the theory of socioemotional selectivity, which posits that increased stress from marital dissolutions and the absence of partner support can result in detrimental adjustments and emotional or behavioural issues, thereby potentially harming cognitive function and hastening cognitive decline in later life [53].

We also observed a decrease in physical daily exercise and an increased rate of drinking following the death of a spouse. Prior studies have not determined if or how much physical activity varies after becoming widowed, with divergent results reporting increases [14,16], decreases [15,54], or no changes [55] in physical activity during the transition to widowhood. In a four-year study of 80 944 women aged 46–71, those who became widowed and remained unmarried decreased their intake of alcohol 0.43 servings/week during the aftermath of the loss of their spouse [16]. Another study of 38 865 men aged 40–75 found an association between marital dissolution and increased alcohol consumption [15]. The benefits of marriage are known to affect men and women differently. The marital role of women is more likely to be characterised by caregiving, which may indicate that women do more housework and less physical exercise [56]. Additionally, due to traditional gender roles, women consume less alcohol overall [16]. Therefore, future research should explore the effect of gender and marital roles in this context.

Finally, we determined subjective well-being to be an essential mediating variable in the association between widowhood and cognitive impairment. The loss of a spouse tends to reduce social networking and harm social connections, with widowed older adults receiving less social support and spiritual comfort, negatively impacting their life satisfaction [57]. High levels of subjective well-being are considered a key factor in positive ageing [58]; it has also been found to have a strong relationship with specific cognitive functions [59]. However, few previous studies have tested this mediating effect of subjective well-being. In the context of global population ageing, the mental health of older people has emerged as a topic of societal importance and concern. As subjective well-being seemingly plays a mediating role in the pathway between losing a loved one and these damaging cognitive health effects, targeted interventions and preventive measures could help offset negative outcomes in the future.

Our study has several limitations, emphasizing the need for additional research to fully understand the relationship between widowhood and health. First, we used a cross-sectional design, preventing us from establishing causal inferences between widowhood and cognitive impairment. Despite this, our findings are based on a nationally representative sample, providing relevant data for health promotion among widowed older adults. Second, we did not account for factors like marriage quality and widowhood duration in our analysis, although they have been found to harm the cognitive function of elderly individuals [60]. We were also unable to determine the impact of changes in lifestyle behaviours and social participation, especially if they progressively diminish over time following the loss of a spouse. Third, previous investigations have explored that widowhood might potentially increase the mortality of elderly individuals. Our sample also did not include people who died due to health problems brought on by he loss of a spouse, which caused the survivor bias in our present samples leading to an underestimate of the detrimental impact between widowhood and cognitive function.

CONCLUSIONS

By analysing data of the 2018 wave of the CLHLS, we explored the mediating role of social engagement, health behaviours, and subjective well-being in the association between widowhood and cognitive impairment. We found that widowed older adults had a higher risk of having cognitive impairment than their married counterparts, and that the effect of widowhood on cognitive impairment was partially mediated by social engagement, lifestyle behaviours, and subjective well-being. Our findings highlight a need for policies that focus on the specific requirements of this vulnerable population, such as the maintenance of social interaction, adoption of a healthy lifestyle, improvement of subjective well-being, and provision of necessary support systems.

Acknowledgements

We thank the CLHLS team for providing data and all respondents for their contribution.

Ethics statement: The research ethics committees of Duke University and Peking University (IRB00001052e13074) approved the CLHLS study. All participants provided informed consent before enrollment. We did not require ethical approval for the reanalysis of this data, as it is publicly available at https://doi.org/10.18170/DVN/WBO7LK.

Data availability: The datasets generated and/or analysed during the current study are available in the Peking University Open research data repository: https://doi.org/10.18170/DVN/WBO7LK.

Funding: This research was funded by the Science Foundation of School of Humanities and Management, Zhejiang Chinese Medical University (ZD22002).

Authorship contributions: SJ made substantial contributions to the conception or design of the work, reviewing it critically for important intellectual content. MS has made substantial contributions to the conception and design of the work, as well as the drafting of the work. KY, JW, and YZ critically reviewed the manuscript for important intellectual content. All authors approved the final version of the manuscript prior to submission and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Disclosure of interest: The authors completed the ICMJE Disclosure of Interest Form (available upon request from the corresponding author) and disclose no relevant interests.
==== Refs
REFERENCES

1 United Nations. World Population Ageing. New York, USA: United Nations; 2017. Available: https://www.un.org/en/development/desa/population/publications/pdf/ageing/WPA2017_Highlights.pdf. Accessed: 21 August 2024.
2 Deng Q Liu W Inequalities in cognitive impairment among older adults in China and the associated social determinants: a decomposition approach. Int J Equity Health. 2021;20 :82. 10.1186/s12939-021-01422-5 33741012
3 Grolli RE Mingoti MED Bertollo AG Luzardo AR Quevedo J Réus GZ Impact of COVID-19 in the Mental Health in Elderly: Psychological and Biological Updates. Mol Neurobiol. 2021;58 :1905–16. 10.1007/s12035-020-02249-x 33404981
4 Bronskill SE Maclagan LC Walker JD Guan J Wang X Ng R Trajectories of health system use and survival for community-dwelling persons with dementia: a cohort study. BMJ Open. 2020;10 :e037485. 10.1136/bmjopen-2020-037485 32709654
5 Livingston G Huntley J Sommerlad A Ames D Ballard C Banerjee S Dementia prevention, intervention, and care: 2020 report of the Lancet Commission. Lancet. 2020;396 :413–6. 10.1016/S0140-6736(20)30367-6 32738937
6 Fan Z Lv X Tu L Zhang M Yu X Wang H Reduced social activities and networks, but not social support, are associated with cognitive decline among older Chinese adults: a prospective study. Soc Sci Med. 2021;289 :114423. 10.1016/j.socscimed.2021.114423 34597879
7 Wei Y Pei H Lin J Yan X Chen G Changes in Sleep Duration and the Risk of Cognitive Impairment Among Older Adults Aged 65 Years and Over – China, 2014–2019. Zhonghua Liu Xing Bing Xue Za Zhi. 2022;4 :941–4. 10.46234/ccdcw2022.193
8 Guang W Yan E Status of Widowed Elderly Population in China and its Development Trend. Research on Aging. 2013;1 :44–55.
9 Zhang Z Li LW Xu H Liu J Does widowhood affect cognitive function among Chinese older adults? SSM Popul Health. 2018;7 :100329. 10.1016/j.ssmph.2018.100329 30581964
10 Roh M Dan H Kim O Influencing Factors of Subjective Cognitive Impairment in Middle-Aged and Older Adults. Int J Environ Res Public Health. 2021;18 :11488. 10.3390/ijerph182111488 34770002
11 Lee Y Jean Yeung WJ Gender matters: Productive social engagement and the subsequent cognitive changes among older adults. Soc Sci Med. 2019;229 :87–95. 10.1016/j.socscimed.2018.08.024 30177360
12 Waite J, Gallagher M. The case for marriage: Why married people are happier, healthier, and better off financially. USA: The Crown Publishing; 2000.
13 Schoenborn CA Marital status and health: United States, 1999-2002. Adv Data. 2004;351 :1–32.15633583
14 Wilcox S Evenson KR Aragaki A Wassertheil-Smoller S Mouton CP Loevinger BL The effects of widowhood on physical and mental health, health behaviors, and health outcomes: The Women’s Health Initiative. Health Psychol. 2003;22 :513–22. 10.1037/0278-6133.22.5.513 14570535
15 Eng PM Kawachi I Fitzmaurice G Rimm EB Effects of marital transitions on changes in dietary and other health behaviours in US male health professionals. J Epidemiol Community Health. 2005;59 :56–62. 10.1136/jech.2004.020073 15598728
16 Lee S Cho E Grodstein F Kawachi I Hu FB Colditz GA Effects of marital transitions on changes in dietary and other health behaviours in US women. Int J Epidemiol. 2005;34 :69–78. 10.1093/ije/dyh258 15231759
17 Isherwood M King S Luszcz A Widowhood in the fourth age: support exchange, relationships and social participation. Ageing Soc. 2017;37 :188–212. 10.1017/S0144686X15001166
18 Padayachey U Ramlall S Chipps J Depression in older adults: prevalence and risk factors in a primary health care sample. South African Family Practice. 2017;59 :61. 10.1080/20786190.2016.1272250
19 Kim Y Kim S Will the Pain of Losing a Husband Last Forever? The Effect of Transition to Widowhood on Mental Health. Dev Soc. 2016;45 :165–87. 10.21588/dns.2016.45.1.007
20 Siflinger B The Effect of Widowhood on Mental Health – An Analysis of Anticipation Patterns Surrounding the Death of a Spouse. Health Econ. 2017;26 :1505–23. 10.1002/hec.3443 27747997
21 Tuinman MA Van Nuenen FM Hagedoorn M Hoekstra-Weebers JE Distress, problems and referral wish of cancer patients: differences according to relationship status and life phase. Psychooncology. 2015;24 :699–704. 10.1002/pon.3712 25345693
22 Aggarwal NT Wilson RS Beck TL Rajan KB Mendes de Leon CF Evans DA Perceived stress and change in cognitive function among adults 65 years and older. Psychosom Med. 2014;76 :80–5. 10.1097/PSY.0000000000000016 24367123
23 Morley JE Cognition and Chronic Disease. J Am Med Dir Assoc. 2017;18 :369–71. 10.1016/j.jamda.2017.02.010 28433119
24 Shin SH Kim G Park S Widowhood Status as a Risk Factor for Cognitive Decline among Older Adults. Am J Geriatr Psychiatry. 2018;26 :778–87. 10.1016/j.jagp.2018.03.013 29748078
25 Lyu J Min J Kim G Trajectories of cognitive decline by widowhood status among Korean older adults. Int J Geriatr Psychiatry. 2019;34 :1582–9. 10.1002/gps.5168 31276237
26 Zhang ZX Zahner GE Román GC Liu XH Wu CB Hong Z Socio-demographic variation of dementia subtypes in china: Methodology and results of a prevalence study in Beijing, Chengdu, Shanghai, and Xian. Neuroepidemiology. 2006;27 :177–87. 10.1159/000096131 17035714
27 Xu PR Wei R Cheng BJ Wang AJ Li XD Li HB The association of marital status with cognitive function and the role of gender in Chinese community-dwelling older adults: a cross-sectional study. Aging Clin Exp Res. 2021;33 :2273–81. 10.1007/s40520-020-01743-5 33156506
28 Ertel KA Glymour MM Berkman LF Effects of social integration on preserving memory function in a nationally representative US elderly population. Am J Public Health. 2008;98 :1215–20. 10.2105/AJPH.2007.113654 18511736
29 Ellwardt L Aartsen M Deeg D Steverink N Does loneliness mediate the relation between social support and cognitive functioning in later life? Soc Sci Med. 2013;98 :116–24. 10.1016/j.socscimed.2013.09.002 24331889
30 Biddle KD Jacobs HIL d’Oleire Uquillas F Zide BS Kirn DR Properzi MR Associations of Widowhood and β-Amyloid With Cognitive Decline in Cognitively Unimpaired Older Adults. JAMA Netw Open. 2020;3 :e200121. 10.1001/jamanetworkopen.2020.0121 32101313
31 Fratiglioni L Paillard-Borg S Winblad B An active and socially integrated lifestyle in late life might protect against dementia. Lancet Neurol. 2004;3 :343–53. 10.1016/S1474-4422(04)00767-7 15157849
32 Shahar DR Schultz R Shahar A Wing RR The effect of widowhood on weight change, dietary intake, and eating behavior in the elderly population. J Aging Health. 2001;13 :189–99. 10.1177/089826430101300202 11787511
33 Takagi D Kondo K Kawachi I Social participation and mental health: moderating effects of gender, social role and rurality. BMC Public Health. 2013;13 :701. 10.1186/1471-2458-13-701 23902596
34 Bourassa KJ Memel M Woolverton C Sbarra DA Social participation predicts cognitive functioning in aging adults over time: comparisons with physical health, depression, and physical activity. Aging Ment Health. 2017;21 :133–46. 10.1080/13607863.2015.1081152 26327492
35 Krueger KR Wilson RS Kamenetsky JM Barnes LL Bienias JL Bennett DA Social engagement and cognitive function in old age. Exp Aging Res. 2009;35 :45–60. 10.1080/03610730802545028 19173101
36 Thomas PA Trajectories of social engagement and limitations in late life. J Health Soc Behav. 2011;52 :430–43. 10.1177/0022146511411922 22144732
37 Wang HX Jin Y Hendrie HC Liang C Yang L Cheng Y Late life leisure activities and risk of cognitive decline. J Gerontol A Biol Sci Med Sci. 2013;68 :205–13. 10.1093/gerona/gls153 22879456
38 Liu H Umberson DJ The times they are a changin’: marital status and health differentials from 1972 to 2003. J Health Soc Behav. 2008;49 :239–53. 10.1177/002214650804900301 18771061
39 Li H Jia J Yang Z Mini-Mental State Examination in Elderly Chinese: A Population-Based Normative Study. J Alzheimers Dis. 2016;53 :487–96. 10.3233/JAD-160119 27163822
40 Hirshkowitz M Whiton K Albert SM Alessi C Bruni O DonCarlos L National Sleep Foundation’s sleep time duration recommendations: methodology and results summary. Sleep Health. 2015;1 :40–3. 10.1016/j.sleh.2014.12.010 29073412
41 Zhang L Bi X Ding Z Health lifestyles and Chinese oldest-old’s subjective well-being-evidence from a latent class analysis. BMC Geriatr. 2021;21 :206. 10.1186/s12877-021-02121-0 33761879
42 Bae JB Kim YJ Han JW Kim TH Park JH Lee SB Incidence of and risk factors for Alzheimer’s disease and mild cognitive impairment in Korean elderly. Dement Geriatr Cogn Disord. 2015;39 :105–15. 10.1159/000366555 25401488
43 Giri M Chen T Yu W Lü Y Prevalence and correlates of cognitive impairment and depression among elderly people in the world’s fastest growing city, Chongqing, People’s Republic of China. Clin Interv Aging. 2016;11 :1091–8. 10.2147/CIA.S113668 27574409
44 Aartsen MJ Van Tilburg T Smits CH Comijs HC Knipscheer KC Does widowhood affect memory performance of older persons? Psychol Med. 2005;35 :217–26. 10.1017/S0033291704002831 15841679
45 Zhang Z Hayward D Gender, the marital life course, and cardiovascular disease in late midlife. J Marriage Fam. 2006;68 :639 57. 10.1111/j.1741-3737.2006.00280.x
46 Liu H Marital dissolution and self-rated health: age trajectories and birth cohort variations. Soc Sci Med. 2012;74 :1107-16. 10.1016/j.socscimed.2011.11.037 22341805
47 Jain U Liu H Langa KM Farron M Kabeto M Lee J Widowhood and cognition among older women in India: New insights on widowhood duration and mediators. SSM Popul Health. 2022;19 :101242. 10.1016/j.ssmph.2022.101242 36193099
48 Feng L Ng XT Yap P Li J Lee TS Håkansson K Marital Status and Cognitive Impairment among Community-Dwelling Chinese Older Adults: The Role of Gender and Social Engagement. Dement Geriatr Cogn Dis Extra. 2014;4 :375–84. 10.1159/000358584 25473404
49 Sundström A Westerlund O Kotyrlo E Marital status and risk of dementia: a nationwide population-based prospective study from Sweden. BMJ Open. 2016;6 :e008565. 10.1136/bmjopen-2015-008565 26729377
50 Dong X Li Y Simon MA Social engagement among U.S. Chinese older adults–findings from the PINE Study. J Gerontol A Biol Sci Med Sci. 2014;69 :S82–9. 10.1093/gerona/glu152 25378453
51 Oh SS Cho E Kang B Social engagement and cognitive function among middle-aged and older adults: gender-specific findings from the Korean longitudinal study of aging (2008-2018). Sci Rep. 2021;11 :15876. 10.1038/s41598-021-95438-0 34354162
52 Carstensen LL Social and emotional patterns in adulthood: support for socioemotional selectivity theory. Psychol Aging. 1992;7 :331–8. 10.1037/0882-7974.7.3.331 1388852
53 Hayes V Demirkol A Ridley N Withall A Draper B Alcohol-related cognitive impairment: current trends and future perspectives. Neurodegener Dis Manag. 2016;6 :509–23. 10.2217/nmt-2016-0030 27855543
54 Janke C Nimrod G Kleiber A Leisure Patterns and Health Among Recently Widowed Adults. Act Adapt Aging. 2008;32 :19–39. 10.1080/01924780802039238
55 Avis NE Brambilla DJ Vass K McKinlay JB The effect of widowhood on health: a prospective analysis from the Massachusetts Women’s Health Study. Soc Sci Med. 1991;33 :1063 70. 10.1016/0277-9536(91)90011-Z 1771433
56 Umberson D Family status and health behaviors: social control as a dimension of social integration. J Health Soc Behav. 1987;28 :306–19. 10.2307/2136848 3680922
57 Mohamad N Alavi K Mohamad S Aun S Intergenerational Support and Intergenerational Social Support Among Elderly – A Short Review in Malaysian Context. Procedia Soc Behav Sci. 2016;219 :513–9. 10.1016/j.sbspro.2016.05.028
58 Rowe JW Kahn RL Successful aging. Gerontologist. 1997;37 :433 40. 10.1093/geront/37.4.433 9279031
59 Siedlecki KL Yazdani N Minahan J Falzarano F Examining processing speed as a predictor of subjective well-being across age and time in the German Aging Survey. Neuropsychol Dev Cogn B Aging Neuropsychol Cogn. 2020;27 :66–82. 10.1080/13825585.2019.1585514 30822256
60 Prior A Fenger-Grøn M Davydow DS Olsen J Li J Guldin MB Bereavement, multimorbidity and mortality: a population-based study using bereavement as an indicator of mental stress. Psychol Med. 2018;48 :1437–43. 10.1017/S0033291717002380 28851470
