Social vulnerability of different dimensions and physical frailty in older Chinese adults: a national cross-sectional study

doi:10.21203/rs.3.rs-4851434/v1

Download PDF

Research Article

Social vulnerability of different dimensions and physical frailty in older Chinese adults: a national cross-sectional study

https://doi.org/10.21203/rs.3.rs-4851434/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Backgroud

Frailty is a significant public health concern among the elderly, with social frailty closely linked to the development and progression of physical frailty. This study examines the current status and influencing factors of social frailty among older adults in China, and explores the impact of different dimensions of social frailty on the degree of physical frailty to provide insights for delaying physical frailty.

Methods

Data were sourced from the Fourth Sample Survey of the Aged Population in Urban and Rural China by the China National Committee on Aging. Physical frailty was assessed using a frailty index model, and principal component analysis determined the dimensions of social frailty. Multifactorial linear regression and logistic regression analyses were conducted to examine the relationships between various dimensions of social frailty and frailty, as well as their impacts on the degree of frailty.

Results

A total of 12,788 older adults were included in the analysis. The frailty index (FI) value was 0.14 ± 0.10, showing an exponential increase with age, and at any age, females had higher FI values than males. Principal component analysis revealed that social frailty in the elderly primarily encompasses dimensions of living conditions, social support, social participation, economic status, and living environment. Multivariate linear regression results indicated that older adults with higher degrees of frailty are more vulnerable in terms of social support, social participation, economic status, and living environment (r = 0.095, 0.198, 0.226, 0.199, all P = 0.000), while the association with living conditions was not statistically significant. Multifactorial logistic regression analysis showed that apart from the living conditions dimension, social support, social participation, economic status, and living environment are associated with an increased risk of frailty, with economic status having the most significant impact, followed by living environment, social support, or social participation.

Conclusions

Social frailty is common among the elderly, and there is a reciprocal and promotive interaction between social and physical frailty. The impacts of different dimensions of social frailty on frailty vary, suggesting that targeted interventions should be developed to address various dimensions of social frailty to mitigate its effects on the frailty of older adults.

Frailty

Social vulnerability

Dimensions of social frailty

National survey

As the global population ages, frailty has become a major public health issue among the elderly. Frailty is defined as a decline in physical strength and physiological function, which increases an individual's dependency, vulnerability, and susceptibility to death, with the accumulation of physical, psychological, and social deficits leading to a complex syndrome[1].Studies have shown that the condition of frailty in the elderly is closely related to falls, dementia, fractures, and cardiovascular diseases, and the degree of frailty greatly affects the survival rate of the elderly, imposing substantial economic and caregiving burdens on families and society[2–4].Understanding the factors influencing frailty and actively intervening early are crucial for delaying the progression of frailty and improving the quality of life in the elderly.

In recent years, with the shift in the medical model, the concept of frailty has evolved from a singular emphasis on physical frailty to a more comprehensive one that includes multiple dimensions (physiological, cognitive, social, or psychological). In the context of an aging society, the elderly increasingly rely on social relationships and the social environment, leading to a growing incidence of social frailty, which can be as high as 66.5% [5]. In 2017, Bunt et al [6]. first defined social frailty as an individual's continuous state of losing one or several vital resources that satisfy basic social needs. The composition of the social frailty concept not only includes the lack of social resources to meet basic social needs but also encompasses a deficiency in social behaviors and activities, as well as a lack of self-management capability. Social frailty in the elderly is associated with cognitive and psychological function decline, reduced quality of life, increased disability and mortality risk [7–9].Many studies consider social frailty to be a multi-dimensional concept, weakened or deficient in aspects such as social resources, social behaviors, social activities, social networks, social functions, and living environments [6, 10–12]; and it is closely related to the occurrence and development of physical frailty. Social frailty can precede physical frailty and lead to its occurrence, with the two interacting and promoting each other [13].Currently, research on frailty among the Chinese elderly still largely focuses on the physical function aspect, with less research on social frailty and few studies that analyze the connection between social and physical frailty based on different dimensions of social frailty. This study utilizes the database from the Fourth Sample Survey of the Aged Population in Urban and Rural China to analyze the current status and influencing factors of social frailty among the elderly, and further explores the relationship between different dimensions of social frailty and physical frailty, as well as their impact on the degree of frailty. The aim is to provide references and bases for early identification of social frailty in the elderly and targeted interventions to delay the onset of physical frailty.

Study Design and Participants

The data for this study were sourced from the database of the Fourth Sample Survey of the Aged Population in Urban and Rural China, which was organized by the China National Committee on Aging. This nationwide survey of the elderly population was conducted from August 1 to August 31, 2015, and targeted Chinese citizens aged 60 and above residing in mainland China. To ensure the representativeness of the sample, the survey employed a multistage, stratified, probability-proportional-to-size (PPS) sampling design with equal probability, which had been introduced in previous studies[14–15].The designed sample size was 223,680 individuals (with a total sampling ratio of approximately 1‰). The survey covered all provinces, autonomous regions, municipalities directly under the Central Government, and the Xinjiang Production and Construction Corps, involving 466 counties (districts), 1,864 townships (streets) (4 townships (streets) from each selected county (district)), and 7,456 villages (neighborhood committees) (4 village (neighborhood) committees from each selected township (street)). Excluding those who refused to participate, were deceased, could not be contacted (after at least 3 attempts), or were residing long-term in nursing institutions during the survey period, the actual number of samples collected was 224,142 older individuals. This survey achieved a nationwide coverage goal for the first time and is to date the largest database of the elderly in the country. The research protocol was approved by the National Bureau of Statistics [No. (2014)87] and the Ethics Committee of Beijing Hospital (2021BJYYEC-294-01). All participants provided written informed consent prior to completing the survey.

Subsequently, the National Committee on Aging conducted annual follow-up surveys from 2016 to 2019 to dynamically monitor the living conditions and health status of the elderly based on the 2015 survey sample. This study focuses on the database of the elderly from the 2017 follow-up survey, which included 12,788 individuals aged between 62 and 101 years, with an average age of (71.7 ± 7.4) years. This group comprised 6,209 females, with an average age of (71.9 ± 7.6) years, and 6,579 males, with an average age of (71.4 ± 7.1) years.

Survey Content

The data collection involved household interviews and questionnaires, encompassing demographic characteristics (age, gender, ethnicity, education level, marital status, etc.), family status (living alone, children's status, etc.), health and medical conditions (vision, hearing, exercise habits, consumption of health products, physical examination history, chronic diseases, medical treatment and insurance status, self-rated health, etc.), care and nursing services [activities of daily living (ADL) capability, incontinence, use of assistive devices, care and nursing needs and willingness, etc.], economic status (income, expenses, home ownership, etc.), living environment (housing area, housing conditions, satisfaction with living conditions, etc.), social participation (involvement in public welfare activities, willingness to help other elderly in need, etc.), rights protection (awareness of elderly rights, whether legal rights are safeguarded, etc.), and cultural and spiritual life (participation in community activities, frequent internet use, etc.). Diseases had to be diagnosed by hospitals at the county level or above, and cases with symptoms but no definite diagnosis were not included in the statistics.

Frailty Assessment

Frailty was evaluated using the Frailty Index (FI) model developed by a team led by Professor Kenneth Rockwood, a geriatric medicine expert from Canada. This method is based on the accumulation of health deficits and quantitatively describes an individual's degree of frailty[16].The FI is calculated as the number of health deficits an individual has divided by the total number of considered health deficit items, with a value range of 0 to 1. Higher values indicate greater frailty.¹⁷ Based on the survey questionnaire, this study selected 31 variables considered as health deficit items, satisfying the conditions for constructing the FI [17].These variables include activities of daily living (ADL) (6 items), chronic diseases (11 items), geriatric syndromes (5 items), health status and mood (4 items), and use of assistive devices (5 items). The variables were scored according to their type, with specific variables and their scores presented in Supplementary Table 1. Following the studies by Thompson et al [18].and Hoover et al [19], the FI is categorized into robust (FI ≤ 0.10), prefrail (0.10 < FI ≤ 0.21), and frail (FI > 0.21) based on threshold values of 0.10 and 0.21.

Social Vulnerability Variables

The selection of items was based on the survey questionnaire, and given the complexity and diversity of social factors, the study aimed to include as many different social factors as possible. Ultimately, 21 variables were included such as marital status, living alone, medical insurance, reimbursement of medical expenses, availability of care when ill, engagement in working, retirement funds with spouse, ownership of a house, economic status, etc., as shown in Table 1. Each variable was scored according to its type; for binary variables, the presence of a social deficit was scored as 1, and absence as 0. For ordinal variables, intermediate values were set [20]. For example, for the variable of economic status, responses of "very wealthy," "relatively wealthy," "just enough," "relatively difficult," and "very difficult" were scored as 0, 0.25, 0.5, 0.75, and 1, respectively. In this way, each variable's frailty score ranged from 0 to 1, and the sum of the scores divided by the total number of considered social deficit items yielded the Social Vulnerability Index (SVI). Principal component analysis (PCA) was used to determine the dimensions of social frailty. The 0–1 scores for the social frailty variables within each dimension were summed and then divided by the total number of items considered as social deficits within that dimension to arrive at the index for that particular dimension.

Table 1

Demographics and the social vulnerability-related factors of older adults by frailty stage
Variables	All					Women					Men
Variables	Total	Robust	Prefrail	Frail	F/x² (P-value)	Total	Robust	Prefrail	Frail	F/x² (P-value)	Total	Robust	Prefrail	Frail	F/x² (P-value)
N	12788	4832	5652	2304		6209	1971	2918	1320		6579	2861	2734	984
Age(x ± s)	71.7 ± 7.4	69.5 ± 6.3	72.0 ± 7.1	75.6 ± 8.2	587.812 (< 0.001)	71.9 ± 7.6	69.3 ± 6.3	72.1 ± 7.2	75.7 ± 8.3	319.279 (< 0.001)	71.4 ± 7.1	69.6 ± 6.3	71.9 ± 6.9	75.3 ± 8.1	258.855 (< 0.001)
Agegroup					971.569 (< 0.001)					561.963 (< 0.001)					399.526 (< 0.001)
60-	6407	48.2%	41.1%	10.7%		2868	42.9%	44.6%	12.6%		3179	52.9%	38.0%	9.0%
70-	4533	32.5%	48.2%	19.3%		2180	26.0%	51.7%	22.3%		2353	38.5%	45.0%	16.4%
80-	2208	20.2%	44.3%	35.5%		1161	15.1%	44.2%	40.7%		1047	26.0%	44.4%	29.6%
Gender					208.387 (< 0.001)
Women	6209	31.7%	47.0%	21.3%
Men	6579	43.5%	41.5%	15.0%
Ethnicity					8.955 (0.011)					7.864 (0.020)					9.684 (0.008)
Han	12166	38.1%	44.0%	17.9%		5924	32.0%	46.6%	21.4%		6242	43.8%	41.5%	14.7%
Minority	622	32.2%	48.7%	19.1%		285	26.7%	55.1%	18.2%		337	36.8%	43.3%	19.9%
Residence place					13.014 (0.001)					15.962 (< 0.001)					5.139(0.077)
Urban	6496	39.1%	43.8%	17.1%		3287	33.9%	46.0%	20.1%		3209	44.5%	41.5%	14.0%
Rural	6292	36.4%	44.6%	19.0%		2922	29.3%	48.1%	22.6%		3370	42.5%	41.6%	15.8%
Education					386.713 (< 0.001)					191.599 (< 0.001)					97.537(< 0.001)
Illiteracy	9133	42.4%	42.7%	14.9%		3834	37.3%	45.7%	17.0%		5299	46.1%	40.6%	13.4%
Primary/Junior high school	3353	25.6%	47.7%	26.7%		2309	22.6%	48.9%	28.5%		1044	32.4%	45.1%	22.5%
Senior high school or above	189	33.9%	51.3%	14.8%		11	18.2%	63.6%	18.2%		178	34.8%	50.6%	14.6%
Social vulnerability-related factors
Marital status					369.183 (< 0.001)					190.485 (< 0.001)					89.856(< 0.001)
Married	9133	42.4%	42.7%	14.9%		3834	37.3%	45.7%	17.0%		5299	46.1%	40.6%	13.4%
Others^a	3542	26.1%	47.9%	26.0%		2320	22.6%	49.0%	28.4%		1222	32.7%	45.9%	21.4%
Living alone	2093	29.1%	49.5%	21.4%	81.869 (< 0.001)	1183	23.7	51.7%	24.6%	44.682 (< 0.001)	910	36.2%	46.7%	17.1%	23.247(< 0.001)
Having medical insurance	12681	37.8%	44.2%	18.0%	4.524 (0.104)	6150	31.8%	47.0%	21.2%	7.824 (0.020)	6531	43.5%	41.6%	15.0%	0.163(0.922)
Reimbursement of medical expenses					84.570 (< 0.001)					46.389 (< 0.001)					54.447(< 0.001)
Very convenient	5660	39.4%	43.7%	16.9%		2716	33.5%	47.3%	19.2%		2944	44.7%	40.4%	14.8%
Relatively convenient	3649	36.3%	44.9%	18.8%		1790	29.3%	46.8%	23.9%		1859	42.9%	43.1%	14.0%
Average	1177	31.0%	47.8%	21.2%		588	27.0%	50.7%	22.3%		589	35.0%	45.0%	20.0%
Relatively inconvenient	295	23.7%	47.8%	28.5%		141	18.4%	51.1%	30.5%		154	28.6%	44.8%	26.6%
Very inconvenient	133	21.8%	48.9%	29.3%		63	17.5%	46.0%	36.5%		70	25.7%	51.4%	22.9%
Availability of care when ill	1695	2.5%	29.2%	68.3%	139.330 (< 0.001)	927	2.2%	27.9%	69.9%	59.879 (< 0.001)	768	2.9%	30.9%	66.3%	80.086(< 0.001)
Engagement in working	1206	62.4%	34.2%	3.4%	398.428 (< 0.001)	339	52.2%	41.6%	6.2%	87.784 (< 0.001)	867	66.3%	31.3%	2.4%	249.214(< 0.001)
Retirement funds with spouse	4284	48.0%	41.2%	10.8%	378.376 (< 0.001)	1953	41.8%	44.8%	13.4%	177.656 (< 0.001)	2331	53.2%	38.3%	8.6%	185.287(< 0.001)
Ownership of a house	9106	41.1%	43.9%	15.0%	246.967 (< 0.001)	4121	35.5%	46.6%	17.9%	123.278 (< 0.001)	4985	45.7%	41.6%	12.7%	91.249(< 0.001)
Economic status					1293.369 (< 0.001)					609.880 (< 0.001)					688.871(< 0.001)
Very wealthy	256	70.3%	23.8%	5.9%		110	64.5%	26.4%	9.1%		146	74.7%	21.9%	3.4%
Relatively wealthy	2210	54.4%	37.1%	8.5%		1007	48.0%	42.0%	10.0%		1203	59.8%	33.0%	7.2%
Just enough	7551	39.5%	45.3%	15.2%		3729	33.0%	48.6%	18.4%		3822	45.7%	42.2%	12.0%
Relatively difficult	2298	17.8%	50.0%	32.2%		1123	13.7%	50.6%	35.7%		1175	21.6%	49.5%	28.9%
Very difficult	437	10.5%	40.5%	49.0%		222	9.5%	35.6%	55.0%		215	11.6%	45.6%	42.8%
Time of the house built					75.205 (< 0.001)					24.897 (< 0.001)					55.973(< 0.001)
After 2000s	5029	38.8%	43.5%	17.7%		2451	31.9%	47.1%	20.9%		2578	45.3%	40.1%	14.5%
In the 1990’ s	3635	41.0%	43.3%	15.7%		1757	34.8%	45.9%	19.4%		1878	46.8%	40.9%	12.3%
1970s-1980s	3382	34.9%	45.6%	19.6%		1630	29.6%	48.1%	22.3%		1752	39.8%	43.2%	17.0%
Before liberation /1950s-1960s	729	28.0%	47.3%	24.7%		364	25.0%	47.3%	27.7%		365	31.0%	47.4%	21.6%
Housing area(m²)					101.531 (< 0.001)					41.023 (< 0.001)					67.963(< 0.001)
≥ 145	2632	42.7%	41.8%	15.5%		1300	35.8%	44.7%	19.5%		1332	49.5%	39.0%	11.5%
91–144	3495	39.7%	44.1%	16.2%		1733	34.0%	47.1%	18.9%		1762	45.3%	41.1%	13.6%
61–90	3338	37.5%	43.9%	18.5%		1560	31.0%	46.9%	22.2%		1778	43.3%	41.4%	15.3%
< 60	3183	31.5%	46.9%	21.6%		1553	26.4%	49.3%	24.3%		1630	36.3%	44.7%	19.0%
Having private living rooms	12144	38.2%	44.0%	17.8%	16.758 (< 0.001)	5873	32.0%	46.9%	21.1%	4.549 (0.103)	6271	43.9%	41.3%	14.8%	11.272(0.004)
No problem with living conditions	6816	46.1%	41.7%	12.3%	557.399 (< 0.001)	3221	39.6%	45.7%	14.7%	272.522 (< 0.001)	3595	51.8%	38.1%	10.1%	275.679(< 0.001)
Satisfaction with living conditions					331.884 (< 0.001)					152.223 (< 0.001)					182.490(< 0.001)
Satisfied	7404	42.9%	42.2%	14.9%		3546	36.6%	45.8%	17.6%		3858	48.7%	39.0%	12.3%
Average	4166	33.4%	46.3%	20.3%		2070	27.2%	48.5%	24.3%		2096	39.6%	44.1%	16.3%
Not satisfied	1121	20.9%	47.9%	31.2%		550	17.6%	47.8%	34.5%		571	24.0%	48.0%	28.0%
Participating in public welfare activities	6497	43.3%	44.1%	12.6%	324.284 (< 0.001)	2939	37.1%	47.3%	15.6%	133.553 (< 0.001)	3558	48.5%	41.4%	10.1%	168.644(< 0.001)
Participating in seniors association	2199	45.2%	42.2%	12.6%	83.700 (< 0.001)	1014	37.6%	46.4%	16.1%	28.318 (< 0.001)	1185	51.7%	38.6%	9.7%	52.399(< 0.001)
Willingness to help other elderly in need	10026	40.4%	44.1%	15.5%	251.857 (< 0.001)	4787	34.2%	47.2%	18.7%	105.077 (< 0.001)	5239	46.2%	41.3%	12.6%	139.957(< 0.001)
Legal rights are protected	11977	38.3%	44.1%	17.5%	43.702 (< 0.001)	5820	32.1%	47.3%	20.6%	30.836 (< 0.001)	6157	44.2%	41.2%	14.6%	20.853(< 0.001)
Participating in recreational activities	11889	39.2%	44.7%	16.1%	436.610 (< 0.001)	5624	33.4%	47.7%	18.8%	228.089 (< 0.001)	6265	44.4%	41.9%	13.7%	167.412(< 0.001)
Frequent internet use	973	50.8%	42.0%	7.2%	116.576 (< 0.001)	376	43.9%	48.4%	7.7%	53.296 (< 0.001)	597	55.1%	38.0%	6.9%	51.090(< 0.001)
Attending a college/school for older adults	347	45.2%	43.8%	11.0%	15.076 (0.001)	171	38.0%	48.5%	13.5%	7.305 (0.026)	176	52.3%	39.2%	8.5%	8.371(0.015)
Note: ^aOther marital status includes widowed, divorced or never married

Statistical Methods

Data analysis and graphing were performed using SPSS 24.0 and Matlab 2020 software. Missing data were addressed using the multiple imputation method with the Markov Chain Monte Carlo (MCMC) technique [21].Quantitative data were expressed as x̄±s, and comparisons among multiple groups were conducted using analysis of variance (ANOVA). Categorical data were presented as number (percentage) and compared using the chi-squared (χ²) test. PCA was utilized to recombine social frailty-related variables into non-correlated dimensions of social frailty, facilitating a comprehensive understanding of the interrelations among the dimensions and their impact on frailty. Nonlinear regression techniques were applied to fit age-specific frailty index values as functions of age (exponential functions), and linear regression was used to fit the mean values of the dimensions of SVI as functions of age for older adults of different genders. Multifactorial linear regression analysis was used to examine the relationships of various dimensions of social frailty with general demographic characteristics such as age, gender, ethnicity, place of residence, and educational level, as well as with frailty. Using robust as a reference, multinomial logistic regression, adjusted for age, gender, ethnicity, place of residence, and educational level, analyzed the impact of various dimensions of social frailty on the degree of frailty among the elderly. A P-value of < 0.05 was considered statistically significant.

Comparison of General Characteristics and Social Frailty-Related Factors Among Elderly with Different Frailty States

Out of the 12,788 older adults, those who are robust, prefrail, and frail accounted for 37.8% (4,832/12,788), 44.2% (5,652/12,788), and 18.0% (2,304/12,788) respectively. Among the general characteristics, older age, female gender, minority ethnicity, rural residence, and lower education levels were associated with a higher proportion of prefrail and frail elderly. Further gender-based comparison of the general characteristics of the elderly with different frailty states revealed that all variables, except for the place of residence among older men, were statistically significant in influencing the frailty status of both older men and older women (all P < 0.05). Additionally, in terms of social frailty-related factors, all factors impacted the degree of frailty in older adults, regardless of gender, except for having medical insurance and having a separate living room; see Table 1.

Principal Component Analysis of Social Frailty Dimensions

To summarize the main dimensions of factors related to social frailty, PCA was performed on 21 factors, extracting components with eigenvalues greater than 1 using the varimax rotation method. This resulted in five factors with a cumulative contribution rate of 71.27%. The rotated factor loadings are presented in Table 2, with most factor loadings above 0.4, primarily concentrated in the five dimensions of living conditions, social support, social participation, economic status, and living environment. See Table 2.

Table 2

Principal component analysis with five social vulnerability dimensions
Variables	Rotated Component Matrix
	Component
	1	2	3	4	5
	Living conditions	Social support	Social participation	Economic status	Living environment
Living alone	0.912*	0.057	0.026	-0.016	-0.003
Marital status	0.839*	0.083	-0.032	-0.031	0.015
Availability of care when ill	0.003	0.586*	0.026	-0.016	0.023
Having medical insurance	0.034	0.461*	0.027	0.046	0.121
Reimbursement of medical expenses	0.046	0.432*	-0.016	0.022	-0.102
Legal rights properly protected	0.163	0.328*	0.151	0.053	-0.122
Participating in recreational activities	0.124	0.097	0.735*	0.032	-0.205
Participating in volunteer activities	-0.014	0.086	0.657*	-0.061	-0.039
Willing to help other older adults	0.109	0.011	0.629*	0.136	0.083
Participating in seniors association	0.043	0.193	0.572*	0.025	-0.178
Attending a college/school for older adults	0.019	-0.017	0.503*	-0.059	0.072
Often surfing the internet	0.263	0.141	0.419*	-0.069	0.066
Economic status	0.145	-0.001	0.006	0.693*	0.043
Retirement funds with spouse	0.092	0.039	0.027	0.638*	0.150
Engagement in working	0.007	0.016	0.022	0.420*	0.037
Ownership of a house	0.096	0.071	-0.019	0.356*	-0.021
Having private living rooms	0.035	-0.064	0.034	0.093	0.519*
No problem with living conditions	0.105	0.138	0.003	0.045	0.457*
Satisfied with living conditions	-0.122	-0.344	-0.231	0.002	0.412*
Housing area	0.001	0.123	0.006	-0.118	0.383*
Time of the house built	-0.007	0.047	0.085	-0.015	0.356*
Each variable’s factor loading is indicated by*.
Extraction method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. Rotation Converged in 6 iterations.

Trends in FI Values and the Indices of Five Social Frailty Dimensions by Age and Gender

The FI value for the 12,788 elderly participants was 0.14 ± 0.10, with women having a higher FI value of 0.16 ± 0.10 compared to men with 0.13 ± 0.09 (t = 14.163, P = 0.000). Analysis of the trends in FI values with age by gender indicated that both elderly men and women showed an exponential increase in FI values with age: ln(FI) = A + B×age, where for elderly men ln(FI)=-3.946 + 0.027×age (r = 0.912, P = 0.000) and for elderly women ln(FI)=-3.576 + 0.024×age (r = 0.933, P = 0.000). At all ages, women had higher FI values than men, indicating that elderly women have a higher level of frailty. On a logarithmic scale, the average annual relative increase in health deficits and FI values for elderly women was lower than that for men (B = 0.024 compared to B = 0.027), suggesting that elderly men accumulate health deficits more rapidly than women, as shown in Fig. 1.

Further analysis of the trends in the indices of the various dimensions of social frailty with age by gender showed that in aspects such as living conditions, social participation, economic status, and living environment, the social frailty index for both elderly men and women exhibited a linear increase with age: Mean(SVI) = A + B×age. However, for the dimension of social support, only the social frailty index for elderly men showed a linear increasing trend with age, while the trend for elderly women was not statistically significant (r = 0.180, P = 0.137). Moreover, at all ages, women had higher social frailty indices than men in living conditions, social participation, and economic status dimensions, with significant differences (t = 66.130, 10.947, 90.288, P = 0.000, 0.001, 0.000), while there were no significant differences in social support and living environment dimensions (t = 0.429, 1.380, P = 0.514, 0.244). Additionally, the average annual relative increase in the indices of living conditions and social participation for elderly women was lower than for men (B = 0.0128 compared to B = 0.0130, B = 0.0070 compared to B = 0.0078), while it was higher in the economic status dimension (B = 0.0081 compared to B = 0.0042). See Fig. 2.

Multivariate Linear Regression Analysis of the Relationship Between Social Frailty Dimensions and General Demographic Characteristics and Frailty

Further multivariate linear regression analyses were conducted with the indices of social frailty dimensions as dependent variables and age, gender, ethnicity, place of residence, educational attainment, and degree of frailty as independent variables. The results showed that the vulnerability in the dimensions of living conditions, social participation, and economic status increased with age (all P = 0.000), while vulnerability in social support decreased with age (P = 0.002). Compared to elderly men, elderly women were more vulnerable in the aspects of living conditions, social participation, and economic status (all P = 0.000). Minority elderly showed less vulnerability in living conditions but were more vulnerable in social support, social participation, and economic status (all P = 0.000). Rural elderly were more vulnerable than urban counterparts in the dimensions of social support, social participation, economic status, and living environment (all P = 0.000); elderly with lower levels of education exhibited greater vulnerability across all dimensions (all P = 0.000); and those with higher degrees of frailty were more vulnerable in social support, social participation, economic status, and living environment (all P = 0.000), with no statistically significant correlation with living conditions. See Table 3.

Table 3

Multifactorial linear regression analysis of the relationship between social vulnerability index (SVI) dimensions and demographic characteristics or frailty [β’(P-value)]
Variables	Living conditions	Social support	Social participation	Economic status	Living environment
Agegroup (Increasing)	0.025(0.000)	-0.030(0.002)	0.139(0.000)	0.090(0.000)	0.000(0.973)
Gender(Male)	-0.019(0.000)	0.003(0.728)	-0.047(0.000)	-0.074(0.000)	0.013(0.145)
Ethnicity (Minority)	-0.026(0.000)	0.035(0.000)	0.047(0.000)	0.066(0.000)	-0.002(0.802)
Residence place(Rural)	0.004(0.370)	0.036(0.000)	0.076(0.000)	0.183(0.000)	0.040(0.000)
Education(Higher)	-0.880(0.000)	-0.036(0.000)	-0.088(0.000)	-0.177(0.000)	-0.045(0.000)
Frailty (Increasing)	0.011(0.307)	0.095(0.000)	0.198(0.000)	0.226(0.000)	0.199(0.000)

Logistic Regression Analysis of the Impact of Social Frailty Dimensions on the Degree of Frailty

A multifactorial logistic regression analysis was performed with the degree of frailty in the elderly as the dependent variable and the indices of social frailty dimensions as independent variables, adjusting for age, gender (female = 0, male = 1), ethnicity (Han = 1, minority = 2), place of residence (urban = 1, rural = 2), and educational attainment (illiterate = 1, primary or junior high = 2, high school or above = 3). The results indicated that an increase in age and being female increased the risk of frailty in the elderly. Among the dimensions of social frailty, all except living conditions were associated with an increased risk of frailty, with the economic status dimension having the greatest impact, followed by living environment, social support, or social participation. See Table 4.

Table 4

Logistic regression analysis of the impact of social vulnerability index (SVI) dimensions on the degree of frailty
Variables	Pre-frailty vs. Robust			Frailty vs.Robust
Variables	β value	OR(95%CI)	P value	β value	OR(95%CI)	P value
Agegroup (Increasing)	0.392	1.480(1.393–1.573)	< 0.001	0.832	2.298(2.124–2.486)	< 0.001
Gender(Male)	-0.368	0.692(0.638–0.752)	< 0.001	-0.518	0.595(0.532–0.666)	< 0.001
Ethnicity (Minority)	0.161	1.175(0.969–1.425)	0.101	-0.094	0.910(0.701–1.181)	0.478
Residence place(Rural)	-0.040	0.961(0.885–1.043)	0.342	-0.097	0.908(0.811–1.016)	0.092
Education(Higher)	-0.018	0.983(0.811–1.190)	0.857	0.082	1.085(0.851–1.383)	0.510
SVI dimensions
Living conditions	0.174	1.190(0.920–1.540)	0.185	-0.200	0.819(0.590–1.137)	0.233
Social support	1.023	2.780(1.915–4.036)	< 0.001	1.334	3.796(2.399–6.007)	< 0.001
Social participation	0.404	1.498(1.175–1.909)	0.001	2.077	7.977(5.721–11.125)	< 0.001
Economic status	1.421	4.141(3.371–5.088)	< 0.001	3.085	21.877(16.261–29.433)	< 0.001
Living environment	1.294	3.646(2.993–4.442)	< 0.001	2.317	10.148(7.829–13.155)	< 0.001

In recent years, an increasing number of geriatric medical researchers have begun to study the issue of social frailty in the elderly. This study utilized the large sample database from the Fourth Sample Survey of the Aged Population in Urban and Rural China to analyze the status of social and physical frailty among the elderly. The study applied PCA to precisely divide the indicators of social frailty into different dimensions and to analyze the correlation between these dimensions and frailty. PCA is a widely applied multivariate statistical method for comprehensive evaluation, which uses the concept of dimensionality reduction to extract several meaningful principal components that do not overlap in information content from multiple indicators. This transformation of correlated indicators into a smaller number of independent composite indicators effectively reduces bias caused by overlapping information [22]. In this study, PCA was utilized to reduce 21 evaluation indicators related to social frailty to 5 principal components, providing a more concise and accurate description of the correlation between the dimensions of social frailty and frailty, which is beneficial for enhancing the efficiency and targeting of interventions.

The study analyzed the trends in frailty status among elderly men and women with age. The results showed that the FI values increased exponentially with age for both elderly men and women, and that at any age, elderly women were more frail than men, consistent with our previous research findings [23–24]. Generally, elderly women tend to exhibit various symptoms of frailty more readily than elderly men. Some researchers believe that changes in the levels of sex hormones in the elderly play a role in the onset of frailty, with the decline in sex hormone levels being directly related to muscle weakness, fatigue, and reduced function in frail older individuals. After menopause, elderly women experience a significant decrease in estrogen levels, which can lead to physical symptoms such as fatigue, insomnia, and headaches, as well as psychological symptoms like anxiety and depression [25]. In elderly men, frailty is closely related to androgen levels, with higher serum levels of free testosterone and dihydrotestosterone reducing the risk of frailty [26]. The trend analysis of social frailty with age in the elderly showed that as age increases, multiple dimensions of social frailty exhibit a trend of increasing vulnerability with age, which is consistent with current literature [6, 12]. The only exception was that the trend in social support was not pronounced among elderly women, which may be related to the age-related variables included in the social support dimension in this study having a lesser impact on age. Future research should enrich the content of survey questionnaires to include more relevant information for analysis. Elderly women are more vulnerable in several dimensions of social frailty compared to men, consistent with the findings of Park et al [27]. and Shah et al [28]., and in line with the subsequent multifactorial regression analysis results of this study. This could be because most women are primarily engaged in managing household affairs and caring for family members, resulting in elderly women having less time for leisure activities and social contacts, thus making them more susceptible to social frailty. Therefore, elderly women should maintain a cheerful mood, reserve some leisure time for entertainment and social activities, and their family members and community healthcare workers should also pay more attention to their physiological, psychological conditions, and social relationships, taking measures in a timely manner.

Further multifactorial linear regression analysis was conducted to identify factors influencing the various dimensions of social frailty among the elderly. The results indicated that as age increases, the vulnerability of older adults in terms of living conditions, social participation, and economic status also increases. The older the individuals, the more likely they are to live alone or be widowed, and the less they engage in social participation and activities [29].In contrast, vulnerability in social support decreases with age, meaning that older adults receive more social support as they age. On one hand, as individuals age, their adult children may pay more attention to their parents and provide more emotional and material support. On the other hand, the elderly, particularly those of advanced age, may be more inclined to participate in community activities and volunteer services, not only increasing their social interactions but also potentially garnering more social support. Additionally, older adults may have access to more comprehensive medical and social welfare services, thereby obtaining various forms of social support. When comparing different genders, research by Andrew et al [12].suggests that elderly women have higher levels of social participation and support but are more vulnerable in their living conditions (i.e., living alone and/or being single or widowed). The findings of this study indicate that elderly women are more vulnerable than men in living conditions (living alone and/or being single or widowed), and they have higher vulnerability in social participation and economic status. This may be related to cultural and societal constraints in China that lead to gender differences in income and social network differences; although elderly women may have extensive social networks, these are likely based more on family than on professional or social organizations [30]. Ethnic minorities and rural older adults are more vulnerable than Han Chinese and urban-dwelling older adults in terms of social support, social participation, and economic status. This aligns with findings by Qi et al[31]. and may be due to cultural differences or geographic isolation, which results in a more limited social network, fewer social activities, and fewer sources of social support for these groups. Additionally, rural older adults exhibit greater frailty in the living environment dimension, suggesting that this group faces more challenges in housing safety, convenience, and comfort. Health management for rural older adults should pay special attention to improvements in their living environment. Elderly individuals with lower levels of education display greater vulnerability across all dimensions, consistent with literature on the importance of education in many social measures [32–33]. Those with lower levels of education may have a weaker capacity to accept new things and fewer avenues to obtain information, leading to a more monotonous daily life, a significant decline in role function, and consequently, the onset of social frailty [34]. Older adults with higher degrees of frailty are more vulnerable in terms of social support, social participation, economic status, and living environment. This is related to the decline in physical function among frail older individuals, which limits mobility in living spaces, restricts outdoor social activities and social networks, and thus makes social frailty more likely to occur. Therefore, interventions tailored to the characteristics of different groups of older adults (varying by age, gender, ethnicity, residence, education level, and degree of frailty) can be developed based on the vulnerability of different dimensions of social frailty to establish appropriate intervention targets and effective measures.

To further explore the impact of the various dimensions of social frailty on the degree of frailty, this study conducted logistic regression analysis, adjusting for confounding factors such as age, gender, and years of education. The results show that the economic status dimension has the most substantial impact on the degree of frailty, followed by living environment, social support, or social participation. Older adults with better economic status usually have higher health literacy, access to a broader selection of medical resources that are of higher quality, which positively influences disease prevention and self-management of health, and helps in delaying aging. Additionally, economic stability can reduce the psychological stress experienced by older adults in financial aspects, often correlating positively with stronger social networks and support mechanisms, which are beneficial for mental health [35]. Safety features in the living environment, such as handrails, barrier-free access, and non-slip flooring, can reduce the occurrence of falls and other accidental injuries; a well-designed and organized living environment may also decrease psychological stress and slow cognitive and emotional decline [36]. Social support can effectively reduce psychological stress and anxiety levels in the elderly, contributing to mental health and emotional stability. Increased participation in social activities can enhance their sense of social identity and self-efficacy, and interactions with others during these activities can help slow the decline in cognitive function, playing a significant role in preventing frailty [37].

This study still has some limitations that need to be considered. Firstly, the variables included for social frailty are based on self-reported data from the Survey on Living Conditions of Urban and Rural Older Persons, rather than objectively defined social factors, which might introduce information bias. However, self-perceived vulnerability among the elderly may be more related to their health status than objective measurement standards [38]. Secondly, as a cross-sectional study, the analysis of related factors can only reflect correlations between variables and cannot infer causation. Furthermore, the variables included in the analysis are based on the content of the survey questionnaire and may not comprehensively reflect the conditions of all dimensions of social frailty. Future research should aim to improve the survey questionnaire and supplement relevant information for a more in-depth analysis.

Social frailty is common among the elderly and has become a prominent issue affecting their health and quality of life. Social and physical frailty mutually influence and promote each other, and the impact of different dimensions of social frailty on frailty varies. The dimensions of social frailty among older adults also differ; therefore, when analyzing and studying social frailty among the elderly, it is necessary to consider various factors comprehensively. It is also important to recognize the uniqueness of each individual, and targeted interventions should be developed to address the different dimensions of social frailty to reduce its impact on the frailty of older adults.

Ethics approval and consent to participate

In this study, all methods were carried out in accordance with relevant guidelines and regulations or declaration of Helsinki. This study was approved by the Ethics Review Committee of Beijing Hospital (2021BJYYEC-294-01). The participants provided their written informed consent to participate in this study.

Consent for publication

Not applicable.

Availability of data and materials

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.

Funding

This study was supported by grants from the National Key R&D Program of China [Grant nos.2020YFC2003000 and 2020YFC2003001], the National High Level Hospital Clinical Research Funding (Grant No. BJ-2023-018).

Authors’ contributions

JS, CG: had full access to all of the data, performed statistical analysis of the data, and drafted the paper. YT, YC: Assisted with the data analysis and manuscript preparation. LZ, SL, YL, BS, XL, QM: Data cleansing and processing. XZ: Assisted with the data analysis and result interpretation. HL, DL: Initiated and designed the study, revised the paper, and provided final approval for the version to be published. All authors had full access to all the data and contributed to the manuscript revision as well as read and approved the submitted version.

Acknowledgments

The authors thank the participants of the fourth SSAPUR study and the National Key R&D Program of China for their assistance.

Author details

¹The Key Laboratory of Geriatrics, Beijing Institute of Geriatrics, Institute of Geriatric Medicine, Chinese Academy of Medical Sciences, Beijing Hospital/National Center of Gerontology of National Health Commission, Beijing 100730, China.

²Department of Gastroenterology, China-Japan Friendship Hospital, Beijing 100029, China. ³Department of Cardiology, Beijing Hospital, National Center of Gerontology, Institute of Geriatric Medicine, Chinese Academy of Medical Sciences, Beijing 100730, China. ⁴Peking Union Medical College, Chinese Academy of Medical Sciences, Graduate School of Peking Union Medical College, Beijing 100730, China. ⁵Peking University Fifth School of Clinical Medicine, Beijing, China.⁶Savaid Medical School, University of Chinese Academy of Sciences, Beijing 100730, China.

Dent E, Martin FC, Bergman H, Woo J, Romero-Ortuno R, Walston JD. Management of frailty: opportunities, challenges, and future directions. Lancet. 2019;394(10206):1376–86.
Fang X, Shi J, Song X, et al. Frailty in relation to the risk of falls, fractures, and mortality in older Chinese adults: results from the Beijing Longitudinal Study of Aging. J Nutr Health Aging. 2012;16(10):903–7.
Ward DD, Ranson JM, Wallace LMK, Llewellyn DJ, Rockwood K. Frailty, lifestyle, genetics and dementia risk. J Neurol Neurosurg Psychiatry. 2022;93(4):343–50.
Ko D, Bostrom JA, Qazi S, Kramer DB, Kim DH, Orkaby AR. Frailty and cardiovascular mortality: a narrative review. Curr Cardiol Rep. 2023;25(4):249–59.
Jujo K, Kagiyama N, Saito K, et al. Impact of social frailty in hospitalized elderly patients with heart failure: a FRAGILE-HF registry subanalysis. J Am Heart Assoc. 2021;10(17):e019954.
Bunt S, Steverink N, Olthof J, van der Schans CP, Hobbelen JSM. Social frailty in older adults: a scoping review. Eur J Ageing. 2017;14(3):323–34.
Hamid TA, Salih SA, Zillah Abdullah SF, Ibrahim R, Mahmud A. Characterization of social frailty domains and related adverse health outcomes in the Asia-Pacific: a systematic literature review. Peer J. 2024;12:e17058.
Odaci Comertoglu E, Ozturk Y, Hafizoglu M, et al. The effect of social frailty on mental health and quality of life in older people: a cross-sectional study. Eur Geriatr Med. 2024;15(2):453–61.
Li X, Gao L, Qiu Y, et al. Social frailty as a predictor of adverse outcomes among older adults: a systematic review and meta-analysis. Aging Clin Exp Res. 2023;35(7):1417–28.
Bessa B, Ribeiro O, Coelho T. Assessing the social dimension of frailty in old age: A systematic review. Arch Gerontol Geriatr. 2018;78:101–13.
Andrew MK, Dupuis-Blanchard S, Maxwell C, et al. Social and societal implications of frailty, including impact on Canadian healthcare systems. J Frailty Aging. 2018;7(4):217–23.
Andrew MK, Keefe JM. Social vulnerability from a social ecology perspective: a cohort study of older adults from the National Population Health Survey of Canada. BMC Geriatr. 2014;14:90.
Makizako H, Shimada H, Doi T, et al. Social frailty leads to the development of physical frailty among physically non-frail adults: a four-year follow-up longitudinal cohort study. Int J Environ Res Public Health. 2018;15(3):490.
Hu X, Meng L, Wei Z, et al. Prevalence and potential risk factors of self-reported diabetes among elderly people in China: a national cross-sectional study of 224,142 adults. Front Public Health. 2022;10:1051445.
Zeng X, Jia N, Meng L, et al. A study on the prevalence and related factors of frailty and pre-frailty in the older population with diabetes in China: A national cross-sectional study. Front Public Health. 2022;10:996190.
Mitnitski AB, Mogilner AJ, Rockwood K. Accumulation of deficits as a proxy measure of aging[J]. Sci World J. 2001;1:323–36.
Searle SD, Mitnitski A, Gahbauer EA, Gill TM, Rockwood K. A standard procedure for creating a frailty index. BMC Geriatr. 2008;8:24.
Thompson MQ, Theou O, Adams RJ, Tucker GR, Visvanathan R. Frailty state transitions and associated factors in south Australian older adults. Geriatr Gerontol Int. 2018;18(11):1549–55.
Hoover M, Rotermann M, Sanmartin C, Bernier J. Validation of an index to estimate the prevalence of frailty among community-dwelling seniors. Health Rep. 2013;24(9):10–7.
Andrew MK, Mitnitski A, Rockwood K. Social vulnerability, frailty, and mortality in elderly people. PLoS ONE. 2008;3(5):e2232.
Liu XL, Wang XX, Li N, et al. Application of multiple filling method in public health investigation. Natl Med J China. 2019;99(46):3675–7. 10.3760/cma.j.issn.0376-2491.2019.46.018.
Moreira J, Silva B, Faria H, Santos R, Sousa ASP. Systematic review on the applicability of principal component analysis for the study of movement in the older adult population. Sens (Basel). 2022;23(1):205.
Shi J, Song X, Yu P, et al. Analysis of frailty and survival from late middle age in the Beijing Longitudinal Study of Aging. BMC Geriatr. 2011;11:17.
Shi J, Tao Y, Meng L, et al. Frailty status among the elderly of different genders and the death risk: a follow-up study. Front Med (Lausanne). 2021;8:715659.
Ruan H, Hu J, Zhao J, et al. Menopause and frailty: a scoping review. Menopause. 2020;27(10):1185–95.
Swiecicka A, Eendebak RJAH, Lunt M, et al. Reproductive hormone levels predict changes in frailty status in community-dwelling older men: European male ageing study prospective data. J Clin Endocrinol Metab. 2018;103(2):701–9.
Park H, Jang IY, Lee HY, Jung HW, Lee E, Kim DH. Screening value of social frailty and its association with physical frailty and disability in community-dwelling older Koreans: Aging Study of PyeongChang Rural Area. Int J Environ Res Public Health. 2019;16(16):2809.
Shah SJ, Oreper S, Jeon SY, Boscardin WJ, Fang MC, Covinsky KE. Social frailty index: development and validation of an index of social attributes predictive of mortality in older adults. Proc Natl Acad Sci U S A. 2023;120(7):e2209414120.
Ravenel JR, Perkins AE, Tomczik A, et al. Age-related decline in social interaction is associated with decreased c-Fos induction in select brain regions independent of oxytocin receptor expression profiles. Aging Brain. 2024;5:100107.
Zhang W, Feng Q, Liu L, Zhen Z. Social engagement and health: findings from the 2013 survey of the Shanghai elderlylife and opinion. Int J Aging Hum Dev. 2015;80(4):332–56.
Qi X, Li Y, Hu J, et al. Prevalence of social frailty and its associated factors in the older Chinese population: a national cross-sectional study. BMC Geriatr. 2023;23(1):532.
Ye L, Elstgeest LEM, Zhang X, Alhambra-Borrás T, Tan SS, Raat H. Factors associated with physical, psychological and social frailty among community-dwelling older persons in Europe: a cross-sectional study of Urban Health Centres Europe (UHCE). BMC Geriatr. 2021;21(1):422.
Luo D, Yu S, Wang J, et al. Social participation of community-dwelling older adults in western China: A latent profile analysis. Front Public Health. 2022;10:874204.
Chehrehgosha M, Sharifi A, Meftah AM, et al. Demographic and biological factors in interrelationships between physical, cognitive, psychological, and social frailty in community-dwelling older adults: Data from the Birjand Longitudinal Aging Study (BLAS). Maturitas. 2024;181:107905.
Sirven N, Dumontet M, Rapp T. The dynamics of frailty and change in socio-economic conditions: evidence for the 65 + in Europe. Eur J Public Health. 2020;30(4):715–9.
Ganesan B, Gowda T, Al-Jumaily A, Fong KNK, Meena SK, Tong RKY. Ambient assisted living technologies for older adults with cognitive and physical impairments: a review. Eur Rev Med Pharmacol Sci. 2019;23(23):10470–81.
Wang Y, Chen Z, Zhou C. Social engagement and physical frailty in later life: does marital status matter? BMC Geriatr. 2021;21(1):248.
Ye B, Chen H, Huang L, et al. Changes in frailty among community-dwelling Chinese older adults and its predictors: evidence from a two-year longitudinal study. BMC Geriatr. 2020;20(1):130.

No competing interests reported.

AppendixTable1.docx

Download PDF

Editorial decision: Revision requested
07 Aug, 2024
Editor assigned by journal
06 Aug, 2024
Submission checks completed at journal
06 Aug, 2024
First submitted to journal
03 Aug, 2024

You are reading this latest preprint version

Social vulnerability of different dimensions and physical frailty in older Chinese adults: a national cross-sectional study

Status:

Version 1

Abstract

Backgroud

Methods

Results

Conclusions

Figures

Background

Methods

Study Design and Participants

Survey Content

Frailty Assessment

Social Vulnerability Variables

Statistical Methods

Results

Comparison of General Characteristics and Social Frailty-Related Factors Among Elderly with Different Frailty States

Principal Component Analysis of Social Frailty Dimensions

Logistic Regression Analysis of the Impact of Social Frailty Dimensions on the Degree of Frailty

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1