Life Satisfaction in a Clinical and a Nonclinical Group of Older People: The Role of Self-Compassion and Social Support

Abstract

Despite the health-promoting role of life satisfaction, little is known about its determining factors in older people with mental health problems compared to nonclinical participants. This study provides preliminary data into the role of social support, self-compassion, and meaning in life on older people’s life satisfaction within both clinical and non-clinical populations. In total, 153 older adults (age ≥60) completed the Satisfaction With Life Scale (SWLS), the Self-Compassion Scale (SCS), the Meaning in Life Questionnaire (MLQ), and questions for relational variables. Hierarchical logistic regression analysis revealed that determinants of life satisfaction were self-kindness (B = 2.036, p = .001) and intimate friends’ network (B = 2.725, p = .021), while family relationships were found to be significant among the clinical group (B = 4.556, p = .024). Findings are discussed in relation to incorporating self-kindness and rapport with family in clinical work with older adults to better promote their well-being.

Introduction

Given that older adults’ well-being is an indicator of the progress of society (Steptoe et al., 2015), a contemporary challenge is to facilitate the factors that relate to life satisfaction in late adulthood and incorporate them into clinical practice. Life satisfaction, defined as one’s evaluative global aspect of quality of life and happiness (E. D. Diener, 2009) stands as an important objective for health and social policies (World Health Organization [WHO], 2015). Yet, the factors contributing to life satisfaction of older adults with mental health needs, compared to those without any diagnosed mental health difficulties, have received little attention to date.

Life satisfaction is related to health maintenance and healthy aging (Celik et al., 2018; López Ulloa et al., 2013). Particularly, life satisfaction of individuals seeking mental health care services has been correlated with fewer psychiatric symptoms, decreased depression, and lower rates of comorbidity (Barnes et al., 2012; Koivumaa-Honkanen et al., 1996). The current body of research has documented that both personal psychological resources as well as interpersonal ones play a key role in life satisfaction. Studies conducted in Western and Eastern communities (Khodabakhsh, 2022; Rook & Charles, 2017; Wilson et al., 2007) have provided cross-cultural evidence that social support has a vital impact on older adults’ life satisfaction. In a longitudinal study (Harasemiw et al., 2019), less diverse social networks were related to lower life satisfaction and more depressive symptoms. On the other hand, higher levels of social support were found to buffer the impact of depressive symptoms on community-dwelling older adults’ life satisfaction (Adams et al., 2016). Similarly, poor social support was correlated with dissatisfaction with life for psychiatric inpatients (Koivumaa-Honkanen et al., 1996) and assisted-living residents (Young, 2006), while findings from outpatients showed that living in a nuclear family and having the perception of being ignored due to old age were inversely associated with life satisfaction and positively associated with higher depression (Ghimire et al., 2018). Furthermore, peer support has been underlined to be of importance for older adults in modern society (Aziz Marzuki et al., 2023). Although relational well-being may prevail as an integral component of older adults’ overall well-being, more empirical attention is warranted to the interplay between relational variables and dispositional ones.

A growing body of research has showed that self-compassion was related to higher levels of well-being and lower levels of depression and anxiety in later life (Allen et al., 2012; Brown et al., 2019; Homan, 2016; Phillips & Ferguson, 2013). Self-compassion refers to an accepting, kind, and forgiving attitude toward the self (K. Neff, 2003). In a scoping review (Tavares et al., 2020), self-compassion was found beneficial for older people’s psychological adjustment, as it has been found to protect them from developing mental health and sleep disturbance symptoms (Kim & Ko, 2018), and it exerted influence on the subjective well-being of older adults who use mental health services (Cunha et al., 2017). Despite the promising role of self-compassion in mental health (Muris & Petrocchi, 2017), its associations with meaning in life at older ages are yet to be clarified.

Meaningfulness is particularly pertinent to the promotion of healthy aging (Steptoe & Fancourt, 2019). Studies have highlighted that having lower levels of meaning in life increased the likelihood of suffering from depression in late adulthood (Volkert et al., 2019). Conversely, sustaining the presence of meaning despite the age-dependent changes was found to be an important inner health resource for older adults, as it was associated with stronger close relationships, broader social engagement, and less loneliness (Hupkens et al., 2018; Krause & Rainville, 2020). These findings indicate that personal and interpersonal factors may blend together to fill life with satisfaction, especially at older ages.

The aim of this study was to provide greater detail into the role of (1) self-compassion and meaning in life as modifiable dispositional variables, and (2) relationships with others (i.e., family members, friends, and the community), on older adults’ life satisfaction, while taking into account their mental health condition. Should clinicians be informed about the factors that are associated with life satisfaction of older adults with mental health issues, they could target tailor-made interventions to amplify resources in response to specific needs in the face of increased frailty and losses.

Methodology

Participants

In total, 153 older adults (age ≥60) participated in the study. The clinical sample (n = 54) was consecutively recruited from two psychogeriatric units of the public health system (i.e., 2nd Department of Psychiatry and the Patras Office of The Hellenic Red Cross), while the nonclinical sample (n = 99) was recruited from public community centers for older people, all located in an urban area in Greece. Participants in the clinical group were stabilized outpatients, currently receiving psychiatric medication for affective and/or anxiety disorders, diagnosed with a Mini-International Neuropsychiatric Interview (M.I.N.I.)-5.0.0-based (Sheehan et al., 1998) clinical interview according to the DSM-IV-TR criteria (American Psychiatric Association [APA], 2000). Their symptomatic distress levels were not of a clinical magnitude according to the SCL-90-R scores (Derogatis, 1992). Patients were excluded if other conditions, such as neurocognitive disorders, organic mental disorders, mental retardation, catatonic or psychotic features, and severe personality disorders, were present.

Table 1 presents the demographic characteristics of the two samples (clinical and nonclinical). The two groups were not significantly different, with the exception of female gender (p = .032) and less involvement in a community center for older people (p = .015) in the clinical group compared to the nonclinical one.

Table 1.

Sociodemographic Characteristics Across Samples.

Characteristics	Clinical group		Nonclinical group		Total		p-Value*
	n	%	n	%	n	%
Gender (female)	40	74.1	56	56.6	96	62.7	.032
Age (60–69)	32	59.3	61	61.6	93	60.8	.775
Retired	36	67.9	74	74.7	110	72.4	.370
Married	27	50.0	62	62.6	89	58.2	.130
Children	47	88.7	93	93.9	140	92.1	.252
Grandchildren	36	67.9	65	66.3	101	66.9	.842
Non-member of senior center	43	82.7	63	63.6	106	70.2	.015
Physical health problems	45	84.9	75	75.8	120	78.9	.187

Note. N = 153 (n = 54 for the clinical sample and n = 99 for the nonclinical sample).

^*p-Value from Pearson’s chi-square test.

Materials

Satisfaction With Life Scale

The SWLS (E. Diener et al., 1985) contains five items rated on a 7-point Likert scale. Total scores range from 5 to 35, with higher scores indicating greater life satisfaction. The SWLS has been used in both clinical and nonclinical samples (Pavot & Diener, 2008) as well as in older adults (Pavot et al., 1991), and it has been validated in the Greek population (Lyrakos et al., 2013) . Cronbach’s alphas range from .79 to .89 (Pavot & Diener, 1993). In the present study, Cronbach’s alpha was .88.

Self-Compassion Scale (SCS)

The SCS contains 26 items rated on a 5-point Likert scale to indicate the degree of a compassionate attitude toward the self during hardship (K. Neff, 2003). It consists of six subscales: self-kindness, common humanity, mindfulness, self-judgment, isolation, and over-identification. Cronbach’s alphas have been found to be .92 for the total score and from .75 to .81 for the subscales (Allen et al., 2012; K. Neff, 2003; K. D. Neff et al., 2017). The reliability and validity of the SCS were confirmed in the Greek context (Mantzios et al., 2015) . In this study, the internal consistency was 0.79, while Cronbach’s alphas for the six subscales ranged from .57 to .72 with over-identification having poor reliability (α = .57).

Meaning in life

The 10-item MLQ comprises Presence (perceived meaning) and Search (motivation to discover meaning) scales, rated on a 7-point Likert scale. Factor scores range from 5 to 35. MLQ demonstrated adequate internal consistency with Cronbach’s alphas of .86 and .87 for the Presence and the Search subscales, respectively (Steger et al., 2006). It has been valid in older-adulthood (Hallford et al., 2018) and clinical populations (Schulenberg et al., 2011). The Greek version showed adequate internal consistency estimates (Stalikas et al., 2018) . Cronbach’s alphas in this study were .87 for the total scale.

Relational variables

Seven closed-ended questions were asked, in order to capture both objective and subjective dimensions of social support (Oxman et al., 1992). Two questions on the frequency of time spent with children and grandchildren were rated on a 7-point Likert scale ranging from 0 to 6 (“Never,” “Several times in a year or less,” “Once a month or less,” “Several times in a month,” “Once a week,” “Several times in a week,” and “Everyday”). The number of older adults’ close friends (1 = “None” to 5 = “More than 5”), self-rated social participation in community activities, and the degree of satisfaction with their relationship with (i) close family members, (ii) other relatives, and (iii) social network, using a 5-point Likert scale ranging from 1 (“Not at all satisfied”) to 5 (“Very much satisfied”), were also included.

Procedure

The study protocol was approved by the Research Ethics Committee of the University Hospital (EBD 58/4-2-2020). Participation in the study was on a voluntary basis. The data were anonymized and a written informed consent was provided by all participants. The measures were administered individually in paper-and-pencil form. In case of difficulty, trained research assistants helped the participants to fill them out. No missing data were detected in the questionnaire upon delivery.

Statistical Analyses

Descriptive statistics as well as means and standard deviations of all scales and subscales were examined. Pearson’s Chi square test and t-test for independent samples were used to examine distribution differences of demographic variables and to compare mean scale scores between their categories as well as mean differences in the scale scores and the relational characteristics between clinical/nonclinical groups, respectively. Associations of scale scores with demographic and relational variables were also examined. The normality of distributions was explored within each group via Q-Q plots, and was found to be close to normal for scale scores. Relational variables were treated as ordinal; thus, Spearman’s correlation was used for their correlation with scale scores and Pearson correlation analysis was performed to investigate the correlation between scale scores. A hierarchical linear model was conducted with SWLS as dependent variable, including the clinical/nonclinical variable in the first step, while adding in the second step the relational factors, and in the third step the scale scores that had been found to be significantly associated with SWLS in the two-way full factorial ANOVA models. All VIF values were between 1 and 3, so no multicollinearity was detected. Post hoc power calculations given the achieved sample size using the G*Power program (version 3.1.9.4) (Faul et al., 2007) showed power of 90% in independent samples t-tests for medium effect sizes (Cohen’s d = 0.5) and 32% for small effect sizes (d = 0.2). Tests for adding a predictor in multiple regressions with five predictors had power of over 99% for medium effect sizes (f2 = 0.15) and 41% for small effect sizes (f2 = 0.02). Data were analyzed using SPSS, Version 25 (IBM Corp. Released, 2017).

Results

Descriptive Statistics

Participants rated their relationship with close family members and relatives as very (4.0 ± 1.1) and moderately (3.6 ± 1.1) satisfying, respectively. However, the clinical group had significantly lower means compared to the nonclinical group in most of the relational variables (Table 2).

Table 2.

Descriptive Statistics for Relational Variables Across Samples.

Variable	Clinical group	Nonclinical group	Total	p-Value*
	M ± SD	M ± SD	M ± SD
Time spent with children	4.0 ± 2.3	5.0 ± 1.6	4.7 ± 2.0	.006
Time spent with grandchildren	2.94 ± 2.9	3.3 ± 2.6	3.2 ± 2.6	.393
Satisfaction with relationship with close family members	3.6 ± 1.3	4.2 ± 0.9	4.0 ± 1.1	.004
Satisfaction with relationship with relatives	3.3 ± 1.2	3.7 ± 1.1	3.6 ± 1.1	.012
Number of intimate friends	3.0 ± 1.2	3.4 ± 1.2	3.3 ± 1.2	.028
Satisfaction with relationship with friends and social network	3.0 ± 1.2	3.4 ± 1.2	3.3 ± 1.2	.230
Degree of social participation in community activities	1.9 ± 1.2	2.1 ± 1.3	2.0 ± 1.3	.462

^*p-Value from t-test for independent samples.

Table 3 presents the means and standard deviations for the study variables. The mean scores of SWLS, SCS, and MLQ scales and subscales for the total sample were moderate to high. The nonclinical group scored higher than the clinical one in satisfaction with life (t(150) = 5.59, p < .001), presence of meaning in life (t(76.8) = 2.41, p = .018), search for meaning in life (t(148) = 2.16, p = .033), and the total score of meaning in life (t(76.9) = 2.71, p = .008). Significant differences on self-compassion between the two groups were found for self-kindness (t(147) = 2.66, p = .009), common humanity (t(148) = 2.95, p = .004), and mindfulness (t(148) = 2.01, p = .046), with the clinical group presenting lower scores than the nonclinical one.

Table 3.

Means, Standard Deviations, and Differences of Scale and Subscales Scores Across Samples.

Variable	Clinical group	Nonclinical group	Total	p-Value*
	M ± SD	M ± SD	M ± SD
Satisfaction with life	18.1 ± 7.7	24.8 ± 6.8	22.5 ± 7.8	<.001
Meaning in life	42.3 ± 14.9	48.6 ± 10.2	46.0 ± 12.3	.008
Presence	22.4 ± 7.1	25.1 ± 5.1	24.1 ± 5.9	.018
Search	20.8 ± 8.5	23.5 ± 6.7	22.6 ± 7.4	.033
Self-compassion	2.9 ± 0.5	3.0 ± 0.6	3.0 ± 0.6	.037
Self-kindness	2.9 ± 1.0	3.4 ± 0.9	3.2 ± 0.9	.009
Self-judgment	3.2 ± 0.9	3.1 ± 0.8	3.1 ± 0.9	.281
Common humanity	2.9 ± 1.0	3.4 ± 0.8	3.2 ± 0.9	.004
Isolation	3.2 ± 0.9	3.4 ± 1.0	3.3 ± 1.0	.360
Mindfulness	3.0 ± 1.0	3.4 ± 0.9	3.3 ± 1.0	.046
Over-identification	3.2 ± 0.9	3.1 ± 0.9	3.1 ± 0.9	.541

^*p-Value from t-test for independent samples.

Associations of Satisfaction With Life With Self-compassion and Meaning in Life

In the total sample, SWLS score was correlated with older adults’ satisfaction with their relationship with close family (Spearman’s rho = .241, p = .003) and other relatives (Spearman’s rho = .186, p = .022) as well as with the number of close friends (Spearman’s rho = .253, p = .002). SWLS score was also significantly correlated with MIL total score (r = .207, p = .011) and presence of meaning in life (r = .215, p = .008), but not with search for meaning in life (r = .120, p = .142). Furthermore, SWLS was correlated with self-kindness (r = .363, p < .001), common humanity (r = .175, p = .032), and mindfulness (r = .167, p = .042). Interestingly, no significant association was found between SWLS and self-rated physical health problems (t(65.4) = −0.86, p = .392).

Both the MIL total and the MIL-Presence scores were significantly correlated with self-kindness (r = .308, p < .001 and r = .346, p < .001), common humanity (r = .189, p = .020 and r = .274, p = .001), and mindfulness (r = .172, p = .036 and r = .211, p = .010), whereas the MIL-Search subscale was significantly correlated only with self-kindness (r = .213, p = .009). Self-kindness was also correlated with peer support satisfaction (Spearman’s rho = .256, p = .002). Common humanity was significantly correlated with the frequency of spending time with grandchildren (Spearman’s rho = −.186, p = .029), and satisfaction with relationship with relatives (Spearman’s rho = −.163, p = .047). On the other hand, the isolation score was found to be negatively correlated with the frequency of spending time with grandchildren (Spearman’s rho = .170, p = .048), but positively associated with physical problems (t(146) = −2.19, p = .030) with a higher mean score (3.63 ± 0.85 vs. 3.22 ± 0.96) in those having physical problems compared to those without any such problems.

Satisfaction With Life Among Clinical and Nonclinical Groups

The linear hierarchical regression model revealed that clinical versus nonclinical group was significantly associated with SWLS, in all the three steps of the model (Table 4). The relationship with family was found to be significantly associated with SWLS among the clinical group (B = 4.556, p = .024). However, this association marginally lost its significance after the addition of self-kindness in the last step of the model and its significant association with SWLS (B = 2.036, p = .001). Having four or more intimate friends was found to be significantly associated with SWLS (B = 2.725, p = .021) in both the second and the third step of the model. No significant impact of the variable of gender was found.

Table 4.

Hierarchical Linear Regression Analysis for Satisfaction With Life of Older Adults.

Model	Unstandardized coefficients		Standardized coefficients	t	Sig.
	B	SE	β
1. (Constant)	31.431	1.746		18.002	<.001
Clinical vs. Nonclinical	−6.646	1.230	−0.408	−5.401	<.001
	R2 = .167; adjusted R2 = .161
	ΔR2 = .167; F(1.146) = 29.17; p < .001
2. (Constant)	32.651	3.277		9.965	<.001
Clinical vs. Nonclinical	−9.091	2.127	−0.558	−4.274	<.001
relationshipfamily_GOOD among clinical	4.556	2.003	0.241	2.274	.024
relationshipfamily_GOOD among nonclinical	0.069	1.657	0.004	0.042	.967
Four or more close friends	3.026	1.207	0.188	2.508	.013
	R2 = .233; adjusted R2 = .212
	ΔR2 = .067; F(3.143) = 4.16; p = .007
3. (Constant)	24.299	3.995		6.083	<.001
Clinical vs. Nonclinical	−7.544	2.101	−0.463	−3.591	<.001
relationshipfamily_GOOD among clinical	3.769	1.946	0.199	1.937	.055
relationshipfamily_GOOD among nonclinical	0.287	1.600	0.019	0.180	.858
Four or more close friends	2.725	1.167	0.169	2.334	.021
Self-compassion_Self-kindness	2.036	0.596	0.251	3.419	.001
	R2 = .292; adjusted R2 = .267
	ΔR2 = .058; F(1.142) = 11.69; p = .001

Discussion

The purpose of this study was to examine and compare social support, self-compassion and meaning in life in relation to life satisfaction between a clinical and a nonclinical group of older people. The main findings of this study are: (1) life satisfaction, meaning in life, and self-compassion in all the positive factors of self-kindness, common humanity and mindfulness were significantly lower in older adults with mental health issues compared to those in the nonclinical group; (2) having an emotionally satisfying relationship with family members was a main determinant of life satisfaction for older adults in the clinical group compared to the nonclinical one; (3) having self-kindness and a peer support network prevailed as major factors of older adults’ life satisfaction in both groups. These findings are in line with previous research showing a negative association of older persons’ life satisfaction with psychiatric morbidity (Barnes et al., 2012; Ghimire et al., 2018; Koivumaa-Honkanen et al., 1996). Additionally, it demonstrated the effect of mental health difficulties on older people’s subjective wellbeing, as this was found to be independent from the sociodemographic variables, or the self-rated physical health problems.

The current study builds on the existing literature by presenting new perspectives on the link between non-institutionalized older adults’ life satisfaction with family relationships, empirically supporting the proposal that the treatment of older adults may also entail strengthening the quality of rapport with family members. Furthermore, the results are in line with research focusing on the role of increasing satisfaction with peer and social support as a moderator of treatment effects on older adults’ quality of life (Aziz Marzuki et al., 2023; LaRocca & Scogin, 2015) as well as new evidence about how the integration of positive psychology in the clinical milieu (Yotsidi, 2020), especially self-compassion interventions in older adults (Hodgetts et al., 2021; Perez-Blasco et al., 2016; Stroud & Griffiths, 2021) may improve subjective wellbeing.

What is perhaps striking in our study, is that all the positive factors of self-compassion (i.e., self-kindness, common humanity, and mindfulness) were found to be at higher levels among the older adults who did not use mental health services. Additionally, our finding that self-kindness was associated with older persons’ life satisfaction and meaning in life both within clinical and nonclinical populations extends emerging research on assessing self-kindness as a unique well-being and mental health protective factor in late adulthood (Ceclan & Nechita, 2021; Dreisoerner et al., 2020; Smith et al., 2018). Based on evidence gleaned so far, treating oneself with a compassionate attitude when faced with personal suffering can serve as a buffer against psychopathology in later life (Allen et al., 2012; Brown et al., 2019; Homan, 2018; Tavares et al., 2020). Yet, to our best knowledge this is the first study to assess the role of self-compassion among different groups of older adults (clinical vs. nonclinical) in relation to their life satisfaction and other interpersonal and intrapersonal variables.

Previous research showed that the negative indicators of self-compassion increase vulnerability to mental health problems, while the positive ones have a protective influence against psychopathology (Muris & Petrocchi, 2017). In our study, the positive indicators of self-compassion were associated with different relational variables, such as older people’s relationships with their grandchildren, relatives, or friends. On the other hand, isolation was found to be associated with perceived physical problems. These findings suggest the need to explore further the interplay between older persons’ self-compassion, mental and physical health indicators, and different patterns of interaction with others. Taking into account that older adults with higher self-kindness and common humanity were found to be more satisfied with their social relationships and spend more time with their grandchildren, respectively, further research is warranted into the beneficial role of self-compassion for people with poorer mental health to help them sustain social and family connectedness.

This is of importance, considering that despite older adults in the clinical group rely more on their perceived satisfaction with relating to intimate others to be satisfied with their life, they reported spending significantly less time with their children, and being less satisfied with family, or having friends. These results confirm recent evidence (Lee et al., 2020) that it is the appraisal of the quality of family interactions that matters more to older adults with mental health issues than quantitative criteria (e.g., frequency of contacts, amount of social participation). Yet, the increase of the population of older adults often co-occurs with the decrease of the offspring’s availability in modern societies due to geographical distance and work demands, leading to a “double-bind” phenomenon that may intensify frustration in older adults’ intimate relationships. Thus, grappling with the developmental tasks of older age in clinical populations may require a “bidirectional intervention,” that is, targeting both the older clients and their family members.

To the best of our knowledge, this study is the first to directly compare the role of social support, self-compassion, and meaning in life in older adults’ life satisfaction among clinical versus nonclinical samples. Overall, our findings highlight the lack of perceived life satisfaction in older people with mental health needs and they encourage future research on how to incorporate subjective well-being as an additional treatment goal. Yet, there are several notes of caution in interpreting the findings. First, no causal inferences can be drawn due to the cross-sectional nature of our study. In addition, we interviewed the higher functioning older adults, who in the case of the clinical group received medical treatment, thus being in remission of symptoms. Taking into account that it is challenging to collect data from a clinical group, a larger sample of older adults with mental health issues is necessary in future research. Furthermore, the study was based on self-reported data. Finally, the results highlighted differences between a clinical and a nonclinical group of older adults in a Greek cultural context, thus cultural issues may be relevant to their interpretation.

Conclusion

A main determinant of life satisfaction for older people with clinically diagnosed mental health difficulties was their perceived satisfaction with their relationship with family members. Nevertheless, it was self-kindness, as a personal resource, and an intimate friends’ network, as an interpersonal one, that were significant for both groups of older people (clinical and nonclinical), inviting thus further scientific inquiry into the beneficial role of self-compassion, and particularly self-kindness, in life satisfaction at late adulthood. These preliminary findings suggest the need to incorporate both intrapersonal and relational well-being indicators into the provision of mental health services, as a means to promote older adults’ life satisfaction and expand the scope of clinical work to become more comprehensive for older people.