The protective influence of family support on anxiety, depressive symptoms, and suicidal ideation among elderly Chinese nursing home residents: A study of serial mediation

Abstract

In light of the aging demographic in China, heightened attention is warranted for the mental well-being of elderly individuals. Nevertheless, the prevalence of suicidal ideation among older residents in Chinese nursing homes and the nuanced impact of family support on this phenomenon, mediated by anxiety and depressive symptoms, remain unclear. A cohort of 506 Chinese elderly adults participated in the study. Psychosocial traits were assessed using the Perceived Social Support from Family scale (PSS-Fa) for family support, the 7-item Generalized Anxiety Disorder scale (GAD-7) for anxiety symptoms, the 9-item Patient Health Questionnaire (PHQ-9) for depressive symptoms, and suicidal ideation. A structural equation model (SEM) was employed to execute a serial mediation model. The analysis of 506 elderly adults revealed that 8.1% reported varying levels of suicidal ideation within Chinese nursing homes. The pathway from family support to anxiety symptoms (standardized beta = −0.025, P = .241), family support to depressive symptoms (standardized beta = −0.072, P < .05), and family support to suicidal ideation (standardized beta = −0.082, P < .05) were explored. Additionally, pathways from anxiety symptoms to suicidal ideation (standardized beta = 0.364, P < .001), anxiety symptoms to depressive symptoms (standardized beta = 0.647, P < .001), and depressive symptoms to suicidal ideation (standardized beta = 0.369, P < .001) were examined. This study elucidated the underlying mechanisms connecting family support to suicidal ideation, with depressive symptoms partially mediating this association. Additionally, our discoveries shed light on the partial mediation of anxiety symptoms by depressive factors when it came to the realm of suicidal ideation.

1. Introduction

Suicide has become a significant public health and societal concern on a global scale. It not only leads to shattered personal lives and reduced productivity but also imposes substantial economic and healthcare burdens on society. In China, despite an overall decrease in suicide rates,^[1] rates among young males and elderly individuals have remained elevated.^[2] Several studies indicate that elderly adults face a higher risk of suicide.^[3,4] A recent meta-analysis has reported a prevalence rate of 11.5% for suicidal ideation among the elderly population in China. As China experiences population aging, with the number of individuals aged 60 and over reaching 260 million, constituting 18.7% of the total population according to the 2021 Census,^[5] it becomes imperative to prioritize the development and maintenance of psychological well-being among older adults. This approach is crucial for achieving high-quality aging, preventing adverse health outcomes, and relieving the burden on healthcare systems.

Suicidal ideation plays a crucial role in assessing suicidal behavior.^[6,7] Typically, it precedes suicide completion and is viewed as an integral aspect of the overall process.^[8] Furthermore, older adults living in nursing homes face a higher suicidal risk compared to those residing in the community, possibly due to factors like social isolation and cognitive decline.^[9] Timely prevention and detection of suicidal ideation are vital for the elderly population. However, current studies on older adults in China have predominantly focused on those living in the community, with limited attention directed toward residents of nursing homes.^[10]

Family support emerges as a pivotal protective factor against suicidal ideation among older adults. The perception of a supportive family environment not only serves as a shield against the psychological burden stemming from physical decline due to aging but also plays a crucial role in maintaining the overall well-being of older individuals. As individuals age, the family often evolves into the primary source of support, offering both practical assistance, such as meal preparation and medication administration, and invaluable emotional support.^[11]

Previous studies underscore the significant impact of emotional support from family on mental health. For instance, Fitzgerald analysis of MIDUS data has revealed an inverse association between family support and symptoms of depression and anxiety.^[12] Moreover, feeling valued within the family context has been linked to a reduced likelihood of expressing suicidal thoughts.^[13] Additionally, research indicates that active engagement in formal support activities may be correlated with a lower incidence of suicide attempts.^[14] By emphasizing the multifaceted contributions of family support, we gain a deeper understanding of its role as a protective buffer, not only addressing practical needs but also fostering emotional well-being in older adults.

Depressive and anxiety disorders emerge as poignant contributors to suicide attempts in older individuals. In a community-based cross-sectional study spanning 3 districts of Kashmir, Sheikh et al (year) have illuminated a significant correlation (R = 0.35) between depression and suicidal ideation.^[15] In a symphony of research, Artero and colleagues (2006) have orchestrated a poignant revelation, a staggering 71% of individuals who have attempted suicide have weathered at least one prior episode of major depression, a stark contrast to the 29% observed in non-attempters. Moreover, 26.1% have reported a history of generalized anxiety, a notable figure compared to the 10.2% prevalence in non-attempters, although anxiety without concurrent depression no longer stands as statistically significant.^[16]

Unveiling the intricate threads of this narrative, assessments utilizing symptom scales such as the Beck Depression Inventory Scale and Spielberger State Trait Anxiety Inventory have painted a nuanced canvas. The brushstrokes of these scales reveal intricate associations between depression, anxiety, and the delicate dance of suicide attempts.^[17] In the realm of mental well-being, these findings serve as illuminating beacons, guiding our understanding of the intricate interplay between emotional struggles and the profound vulnerability of our elders.

Previous research has established a connection between family support, anxiety symptoms, depressive symptoms, and suicidal ideation. Shao study reports a correlation between lower levels of family support and higher levels of anxiety and depression,^[18] while elevated levels of anxiety and depression symptoms can further escalate the risk of suicidal ideation.^[19] However, the underlying mechanisms that connect these factors remain unclear. Therefore, the objectives of this study are threefold: to investigate the prevalence of suicidal ideation among older Chinese nursing home residents; to explore the expected negative association between perceptions of emotional support from family and both depressive and social anxiety symptoms among older Chinese nursing home residents. Furthermore, a negative correlation is anticipated with suicidal ideation as well; to examine how anxiety symptoms and depressive symptoms may mediate the association between family support and suicidal ideation.

Our proposed model is Figure 1.

Figure 1.

The conceptual model for the whole sample based on previous studies.

2. Materials and methods

2.1. Participants

The research employed cluster random sampling techniques for a baseline evaluation of older adults residing in nursing homes, utilizing a cross-sectional study design. Four survey locations were randomly selected based on the economic classification of Nantong City, using a cluster random sampling technique. Between March 3 and March 25, 2023, participants were recruited from 4 nursing homes: Nantong Yincheng Health Care Hospital, Nantong Sunshine Nursing Home, Elderly Yiyuan, and Nantong Sunshine New City Rehabilitation Nursing Home, all situated in Nantong City, Jiangsu Province, China.

The survey was administered through the widely used online questionnaire platform, Wenjuanxing Software, facilitating participants in easily completing the survey. The cross-sectional survey formula was applied to calculate the sampling as follows:

$${\displaystyle \begin{array}{l}{\displaystyle \mathrm{N}=\frac{Z_{1-\mathrm{\partial }/2}^2\times pq}{d^2},}\end{array}}$$

Z_1-α/2 is the critical value for significance testing, where α = 0.05, and its value is 1.96. P represents the prevalence rate of outcome variables, where q = 1-p, and d denotes the allowable error, where d = 0.2p. Previous research indicates that the prevalence rate of outcome variables in older adults was 19.5%. For our calculations in this study, considering a 10% non-response rate, a minimum sample size of 436 subjects was deemed necessary.

To participate in our survey, individuals had to meet specific inclusion criteria: be aged ≥ 60 years old (in accordance with Chinese Elderly Standards); reside in one of the 4 nursing homes selected for the study; be free of any major mental illness; and express a willingness to volunteer for the survey and provide written informed consent. The study purpose was effectively communicated to each participant by our knowledgeable researcher, and the questionnaire link was distributed to ensure anonymity throughout the survey process. Participants were instructed to independently complete the questionnaire and were encouraged to reach out to our researchers for any clarifications. Moreover, participants were fully informed of their rights, including the option to withdraw from the survey at any point. The majority of participants spent approximately 15 minutes successfully completing the questionnaire. Assisted by nursing home staff, the survey included 602 elderly individuals, of which 506 questionnaires were completed, resulting in an impressive response rate of 84.1%. Ethical approval for this study was obtained from the Ethics Committee of Nantong First People Hospital under the identification number 2023KT091.

2.2. Measures

2.2.1. Basic socio-demographic variables.

The study comprehensively collected basic socio-demographic data, encompassing age, sex, ethnicity, marital status, place of residence, monthly income, and educational level.

2.2.2. Family support.

Family support was assessed using the Perceived Social Support from Family scale (PSS-Fa),^[20] consisting of 15 items answered with a simple “yes” (scored 1) or “no” (scored 0). The total score ranged from 0 to 15, with a higher overall score indicating greater family support. The internal consistency, as measured by Cronbach α coefficient, was 0.75 in this study.

2.2.3. Anxiety symptoms.

Anxiety symptoms were evaluated through The Generalized Anxiety Disorder 7-item scale (GAD-7), comprising 7 items with response options of “not at all” (scored 0), “several days” (scored 1), “more than half the days” (scored 2), and “nearly every day” (scored 3). The total scores ranged from 0 to 27, with higher scores reflecting more severe anxiety symptoms. The Cronbach α coefficient for this study was 0.94.

2.2.4. Depressive symptoms and suicidal ideation.

Depressive symptoms were assessed using the 9-item Patient Health Questionnaire (PHQ-9) developed by Kroenke et al (2001). The scale, with a Cronbach α coefficient of 0.87 in our study, excluded item 9 regarding suicidality.^[21] Suicidal ideation was separately examined using PHQ item 9 as described by Viner et al^[22] Participants with a score of 0 (“not at all”) were considered as having no suicidal ideation, while scores of 1 to 3 (“several days,” “more than half the days,” or “nearly every day”) indicated suicidal ideation, following the categorization method described by Uebelacker et al.^[23]

2.3. Statistical analysis

The statistical analysis encompassed descriptive analysis, utilizing frequency and percentages for categorical variables, and mean and standard deviation (SD) for continuous variables. Pearson correlation analysis was conducted to explore the associations between key variables, including family support, anxiety symptoms, depressive symptoms, and suicidal ideation. Additionally, a structural equation model (SEM) was executed to investigate the impact of family support on suicidal ideation through anxiety symptoms and depressive symptoms. The goodness-of-fit of the model was assessed using various indices, including Comparative Fit Index (CFI), Tucker Lewis Index (TLI), Normed Fit Index (NFI), Relative Fit Index (RFI), root mean square error of approximation (RMSEA), standardized root mean squared residual (SRMR), and CMIN/DF. A satisfactory goodness of fit was defined as 1 < CMIN/DF < 5, NFI > 0.90, CFI > 0.90, TLI > 0.90, RFI > 0.90, RMSEA < 0.08, and SRMR < 0.05, following the approach of Catherine et al^[24]

Confounding factors such as age, sex, and monthly income were controlled in the SEM analysis based on previous studies.^[25,26] Finally, hierarchical linear regression models were employed to assess the stability of our results. All statistical analyses were conducted using SPSS 21.0, with the SEM model performed using SPSS Amos software. The significance level was set at P < .05 (two-tailed) for all tests in this study.

3. Results

This study investigated a diverse sample of 506 elderly adults, comprising 313 females and 193 males, with an average age of 83.56 years (SD = 7.31) as detailed in Table 1. The entire cohort belonged to the Han ethnicity (100.0%), with 239 individuals (47.2%) indicating their marital status. Among the elderly participants, 388 (76.6%) were either married or divorced, while 106 (20.9%) reported a monthly disposable income ranging from 3001 to 6000 yuan. Notably, 41 (8.1%) of the elderly adults disclosed varying levels of suicidal ideation within Chinese nursing homes. The key variables exhibited mean scores of 13.49 (SD = 2.06) for family support, 1.00 (SD = 2.50) for anxiety symptoms, 2.72 (SD = 3.72) for depressive symptoms, and 0.09 (SD = 0.34) for suicidal ideation.

Table 1.

Socio-demographic characteristics and key variables outcomes of participants (N = 506).

Characteristic	Number	Percent (%)
Age (yr)	83.56 ± 7.31
60–69	21	4.2
70–79	99	19.6
80–89	273	54.0
90–100	94	18.6
Missing	19	3.8
Sex
Women	313	61.9
Men	193	38.1
Ethnic
Han	506	100.0
Marital status
Married	239	47.2
Divorce	149	29.4
others	8	5.2
Missing	110	21.7
Residence
Rural	51	10.1
Urban	454	89.7
Missing	1	0.2
Education level
High school or lower	372	73.5
Bachelor degree or above	130	25.7
Missing	4	0.8
Income (monthly)
500 or lower	36	7.1
501–1000	52	10.3
1001–3000	81	16.0
3001–6000	106	20.9
6001–9000	70	13.8
9001 or higher	74	14.6
Missing	87	17.2
Key variables (Mean ± SD)
Family support	13.49 ± 2.06
Anxiety symptoms	1.00 ± 2.50
Depressive symptoms	2.72 ± 3.72
Suicidal ideation	0.09 ± 0.34

Table 2 illustrates the results of Pearson correlation analysis, revealing a negative correlation between family support and both depressive symptoms and suicidal ideation. However, no significant correlation was found between family support and anxiety symptoms. Moreover, positive correlations were observed among anxiety symptoms, depressive symptoms, and suicidal ideation.

Table 2.

The correlation coefficient of key variables using Spearman correlation.

Variables	1	2	3	4
1. Family support	1
2. Anxiety symptoms	−0.574	1
3. Depressive symptoms	−0.088*	0.649***	1
4. Suicidal ideation	−0.128**	0.603***	0.613***	1

^*P < 0.05;

^**P < 0.01;

^***P < 0.001.

The findings from the SEM analysis are presented in Figure 2. The final model demonstrated strong fit indices: CMIN/DF = 2.276, NFI = 0.956, CFI = 0.974, TLI = 0.955, RFI = 0.923, RMSEA = 0.050, and SRMR = 0.042. After controlling for age, sex, and monthly disposable income, the pathway effects of key variables were as follows: family support to anxiety symptoms (standardized beta = −0.025, P = .241), family support to depressive symptoms (standardized beta = −0.072, P < .05), family support to suicidal ideation (standardized beta = −0.082, P < .05), anxiety symptoms to suicidal ideation (standardized beta = 0.364, P < .001), anxiety symptoms to depressive symptoms (standardized beta = 0.647, P < .001), depressive symptoms to suicidal ideation (standardized beta = 0.369, P < .001).

Figure 2.

Association between family support, anxiety symptoms, depressive symptoms, and suicidal ideation and its mediation role of anxiety symptoms and depressive symptoms.

Furthermore, bootstrap methods indicated that the indirect effect of family support on suicidal ideation via depressive symptoms was −0.027 (P < .05), and the indirect effect of anxiety symptoms on suicidal ideation via depressive symptoms was 0.239 (P < .05).

As presented in Table 3, the hierarchical linear regression models support and validate our observations from the SEM analysis. Even after adjusting for potential confounding factors such as age and sex, family support demonstrated no significant correlation with anxiety symptoms (b = 0.054, P = .580). However, it emerged as a protective factor against depressive symptoms (b = −0.089, P = .045) and suicidal ideation (b = −0.129, P = .004). When scrutinizing the linear regression outcomes for suicidal ideation, the influence of family support gradually diminished across Model 1, Model 3, and Model 4, indicating a partial mediating effect in the onset of depression. These results underscored the nuanced and intricate relationships between family support, mental health indicators, and the complex dynamics of suicidal ideation among elderly individuals in nursing homes.

Table 3.

The hierarchical linear regression model results for our key variables.

Variables	Anxiety symptoms		Depressive symptoms				Suicidal ideation
	Model 1		Model 1		Model2		Model 1		Model 2		Model 3		Model 4
	b	P	b	p	b	p	b	P	b	P	b	P	b	P
Age	−0.010	0.816	−0.008	0.855	−0.001	0.967	−0.030	0.504	−0.026	0.476	−0.023	0.508	−0.024	0.488
Sex (male)	−0.010	0.828	0.026	0.554	0.033	0.335	−0.015	0.742	−0.031	0.410	−0.009	0.805	−0.019	0.585
Family support	0.054	0.580	−0.089	0.045	−0.073	0.031	−0.127	0.004	−0.082	0.027	−0.113	0.002	−0.094	0.006
Anxiety symptoms					0.647	<0.001					0.600	<0.001	0.424	<0.001
Depressive symptoms									0.548	<0.001			0.280	<0.001

4. Discussion

This study shed light on the intricate pathway of family support influencing suicidal ideation among the elderly population, unveiling the pivotal roles of anxiety symptoms and depressive symptoms. Importantly, our findings underscored that, in the interplay between family support and suicidal ideation, depressive symptoms acted as a partial mediator, while anxiety symptoms did not significantly mediate the relationship. Moreover, our results suggested a reciprocal relationship, as depressive symptoms partially mediated the association between anxiety symptoms and suicidal ideation. Providing context to our discoveries, we found that 8.1% of elderly adults in Chinese nursing homes reported varying levels of suicidal ideation. Although this prevalence was concerning, it was notably lower than that reported in Zhang study, which has documented a prevalence rate of 19.5% among Chinese rural elderly residents in nursing homes.^[10] A plausible explanation for this difference could be attributed to variations in monthly disposable income between the 2 studies. Notably, our study revealed that only 17.4% of the sample reported a monthly disposable income of less than 1000 yuan, in contrast to Zhang study where 57.3% reported such income levels.^[10]

In alignment with Park findings, which associate higher income status with a reduced risk of suicidal ideation,^[27] our study suggested the potential for increased economic support for Chinese nursing homes as an intervention to mitigate the risk of suicidal ideation among elderly residents. This underscored the multifaceted nature of mental health in elderly populations and emphasized the importance of considering economic factors when crafting effective interventions for this vulnerable demographic. Our study illuminated the intricate dynamics among anxiety symptoms, depressive symptoms, family support, and suicidal ideation in elderly individuals residing in nursing homes. Notably, our findings highlighted the crucial partial mediating role of depressive symptoms in the association between anxiety symptoms and suicidal ideation. This aligned with prior research, emphasizing the co-occurrence of anxiety and depressive symptoms, with depressive symptoms often serving as a mediator in the link between anxiety and suicidal ideation.^[28–30] This underscores the necessity of a comprehensive assessment that considers both anxiety and depression symptoms when evaluating suicide risk in elderly individuals within nursing home settings.

Moreover, our study revealed a significant insight, the partial mediating role of depressive symptoms in the link between family support and suicidal ideation. Feelings of exploitation, rejection, and emotional distance from family members exert a substantial impact on depression and suicidal ideation.^[31,32] It is plausible that negative interactions, though perhaps less frequent, hold greater significance than positive ones, particularly in the domain of family relationships, and contribute significantly to these outcomes. Negative interactions have the potential to undermine one sense of personal control or self-worth, triggering adverse psychological characteristics such as depressive symptoms, which, in turn, can influence suicidal ideation.^[33] These insights underscore the intricate interplay between family dynamics, mental health, and vulnerability to suicidal ideation in elderly individuals, emphasizing the need for targeted interventions that address negative family interactions in the context of nursing home environments.

Furthermore, our findings indicated no significant correlation between family support and symptoms of anxiety, which deviated from the outcomes of previous studies.^[18] This inconsistency could be attributed to our distinctive sampling approach. Due to the influence of traditional Chinese culture, elderly individuals in China often strive to cultivate a tranquil state of mind as they age. Consequently, the impact of family support on their anxiety levels becomes less pronounced. Thus, their overall anxiety tends to remain unaffected by variations in family support.

Building on the Social-Emotional Selective Theory,^[34] our research highlighted that older adults often preferred relying on their existing family and friends rather than cultivating new relationships. Therefore, strategically siting nursing homes within communities becomes imperative to facilitate regular visits from family members to their elderly relatives. The environmental design of nursing homes should prioritize creating a warm and secure, home-like atmosphere. Achieving this design necessitates caregivers to demonstrate patience and consistently provide love, ensuring that every elderly resident feels genuinely cared for and valued, even when residing away from their original homes.

Moreover, nursing homes should establish and maintain close communication channels with the family members of elderly residents. This approach enables family members to comprehend and actively participate in the lives of their elderly relatives. The emotional connection within families plays a pivotal role in sustaining various psychological functions among elderly individuals, contributing significantly to their overall well-being.

This study comes with several limitations that should be considered. Firstly, the cross-sectional design impedes the ability to establish causality and extend the results to broader populations. Secondly, participants were recruited through convenient sampling from a single location in China, potentially limiting the generalizability of the findings to a nationally representative sample. Thirdly, certain crucial variables were overlooked as controlled factors in this study, such as physical health, the presence of diseases, medication intake, and cognitive impairment. Future research endeavors should strive to incorporate these previously omitted crucial variables, enhancing the comprehensiveness of our understanding of the subject matter. Addressing these factors in subsequent studies will contribute to a more thorough comprehension of the subject matter.^[35,36] Fourthly, due to the high age characteristics of our elderly care facility population, with a mean age of 83.56 (SD = 7.31) years, there may be a need to account for potential biases in testing certain variables, such as suicidal ideation. Lastly, despite efforts to minimize them, there may still be reporting and recall biases present in this study.

5. Conclusion

In this study, we unveiled the underlying mechanisms that connect family support to suicidal ideation, with depressive symptoms serving as a partial mediator in this association. Furthermore, our research indicated that anxiety symptoms were partially mediated by depressive symptoms concerning suicidal ideation. Notably, our findings highlighted that 8.9% of elderly adults living in Chinese nursing homes reported experiencing varying levels of suicidal ideation. These results suggested that strategies aimed at preventing suicidal ideation among elderly adults in nursing homes should extend beyond direct interventions targeting suicidal thoughts and behavior. Instead, efforts should also encompass addressing family support and promoting social connectedness to ameliorate depressive symptoms.

Acknowledgments

We extend our gratitude to all the nursing homes, trained medical staff, and participants who enthusiastically contributed to the acquisition of data.

Author contributions

Conceptualization: Hongjian Lu, Chunhua Wang.

Formal analysis: Huiying Miao, Xiangfan Chen.

Investigation: Chunhua Wang.

Methodology: Huiying Miao, Xiangfan Chen.

Software: Huiying Miao, Wenbin Ding, Xiangfan Chen.

Supervision: Hongjian Lu.

Validation: Huiying Miao, Xiangfan Chen.

Writing – original draft: Huiying Miao.

Writing – review & editing: Yimei Sun, Jianfeng Ji, Yongqi Lu, Yajing Meng, Wenbin Ding, Xiangfan Chen.