jfa journal

AND option

OR option

LONELINESS PREDICTS PROGRESSION OF FRAILTY IN MARRIED AND WIDOWED, BUT NOT UNMARRIED COMMUNITY DWELLING OLDER ADULTS

 

C. Pollak1, J. Verghese1,2, A.S. Buchman3, Y. Jin4, H.M. Blumen1,2

 

1. Department of Medicine, Albert Einstein College of Medicine, Bronx, NY, USA; 2. Department of Neurology, Albert Einstein College of Medicine, Bronx, NY, USA; 3. Department of Neurological Sciences, Rush University Medical Center, Chicago, IL, USA; 4. Department of Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, NY, USA

Corresponding Author: Chava Pollak, PhD, Albert Einstein College of Medicine, Department of medecine, 1225 Morris Park Ave, Van Etten 308a, Bronx, NY 10461. Email: chava.pollak@einsteinmed.edu

J Frailty Aging 2024;13(2)163-171
Published online March 19, 2024, http://dx.doi.org/10.14283/jfa.2024.27

 


Abstract

BACKGROUND: Loneliness is highly prevalent among older adults and is associated with frailty. Most studies consider loneliness in isolation without consideration for structural and functional measures of social relationships – and longitudinal studies are scarce.
OBJECTIVES: This study examined longitudinal associations between loneliness and frailty and analyzed how structural and functional social measures influence these associations.
DESIGN: Linear mixed effects models examined longitudinal associations between loneliness and frailty assessed with the frailty index (scale 0-100). Models were adjusted for baseline age, gender, education, depressive symptoms, global cognition, and structural (e.g., social network, marital status), and functional social measures (e.g., social, cognitive, and physical activity, and social support).
PARTICIPANTS: Loneliness and frailty data from 1,931 older adults without dementia at baseline from the Rush Memory and Aging Project were examined (mean age 79.6 ± 7.7 years, 74.9% female).
MEASUREMENTS: Baseline loneliness assessed by the de Jong Gierveld Loneliness Scale was the predictor of interest.
RESULTS: Frailty increased significantly over a mean follow-up period of 4.6 years. Effects of loneliness on frailty were modified by marital status. Loneliness predicted an additional accumulation of 0.37 and 0.34 deficits on the frailty index per year in married and widowed individuals respectively, compared to those who were not lonely (married: p=0.009, CI 0.09, 0.64; widowed: p=0.005, CI 0.1, 0.58). Loneliness did not predict frailty progression in unmarried individuals.
CONCLUSIONS: Loneliness predicts frailty progression, highlighting the importance of social determinants on physical health in aging.

Key words: Loneliness, frailty, psychosocial.


 

Introduction

Loneliness is Associated with Poor Health Outcomes

Loneliness is defined as a negative feeling of dissatisfaction with the quality or quantity of social relationships, or the perception of being socially isolated. Loneliness is associated with cognitive and functional decline and an increased risk of Alzheimer’s disease type dementia and mortality (1-4). The pooled prevalence of loneliness among adults aged 65 and older from 15 countries across 4 continents was 28.6%, and the prevalence of loneliness has increased since the onset of the COVID-19 pandemic (5). The prevalence of loneliness was estimated to be 43.2% in a sample of over twenty thousand US middle-aged and older adults (6). Loneliness is increasingly recognized as a public health epidemic and the US Surgeon General recently called for the prioritization of research investigating this issue commensurate with the significant morbidity and mortality associated with loneliness (7), including depression (8), cognitive and functional decline (2, 3), dementia (4), and mortality (1). While the associations between loneliness and functional decline in aging are well-established (3, 9, 10) the longitudinal relationship between loneliness and frailty is less clear.

Current Evidence on Loneliness and Frailty

Frailty is a multi-dimensional construct characterized by decreased physiologic reserves that results in an increased vulnerability to a range of adverse health outcomes including institutionalization, worsening health status, predementia syndromes, dementia, and mortality (11-13). Two of the most commonly used definitions for frailty are the Fried phenotypic approach (14) and the Rockwood deficit accumulation approach (15). The phenotypic approach operationalizes frailty as a biologic syndrome defined by unintentional weight loss, low grip strength, exhaustion, decreased physical activity, and slow gait speed and is associated with falls, disability, hospitalization and mortality (14). The deficit accumulation approach defines frailty as a multidimensional risk state characterized by the quantity of accumulated deficits that reduces the ability of the system to respond to stressors (15). Data on the longitudinal relationship between loneliness and frailty is scarce – and extant studies show mixed results; some showed loneliness was longitudinally associated with frailty and some showed no association (16-19). Studies showed loneliness predicted lower likelihood of transitioning between frailty states (19-21). Studies commonly defined frailty based on the frailty phenotype, which may underestimate frailty levels due to its categorical approach and focus on only one dimension of frailty (22). Only two longitudinal studies reported associations of loneliness and frailty in community dwelling older adults in the same sample using the frailty index, rather than the frailty phenotype (16, 17). One study showed loneliness predicted frailty based on the frailty phenotype definition but not the frailty index in a sample of 2,817 community dwelling older adults with a mean age of 69.3 years (SD 6.9) over 6 years of follow up. The study was replicated in a larger sample of 9,171 individuals from the same cohort with a mean age of 66.3 years (SD 10.2) using the frailty index and showed loneliness predicted frailty over 13 years of follow up 17 however, in this study the frailty index was categorized into 3 categories (frail, pre-frail, robust) rather than a continuous scale. Finally, the risk for mortality associated with frailty was compounded in individuals with frailty combined with loneliness and social isolation, compared to individuals without loneliness or social isolation (23).

Interrelationships between Multiple Aspects of Social Relationships

Three broad aspects of social relationships have been used to categorize how people relate to each other – structural (e.g., social network, marital status), functional (e.g., social support and engagement), and quality (e.g., loneliness, relationship quality) (24). Studies of loneliness and frailty tend not to consider other aspects of social connection. Loneliness measures the perception of social isolation, but an individual who is objectively alone (e.g., living alone or small social network) may not feel lonely, and an individual who is surrounded by others may feel alone (24). Considering multiple aspects of social relationships in an analysis of loneliness and health outcomes is clinically advantageous because it offers a more comprehensive consideration of risk compared to analyzing each construct individually. The inclusion of multiple aspects of social relationships is also a best-practice recommendation from the US Surgeon General’s recent advisory on loneliness and social isolation (7).
The purpose of this study was to fill these gaps in the literature in the following ways: 1) examining associations between loneliness and frailty using the frailty index as a continuous measure to avoid underestimation and capture the multidimensionality of frailty, 2) adding to the literature on loneliness and frailty in a large, longitudinal sample of community dwelling older adults, given prior mixed results, and 3) assessing whether structural and functional social measures modify the relationship between frailty and loneliness – a quality social measure. We hypothesized that loneliness would predict frailty progression over time and that structural and functional support measures would not modify the relationship between loneliness and frailty because it is possible that one might be surrounded by others and feel alone while one might be alone and not feel lonely (24).

 

Methods

Participants

This study includes data from the Rush Memory and Aging Project (MAP). The MAP study is a longitudinal, clinical-pathologic study of chronic conditions of aging. Study procedures are described in detail elsewhere (25). Briefly, participants were recruited from retirement and subsidized housing facilities, church groups, and social services agencies around northeastern Illinois. Participants agreed to detailed clinical evaluation annually conducted in participants’ homes and anatomical gift donation at the time of death (25). There were 2,252 participants in the original dataset that included participants from study inception until the time the data was received (study years 1-20). 237 participants were excluded for missing baseline loneliness measures. An additional 84 participants who had dementia at baseline were excluded since dementia is associated with both loneliness and frailty and can exert influence on both 4,13, and reliability of responses on self-report items such as a loneliness scale or self-reported frailty index items may differ in individuals with dementia. This left 1,931 participants eligible for this investigation with a median follow up of 4 years (IQR 1, 7). The MAP study was approved by the Rush Medical Center Institutional Review Board. Ethical approval for secondary analyses was obtained from the Albert Einstein College of Medicine Institutional Review Board.

Frailty Index

For the deficit accumulation approach, deficits are quantified in a Frailty Index (FI). The index uses a range of deficits like those included in a standard Comprehensive Geriatric Assessment (CGA). Data can be self-reported (e.g., self-rated health), assessment data (e.g., Mini Mental Status Exam), clinical exam (e.g., blood pressure), or laboratory assays (e.g., abnormal creatinine) (15). The more deficits a person has, the higher likelihood they will be frail (15). Candidate variables for the frailty index were screened for inclusion based on previously established procedures using the following criteria: 1) association with health status, 2) increasing prevalence with age, 3) does not saturate too early (e.g., presbyopia), and 4) represents multiple organ systems (11). We included a combination of variables across multiple domains of health and function that were obtained at each clinical evaluation (e.g., physical performance tests, cognitive impairments, comorbidities, depression/mood, and functional measures). The final frailty index included 39 items, displayed in Table 1. Frailty index scores were calculated by adding up the number of deficits present and dividing that number by the total number of deficits included in the index resulting in a range of scores from 0 to 1 with higher scores representing greater frailty. A frailty index measure was calculated for all participants at each time point and none were missing. Scores were multiplied by 100 for ease of interpretation. We tested the predictive validity of the frailty index by assessing the ability of the frailty index to predict mortality using Cox proportional hazards models (Appendix B and Figure e1). A similar frailty index was previously constructed in the same cohort (26). The frailty index was considered as a continuous variable as this allows for examination of a spectrum of health domains and their association with loneliness in the most direct way with the most statistical power to precisely identify individuals with frailty. Genetic studies identified several genetic loci associated with the FI and traits and diseases that are associated with frailty (e.g., cardiovascular disease, depression, neuroticism, body mass index, smoking), which endorses the multidimensionality of the accumulation of deficits approach (27). Additionally, proteomic markers associated with the FI point to the biologic basis for this approach (28). Frailty defined as a proportion of deficits is a robust characteristic and, when sufficiently large numbers of variables are included in the index, the variables yield comparable results in terms of association with adverse outcomes (15).

Table 1. Health Variables Included in the Frailty Index

 

Loneliness

Loneliness was assessed using a modified 5-item de Jong Gierveld Loneliness Scale. The de Jong Gierveld Loneliness Scale is a valid and reliable tool that captures the social and emotional aspects of loneliness 29. In this study, loneliness was characterized by the following: 1) I experience a general sense of emptiness, 2) I miss having people around, 3) I feel like I don’t have enough friends, 4) I often feel abandoned, and 5) I miss having a really good friend. Participants were asked to rate agreement with each item on a 5-point Likert scale and scores were summed and averaged. Higher values indicated higher levels of loneliness. Loneliness was considered as a continuous variable in these analyses as there is no established cut-point for loneliness. Additionally, using a continuous loneliness variable would capture the most complete information as higher scores indicate higher levels of loneliness rather than dichotomizing individuals into two groups.

Other Covariates

Covariates were selected based on reported associations with loneliness and/or frailty (4, 8, 14, 30-35). Gender and education (in years) were recorded at the time of enrollment. Age in years was computed from self-reported date of birth and date of clinical evaluation. Global cognitive function assessed the following cognitive domains: episodic memory, semantic memory, working memory, visuospatial ability/perceptual orientation, and perceptual speed, and was calculated by converting raw scores from a battery of cognitive tests into z-scores and averaged to yield a global cognitive function summary score. Depression was assessed with a modified 10-item Center for Epidemiologic Studies Depression Scale (CES-D) (36). An overall depression score was computed as the sum of symptoms experienced with higher scores indicating higher levels of depressive symptoms. For these analyses, the depression score was re-calculated using only four items from the original 10-item score to exclude a loneliness item, and five items that were included in the frailty index. Marital status was assessed at baseline with questions inquiring if the participant was ever married and if so, their current marital status. Responses were registered as never married, married, widowed, divorced, or separated. For the analyses, we categorized marital status as married, widowed, and unmarried as approximately 80% of participants were married or widowed. We quantified social network size as the number of community members, family, and friends seen at least once a month. Perceived social support was assessed with 4 items from the Multidimensional Scale of Perceived Social Support (37). Items were summed with higher values indicating more social support. Additionally, we included three measures of social engagement – physical, cognitive, and social activity. Physical activity was assessed using questions adapted from the 1985 National Health Interview Survey and measured the number of hours per week the participant engaged in 5 activity categories (e.g., walking, gardening/yardwork, exercise, bicycle riding, swimming). Cognitive activity was assessed by frequency of participation in 7 cognitively stimulating activities during the past year (e.g., reading, playing chess) and summed into a composite score. Social activity was assessed using a 6-item scale that inquired how often the participant engaged in common social activities over the past year (e.g., going to restaurants, volunteering, visiting friends) and summed into a composite score.

Data Analysis

Baseline characteristics of participants were summarized in Table 2. We used t-tests, Wilcoxon rank sum tests, Pearson’s correlation, or Spearman’s rank correlation for comparison of continuous variables and chi-squares tests for comparisons of categorical variables to describe the sample and to examine bivariate associations of frailty and loneliness and covariates. We included all participants with baseline loneliness measures in the main analysis, regardless of follow up time, to maximize the available data in this large sample. We applied linear mixed-effects models to examine the longitudinal associations of loneliness and frailty. Follow-up year was included as the time variable. We assessed for linearity of frailty trajectories over time using likelihood ratio tests. We included a quadratic term for time to account for acceleration of frailty trajectories over time. An interaction term for time and baseline loneliness was added to determine if the frailty trajectory was modified by the presence of loneliness at baseline. We also assessed for linearity of frailty trajectories predicted by the level of loneliness using an interaction term for loneliness and time squared and found them to be linear. Because we wanted to measure the influence of loneliness on rate of change in frailty over time, we restricted our analyses to baseline loneliness. We included covariates that could confound the association of frailty and loneliness based on bivariate associations and previous observations that they are associated with frailty and loneliness (4, 22, 38-40). We added covariates to the models in a stepwise fashion first examining an unadjusted model. The second model was adjusted for demographic covariates (e.g., age, gender, education). The third model was further adjusted for cognition and depressive symptoms. The fourth model was adjusted for structural social connection (e.g., social network, marital status), and the fifth was adjusted for functional connection (e.g., social support, and social, physical, and cognitive activities). Comorbidity was not included as a covariate in these analyses to avoid overadjustment and collinearity because five of the seven chronic conditions that comprise the comorbidity index are included in the frailty index. We used baseline covariate data to be concurrent with loneliness and to ensure any observed change in frailty over time was a change in frailty and not a change in the covariates. Individual participants were included as random effects in the model to account for non-independence in measures within participants. Frailty was also modeled as a random effect to allow frailty trajectories to vary by individual participants. We modeled the covariance structure of the models to reflect decreasing correlation between measures over time, as expected in longitudinal cohort studies. We assessed for effect modification between loneliness and social covariates using product terms for loneliness and each covariate. We used an a priori cut-off of p=0.1 to determine whether effect modification was present. We conducted multiple sensitivity analyses. We ran additional analyses assessing associations of loneliness and the frailty phenotype to evaluate if associations were the same for different frailty measures. We ran additional models excluding individuals with only baseline measures and those with less than ten years of follow up to test whether results were driven by healthier or more engaged participants with many years of follow up. To address the question of potentially nonrandom missing data in a longitudinal cohort study, we ran additional sensitivity analysis including the 237 participants dropped for missing baseline loneliness measures. (For a summary of participants with missing data, refer to Appendix A.) All statistical tests were two-tailed and a p<0.05 was considered statistically significant. Data were inspected graphically and statistically, and model assumptions were found to be adequately met. Stata (StataCorp LLC, College Station, TX) version 17.0 was used for all analyses.

 

Results

Baseline characteristics of participants

The mean age at baseline overall was 79.6 ± 7.7. The sample was mostly female, mostly Caucasian, highly educated, and relatively healthy, without disability, or depression (Table 2). The sample was generally well-supported socially with high social support scores and a median of 5 social contacts (IQR 3, 9). The mean frailty index (100-point scale) at baseline was 17.6 ± 10.5, indicating on average, participants were not frail at baseline. There were significant differences between married, widowed, and unmarried groups in terms of age, gender, ethnicity, and education. The groups also differed significantly by social network size, social support, cognitive activity, and loneliness, cognitive function, frailty, and depressive symptoms.

Table 2. Baseline Characteristics of Participants Overall and By Marital Status

Note. SD=standard deviation; IQR=interquartile range; ADL=activities of daily living; IADL= instrumental activities of daily living

 

Baseline Loneliness and Progression of Frailty by Marital Status

We found that the effects of baseline loneliness on frailty progression over time varied by marital status. We therefore present stratified models for married, widowed, and unmarried individuals separately (Table 3). Loneliness predicted frailty progression in married and widowed, but not in unmarried older adults (p=0.009, CI 0.09, 0.64; p=0.005, CI 0.1, 0.58; p=-0.03, CI -0.36, 0.3, respectively). In married individuals, loneliness predicted a 0.37-point increase in frailty per year and in widowed individuals, loneliness predicted a 0.34-point increase in frailty per year. In all three groups, accumulation of frailty accelerated significantly over time (married p=0.06, CI 0.05, 0.07; widowed p=0.06, CI 0.05, 0.08; unmarried p=0.08, CI 0.06, 0.11). At baseline, loneliness was significantly associated with frailty in married and unmarried individuals, but not widowed individuals (married p<0.001, CI 1.4, 4.0; widowed p=0.2, CI-0.47, 2.27; unmarried p=0.003, CI 1.0, 4.5). Baseline cognitive and physical activity were significantly associated with frailty in all 3 groups, while social activity was only significantly associated with frailty in the widowed group. Baseline social support and social network were not significantly associated with frailty. These results are graphically depicted in Figure 1. As demonstrated in the figure, frailty increased over time in all groups. Marital status was associated with frailty at baseline. Loneliness influenced frailty trajectories in all groups except those who were unmarried at baseline. Individuals who were married and not lonely had the lowest predicted frailty after 5 years. Individuals who were widowed and lonely had the highest predicted frailty after 5 years.

Table 3. Baseline Loneliness and Progression of Frailty by Marital Status

Note. CI=confidence interval. *Represents annual rate of change in frailty index, per one unit increase in loneliness. Other terms represent cross sectional associations of variables with frailty index scores. **Represents nonlinear acceleration of frailty over time

Figure 1. This model derived figure illustrates the rate of change in frailty in participants who were lonely vs. not lonely by marital status

The figure shows the influence of marital status on frailty at baseline. The accumulation of frailty over time is also influenced by marital status in married or widowed participants. Individuals who were married and not lonely had the lowest predicted frailty after 5 years. Individuals who were widowed and lonely had the highest predicted frailty after 5 years.

 

We did not find any significant effect modifications by social covariates, loneliness, and time, using an a priori alpha level of p=0.1 except for cognitive activity (p=0.006, CI 0.02, 0.15).

 

Discussion

The main findings of this study were that in 1,931 individuals without dementia at baseline 1) frailty increased in a nonlinear fashion with time, 2) baseline loneliness was also associated with accelerated frailty over time over and above the predicted increase in frailty in married and widowed individuals, and 3) associations of loneliness and frailty persisted even after adjustment for multiple confounders including demographic, health, and structural and functional connection measures. Increased frailty is associated with increased mortality, underscoring the importance of frailty prevention (11). These results identify loneliness as one of several risk factors for frailty and should be considered when targeting risk groups for frailty prevention and intervention.
We found partial support for our hypothesis that effects of loneliness on frailty would not be modified by structural and functional social factors. Effects of loneliness on frailty progression were modified by marital status, but not by social network, social support, or social, cognitive, and physical activity. Loneliness predicted frailty progression in married and widowed, but not unmarried individuals. A possible explanation for these findings is relationship quality. We have made the point that social relationships are crucial for health, however, not all relationships are positive, and relationship conflict could have a negative impact on health. Loneliness is influenced by marital quality 41 and may contribute to the associations between loneliness and frailty observed here, either positively or negatively (42, 43). Spouses of people living with dementia with good marital quality, for example, were lonelier compared to those with low marital quality related to change in their spouse’s cognition (41). Recently widowed individuals with better marital quality also report increased distress and depressive symptoms compared to individuals with low marital quality (44). It is conceivable then that married individuals who were lonely had poorer relationship quality and that affected their frailty trajectories. Widowhood is a great source of trauma and has significant effects on health and mortality, particularly in the early bereavement period (45, 46). Thus, it is possible that mechanisms for the effects of loneliness on frailty might be different in married vs. widowed individuals, or that the effects might be related to relationship quality in both groups. Individuals who are unmarried by choice may have different frailty trajectories than those who are unmarried unwillingly. Most individuals in this sample were married or widowed, and it is possible that we were not powered to detect subtle differences in frailty in this group. The unmarried group was also significantly younger compared to the other groups, which may also influence frailty trajectories in this group. Several measures such as relationship quality, marital transitions, and whether the individual enjoys being alone may shed light on these findings and should be explored in future studies.
Our findings are consistent with previous studies that showed loneliness was associated with frailty cross-sectionally and longitudinally (22). Loneliness was significantly associated with frailty transitions where lonely individuals were more likely to have worsening frailty and less likely to improve (16, 21, 47). One large, longitudinal study reported an association between loneliness and frailty using the frailty index over 14 years of follow using data from the English Longitudinal Study (ELSA) (17). Participants in the ELSA cohort had a mean age of 66.3 at baseline compared to our much older sample with a mean age of 78.3 at baseline (17). We did not find differential effects of loneliness and frailty by age in our sample, which might point to the perniciousness of loneliness as a risk factor for frailty across different life-stages. In contrast, findings from a longitudinal study of Chinese older adults showed baseline loneliness predicted physical frailty in older adults under 75 years old, but not in those older than 75 years over 4 years of follow up (21). However, the study enrolled participants aged 45 and older which may have affected the results (21). Additionally, our findings, that loneliness predicts frailty over time, highlights loneliness as a risk factor for frailty even in a relatively healthy cohort such as the Rush cohort. This emphasizes loneliness as an area of opportunity for early intervention to prevent frailty and promote healthy aging.
Associations between loneliness and frailty were significant even after accounting for structural and functional social factors. This supports loneliness as a discrete construct, separate from closely related but distinctive social factors such as social support, social engagement, or social network. Objective social measures such as social network might be intervened on by expanding one’s social network and social engagement might be intervened upon with social engagement opportunities. Relatively less is known about whether perceived social isolation – loneliness – can be improved by intervening on objective social measures. Several systematic reviews on loneliness interventions proposed various loneliness interventions that include functional support and engagement type interventions ranging from social technologies (e.g., companion robots), to cognitive behavioral therapy and exercise interventions (48-50). Many of these interventions are adaptable to the dramatically changed social landscape since the COVID-19 pandemic (50). More research is needed on loneliness interventions to guide best practices to prevent frailty and promote healthy aging.
Various mechanisms were proposed to explain associations of loneliness and poor health outcomes such as frailty including the inflammatory hypothesis, self-regulation hypothesis, and the stress buffering hypothesis to name a few, but none singularly explain how loneliness impacts health (51-53). Additionally, frailty and loneliness share overlapping neural substrates and pathologies which may also explain potential causal pathways between loneliness and frailty (54-56). Much like the multifactorial nature of frailty and other geriatric syndromes, the causal mechanisms of loneliness and frailty are likely complex (57). We previously synthesized the evidence on the impact of loneliness on function in a systematic review using a nursing framework to highlight the multiple potential causal pathways for this relationship (10). Further research is needed to enhance our understanding of causal pathways to highlight areas of intervention for loneliness and frailty.

Strengths and limitations

Our well-characterized, large sample of older adults with many years of follow up is a strength of our study. However, the characteristics of our sample limit the generalizability of our findings. The cohort is typical of a volunteer sample. Participants were highly educated and cognitively and socially engaged, which is likely different from the general population. Additionally, the sample was mostly female and Caucasian. It is imperative that future research in this area is inclusive of the populations that we serve, and that are particularly affected by these issues. As we used cohort data for these analyses, we cannot rule out reverse causality. Additionally, while missing data is common in longitudinal studies, none of our participants were missing frailty index data and only 237 of our participants were missing baseline loneliness data, and results were unchanged when including these participants in the analyses. Most participants were married or widowed. It is possible the unmarried group was underpowered to detect changes in frailty predicted by loneliness. Our study has many strengths that lend confidence in our findings. The study uniquely enjoys high follow up participation, which reduces attrition-related bias. Additionally, we used a well-validated frailty tool to capture the multi-dimensionality of frailty and avoid underestimation of frailty in the sample. The frailty index, as a continuous measure, also increases statistical power to identify associations. Additionally, associations persisted even after controlling for several potentially confounding variables, including objective and subjective measures of social connection.

Implications

Our findings add to the literature on longitudinal associations of loneliness and frailty with up to 20 years of follow up. We showed that frailty increased over time and that baseline loneliness predicted frailty progression. Our study responds to calls from the US Surgeon General to prioritize research on social isolation and loneliness with this study on loneliness and frailty and the inclusion of structural and functional aspects of social relationships in this analysis to understand the relationship between loneliness and frailty on a broader level (7). Our results highlight loneliness as a health imperative associated with accelerated frailty and should thus be acknowledged and addressed by healthcare providers, much like any other health condition. The provider can leverage the therapeutic relationship to acknowledge the negative experience of loneliness, educate the patient on the importance of addressing loneliness for their physical health, address reversible causes (e.g., sensory or mobility impairment), and connect patients with existing community resources. Our inclusion of functional and structural aspects of social relationships in this analysis, highlights loneliness as a separate construct and supports the recommendation of additional research on effective interventions for loneliness.

 

Conclusions

In a sample of 1,931 older adults without dementia at baseline, loneliness predicted frailty progression. These results emphasize the importance of psychosocial factors and frailty in aging and suggest the need for interventions and policy investment in the prevention and amelioration of loneliness to promote healthy aging and prevent frailty.

 

Acknowledgements: We thank Melissa Fazzari for her expertise on statistical analysis. We are grateful to the study participants and staff of the Rush Alzheimer’s Disease Center.

Conflict of Interest: We have no conflicts of interest to declare.

Sponsor’s Role: The funders had no role in the data collection, study design, analysis, or manuscript preparation for this study.

Ethical standards: The MAP study was approved by the Rush Medical Center Institutional Review Board. Ethical approval for secondary analyses was obtained from the Albert Einstein College of Medicine Institutional Review Board. All participants provided written informed consent when enrolled in the MAP study; as this was a secondary analyses no additional consent was required.

 

SUPPLEMENTARY MATERIAL

 

References

1. Holt-Lunstad J, Smith TB, Baker M, Harris T, Stephenson D. Loneliness and social isolation as risk factors for mortality: A meta-analytic review. Perspect Psychol Sci. 2015;10(2):227-237. doi:10.1177/1745691614568352
2. Lara E, Caballero FF, Rico-Uribe LA, et al. Are loneliness and social isolation associated with cognitive decline? Int J Geriatr Psychiatry. 2019;34:1613-1622. doi:10.1002/gps.5174
3. Perissinotto CM, Stijacic Cenzer I, Covinsky KE. Loneliness in older persons: A predictor of functional decline and death. Arch Intern Med. 2012;172(14):1078-1083. doi:10.1001/archinternmed.2012.1993
4. Wilson RS, Krueger KR, Arnold SE, et al. Loneliness and risk of Alzheimer Disease. Arch Gen Psychiatr. 2007;64(2):234-240.
5. Su Y, Rao W, Li M, Caron G, D’Arcy CD, Meng X. Prevalence of loneliness and social isolation among older adults during the COVID-19 pandemic: A systematic review and meta-analysis. Int Psychogeriatr. 2023;35(5):229-241. doi:10.1017/S1041610222000199
6. DesHarnais Bruce L, Wu JS, Lustig SL, Russell DW, Nemecek DA. Loneliness in the United States: A 2018 National Panel Survey of demographic, structural, cognitive, and behavioral characteristics. Am J Health Promot. 2019;33(8):1123-1133. doi:10.1177/0890117119856551
7. Our epidemic of loneliness and isolation: the US Surgeon General’s advisory on the healing effects of social connection and community. 2023. 2 May 2023. www.hhs.gov/sites/default/files/surgeon-general-social-connection-advisory.pdf.
8. Cacioppo JT, Hughes ME, Waite LJ, Hawkley LC, Thisted RA. Loneliness as a specific risk factor for depressive symptoms: Cross-sectional and longitudinal analyses. Psychol Aging. 2006;21(1):140-151. doi:10.1037/0882-7974.21.1.140
9. Shankar A, McMunn A, Banks J, Steptoe A. Loneliness, social isolation, and behavioral and biological health indicators in older adults. Health Psychol. 2011;30(4):377-385. doi:10.1037/a0022826
10. Pollak C, Verghese J, Blumen H. Loneliness and functional decline in aging: A systematic review. Res Gerontol Nurs. 2023;Online ahead of printdoi:10.3928/19404921-20230503-02
11. Searle SD, Mitnitski A, Gahbauer EA, Gill TM, Rockwood K. A standard procedure for creating a frailty index. BMC Geriatr. 2008;8(24)doi:10.1186/1471-2318-8-24
12. Sathyan S, Ayers E, Gao T, et al. Frailty and risk of incident motoric cognitive risk syndrome. J Alzheimers Dis. 2019;71(Suppl 1):S85-S93. doi:10.3233/JAD-190517
13. Wallace L, Theou O, Godin J, Andrew MK, Bennett DA, Rockwood K. Investigation of frailty as a moderator of the relationship between neuropathology and dementia in Alzheimer’s disease: a cross-sectional analysis of data from the Rush Memory and Aging Project. Lancet Neurol. 2019;18:177-184.
14. Fried LP, Tangen CM, Walston J, et al. Frailty in Older Adults: Evidence for a Phenotype. J Gerontol Med Sci. 2001;56A(3):M146-M156.
15. Rockwood K, Mitnitski A. Frailty in relation to the accumulation of deficits. J Gerontol Med Sci. 2007;62A(7):722-727.
16. Gale CR, Westbury L, Cooper C. Social isolation and loneliness as risk factors for the progression of frailty: The English Longitudinal Study of Ageing. Age Ageing. 2018;47:392-397. doi: 10.1093/ageing/afx188
17. Davies K, Maharani A, Chandola T, Todd C, Pendleton N. The longitudinal relationship between loneliness, social isolation, and frailty in older adults in England: a prospective analysis. Lancet Health Longev. 2021;2:e70-e77. doi:10.1016/S2666-7568(20)30038-6
18. Mehrabi F, Beland F. Effects of social isolation, loneliness, and frailty on health outcomes and their possible mediators and moderators in community-dwelling older adults: A scoping review. Arch Geron Geriatr. 2020;90(104119)doi:10.1016/j.archger.2020.104119
19. Chu B, Zhang W. Impact of transient and chronic loneliness on progression and reversion of frailty in community-dwelling older adults: four year follow-up. BMC Geriatr. 2022;22(642)doi:10.1186/s12877-022-03283-1
20. Sha S, Xu Y, Chen L. Loneliness as a risk factor for frailty transition among older Chinese people. BMC Geriatr. 2020;20(1):300. doi:10.1186/s12877-020-01714-5.
21. Sha S, Chan SHW, Chen L, Xu Y, Pan Y. The associations between trajectories of loneliness and physical frailty in Chinese older adults: Does age matter? Int J Environ Res Public Health. 2022;19(5105)doi:10.3390/ijerph19095105
22. Kojima G, Taniguchi Y, Aoyama R, Tanabe M. Associations between loneliness and physical frailty in community dwelling older adults: A systematic review and meta-analysis. Ageing Res Rev. 2022;81doi:10.1016/j.arr.2022.101705
23. Hoogendijk EO, Smit AP, van Dam C, et al. Frailty combined with loneliness or social isolation: An elevated risk for mortality in later life. J Am Geriatr Soc. 2020;68(11):2587-2593. doi:10.1111/jgs.16716. Epub 2020 Jul 23
24. National Academies of Sciences E, and Medicine. Social isolation and loneliness in older adults: Opportunities for the health care system. 2020.
25. Bennett DA, Schneider JA, Tang Y, Arnold SE, Wilson RS. The effect of social networks on the relation between Alzheimer’s disease pathology and level of cognitive function in old people: a longitudinal cohort study. Lancet Neurol. 2006;5(5):406-412. doi:10.1016/S1474-4422(06)70417-3
26. Wallace LMK, Theou O, Godin J, et al. 10-year frailty trajectory is associated with Alzheimer’s dementia after considering neuropathological burden. Aging Med. 2021;4:250-256. doi:10.1002/agm2.12187
27. Atkins JL, Jylhava J, Pedersen NL, et al. A genome-wide association study of the frailty index highlights brain pathways in ageing. Aging Cell. 2021;20(e134459)doi:10.1111/acel.13459
28. Sathyan S, Ayers E, Gao T, Milman S, Barzilai N, Verghese J. Plasma proteomic profile of frailty. Aging Cell. 2020;19(e13193)doi:10.1111/acel.13193
29. De Jong Gierveld J, Tilburg TV. The De Jong Gierveld short scales for emotional and social loneliness: tested on data from 7 countries in the UN generations and gender surveys. Eur J Aging. 2010;7:121-130. doi:10.1007/s10433-010-0144-6
30. Wilson-Genderson M, Heid AR, Cartwright F, Collins AL, Pruchno R. Change in loneliness experienced by older men and women living alone and with others at the onset of the COVID-19 pandemic. Res Aging. 2022;44(5-6):369-381. doi:10.1177/01640275211026649
31. Barreto M, Victor C, Hammond C, Eccles A, Richins MT, Qualter P. Loneliness around the world: Age, gender, and cultural differences in loneliness. Pers Individ Differ. 2021;169(2021)doi:10.1016/j.paid.2020.110066
32. Kojima G, Walters K, Iliffe S, Taniguchi Y, Tamiya N. Marital status and risk of physical frailty: A systematic review and meta-analysis. J Am Dir Assoc. 2020;21(3):322-330. doi:10.1016/j.jamda.2019.09.017
33. Felix N, Ayers E, Verghese J, Blumen HM. Increased Social Support Reduces the Incidence of Motoric Cognitive Risk Syndrome. Innov Aging. 2022;6(5)doi:10.1093/geroni/igac048
34. Hajek A, Brettschneider C, Mallon T, et al. How does social support affect functional impairment in late life? Findings of a multicenter prospective cohort study in Germany. Age Ageing. 2017;46:813-820. doi: 10.1093/ageing/afx012
35. Buchman AS, Boyle PA, Wilson RS, Fleischman DA, Leurgans S, Bennett DA. Association between Late-Life Social Activity and Motor Decline in Older Adults. Arch Intern Med. 2009;169(12):1139-1146. doi:10.1001/archinternmed.2009.135
36. Radloff LS. The CES-D Scale: a self-report depression scale for research in the general population. . Appl Psychol Meas. 1977;1:385-401.
37. Zimet GD, Powell SS, Farley GK, Werkman S, Berkoff KA. Psychometric characterisitcs of the Multidimensional Scale of Perceived Social Support J Pers Assess. 1990;55:610-617.
38. Bennett DA, Arnold SE, Valenzuela MJ, Brayne C, Schneider JA. Cognitive and social lifestyle: links with neuropathology and cognition in late life. Acta Neuropathol. 2014;127:137-150. doi:10.1007/s00401-013-1226-2
39. Unger JB, McAvay G, Bruce ML, Berkman L, Seeman T. Variation in the impact of social network characteristics on physical functioning in elderly persons: MacArthur Studies of Successful Aging. J Gerontol Soc Sci. 1999;54B(5):S245-S251.
40. Clegg A, Young J, Iliffe S, Olde Rikkert M, Rockwood K. Frailty in elderly people. Lancet. 2013;381(9868):752-762. doi:10.1016/S0140-6736(12)62167-9
41. Hsu KY, Cenzer I, Harrison KL, Ritchie CS, Waite LJ, Kotwal A. In sickness and in health: Loneliness, depression, and the role of marital quality among spouses of persons with dementia. JAGS. 2023;Online ahead of printdoi:10.1111/jgs.18520
42. Robles TF, Slatcher RB, Trombello JM, McGinn MM. Marital quality and health: a meta-analytic review. Psychol Bull. 2014;140(1):140-187. doi:10.1037/a0031859
43. Thomas PA, Richards EA, Forster AK. Is marital quality related to physical activity across the life course for men and women. J Aging Health. 2022;34(6-8):973-983. doi:10.1177/08982643221083083
44. Jung Lee H, Hwang Han S, Boerner K. Psychological and physical health in widowhood: Does marital quality make a difference? Res Aging. 2022;44(1):54-64. doi: 10.1177/0164027521989083
45. Manor O, Eisenbach Z. Mortality after spousal loss: are there socio-demographic differences? Soc Sci Med. 2003;56(2):405-413. doi:10.1016/s0277-9536(02)00046-1
46. Sasson I, Umberson DJ. Widowhood and depression: New light on gender differences, selection, and psychological adjustment. J Gerontol Psychol Sci Soc Sci. 2014;69(1):135-145. doi:10.1093/geronb/gbt058
47. Jarach CM, Tettamanti M, Nobili A, D’Avanzo B. Social isolation and loneliness as related to progression and reversion of frailty in the Survey of Health Aging Retirement in Europe (SHARE). Age Ageing. 2021;50:258-262. doi:10.1093/ageing/afaa168
48. Fakoya OA, McCorry NK, Donnelly M. Loneliness and social isolation interventions for older adults: A scoping review of reviews. BMC Public Health. 2020;20(129)doi:10.1186/s12889-020-8251-6
49. Poscia A, Stojanovic J, Ignazio La Milia D, et al. Interventions targeting loneliness and social isolation among the older people: An update systematic review. Exp Gerontol. 2018;102:133-144. doi:10.1016/j.exger.2017.11.017
50. Williams CYK, Townson AT, Kapur M, et al. Interventions to reduce social isolation and loneliness during COVID-19 physical distancing measures: A rapid systematic review. PloS One. 2021;16(2):e0247139. doi:10.1371/journal.pone.0247139
51. Ong AD, Uchino BN, Wethington E. Loneliness and heatlh in older adults: A mini-review and synthesis. J Gerontol. 2016;64(4):443-449. doi:10.1159/000441651
52. Cacioppo S, Capitanio JP, Cacioppo JT. Toward a neurology of loneliness. Psychol Bull. 2014;140(6):1464-1504. doi:10.1037/a0037618
53. Cohen S, Wills TA. Stress, social support, and the buffering hypothesis. Psychol Bulletin. 1985;98(2):310-357.
54. Gutierrez-Zuniga R, Davis JRC, Ruddy K, et al. Structural brain signatures of frailty, defined as accumulation of self-reported health deficits in older adults. Front Aging Neurosci. 2023;15(1065191)doi:10.3389/fnagi.2023.1065191
55. Lam JA, Murray ER, Yu KE, et al. Neurobiology of loneliness: a systematic review. J Neuropsychopharmacol. 2021;46:1873-1887. doi:10.1038/s41386-021-01058-7
56. Tian Q, Williams OA, Landman BA, Resnick SM, Ferrucci L. Microstructural neuroimaging of frailty in cognitively normal older adults. Front Med. 2020;7doi:10.3389/fmed.2020.546344
57. Holt-Lunstad J. The major health implications of social connection. Psychol Sci. 2021;30(3):251-259. doi:10.1177/09637214219999630

© Serdi 2024