Loneliness, Social Isolation, and Living Alone Associations With Mortality Risk in Individuals Living With Cardiovascular Disease: A Systematic Review, Meta-Analysis, and Meta-Regression : Psychosomatic Medicine

Journal Logo

SYSTEMATIC REVIEW/META-ANALYSIS

Loneliness, Social Isolation, and Living Alone Associations With Mortality Risk in Individuals Living With Cardiovascular Disease: A Systematic Review, Meta-Analysis, and Meta-Regression

Long, Róisín M. MSc; Terracciano, Antonio PhD; Sutin, Angelina R. PhD; Creaven, Ann-Marie PhD; Gerstorf, Denis PhD; D’Arcy-Bewick, Sinéad MSc; O’Súilleabháin, Páraic S. PhD

Author Information
Psychosomatic Medicine 85(1):p 8-17, January 2023. | DOI: 10.1097/PSY.0000000000001151

Abstract

INTRODUCTION

The need for social connection and belonging is a natural part of the human experience (1), and the intrinsic need to affiliate with others has been demonstrated across the life span, serving an evolutionary and protective function (2). Empirical evidence that indicates a lack of social connections is detrimental to one’s physical health continues to accumulate (3). A number of social health factors including loneliness, social isolation, and living alone are now considered public health priorities (4–6). These aspects of social health represent distinct constructs (7,8), with research suggesting that the correlation between loneliness and social isolation is weak (9,10). For example, one study using data from the Health and Retirement study found that indicators of social isolation and a two-item loneliness scale were not highly correlated (r = 0.201) (11). Loneliness is defined as an aversive emotional state related to the perception of unfulfilled intimate and social needs and the experience of “feeling alone” (12,13). Consequently, an individual could be surrounded by many people and feel lonely, or they could have only a few people around them and not feel lonely at all (14). Loneliness is commonly measured used single items (15) or the University of California Los Angeles (UCLA) Loneliness Scale (16). Social isolation has been described in a variety of ways that often overlap with other constructs including social integration and social support (17). It is, however, generally considered a measure of being alone, having few social ties, infrequent social contact, and little/irregular involvement in clubs or organizations (17–19). Loneliness and social isolation may also be differentiated from living alone, which reflects a one-person household (20). All three aspects of social health are associated with a number of adverse health outcomes such as depression (21,22), frailty (23,24), poorer cognitive functioning (25–27), and early all-cause mortality (7,28), and compelling evidence implicates these aspects of social health in cardiovascular disease (CVD) (17,29–31).

Although findings across studies vary, three aspects of social health (loneliness, social isolation, and living alone) can impair cardiovascular health (32) both in terms of incidence of disease (31) and in terms of prognosis (17,29,30). Although the specific underlying pathophysiological mechanisms are not fully understood, there are several pathways to suggest that these factors may be critically important to consider among those with established CVD, who may be particularly vulnerable to the impact of loneliness, social isolation, and living alone. The multiple, often interacting, pathways that researchers have explored are complex and include those that contribute to the development and progression of CVD. Several theoretical models have been proposed for how social health may shape health outcomes among individuals with CVD (8,17,33). Each model emphasizes complex, multifactorial pathways that contribute to the development and progression of morbidity-related outcomes, making specific reference to CVD. Underlying physiological mechanisms include immunological responses, hypothalamic-pituitary-adrenal axis activation, and stress responsivity (32,34,35). Research has suggested that stress can have the effect of impairing recovery and accelerating progression of the disease (32), including atherosclerosis (36). Notably, loneliness and social isolation constitute some of the most commonly studied adult stressors (34,37). Other pathways may worsen the progression of CVD, including psychosocial factors such as depression (8,38), with prospective research linking depression to both the development of atherosclerosis and a greater risk of cardiovascular events (39). Lastly, lacking important social relationships may be associated with a number of health-risk behaviors such as smoking, excessive alcohol consumption, and poor medication adherence (40,41) (although the findings seem more consistent for social isolation than for loneliness). Moreover, people who live alone may be at increased risk of mortality if the onset of acute symptoms occurs when they are alone and their access to medical assistance is delayed (42).

This review partly overlaps with a number of prior reviews that have implicated aspects of social health in incident CVD (17,31,43) and adverse CVD outcomes including all-cause mortality (17,44). However, prior reviews examined these associations among broad populations (including those with established illnesses) but not specifically individuals with CVD (7), have examined studies of CVD incidence (31), have combined studies of incidence and mortality (43), have looked at the association between a single aspect of social health (17) and a broad range of cardiovascular outcomes (44), or have included studies examining social support more broadly (43,45). In addition, as the published research literature examining cardiovascular populations is increasing relatively quickly, findings continue to emerge and sometimes vary across studies. For example, a recent study that examined both social isolation and loneliness found that social isolation, but not loneliness, was associated with mortality risk among those living with CVD (29), whereas another study found a significant effect for loneliness and increased risk of mortality but not for living alone (46). In addition, findings in relation to living alone and mortality risk among individuals with established CVD seem inconsistent. Although a significant association was found in a number of studies (47–49), others have found no association after adjusting for potential confounders (50,51). Interestingly, a recent study found (after adjustment) that individuals living alone had a lower risk of all-cause death than those living with others (30). Given the growing, yet varying nature of the evidence implicating aspects of social health within the context of CVD, it is timely to synthesize the existing literature of studies on social health and mortality in people with established CVD. Moreover, within the broader literature, research has highlighted that loneliness and social isolation should be assessed individually and in parallel (7). Thus, our review aimed to focus specifically on these aspects of social health.

CVD remains one of the greatest causes of global mortality and as such represents a major societal and medical burden. In addition, those who survive a major adverse cardiovascular event are at increased risk of subsequent cardiac events and death. Thus, it is important to explore social health determinants in this high-risk group (29). Our review sought to elucidate the nature of the relationship between three important aspects of social health among those with established CVD, to synthesize studies, and investigate whether loneliness, social isolation, and living alone are differentially associated with mortality.

METHODS

This study followed the Centre for Reviews and Dissemination’s Guidance for undertaking reviews in healthcare. A protocol was registered with the International Prospective Register for Systematic Reviews (registration number: CRD42021237588). This systematic review followed the updated Preferred Reporting Items for Systematic Review and Meta-Analysis guidelines (Supplemental Digital Content, https://links.lww.com/PSYMED/A881).

Inclusion Criteria

Studies were required to meet the inclusion criteria to be included in the systematic review and meta-analysis. Included studies were those that a) were prospective longitudinal studies; b) investigated the association of loneliness, social isolation, or living alone with mortality (all-cause or cause-specific); c) comprise participants who had a preexisting CVD diagnosis; and d) were peer-reviewed published articles. No study date or demographic restrictions were applied. English language and human subject restrictions were applied. Studies focusing on social support were excluded because it is distinct from loneliness and social isolation, and there are multiple ways to define and measure it. Resolving such issues to synthesize the literature is beyond the scope of the current review. Studies that did not set out to study loneliness or social isolation explicitly but considered the concepts as defined previously were included (7). Studies published at any date up to and including November 25, 2021, were included.

Participants

Study participants were adults older than 18 years living with preexisting CVD at baseline. We defined CVD as encompassing the diagnoses listed in the International Statistical Classification of Diseases and Related Health Problems, Tenth Revision:

  • I10: Hypertension
  • I11: Hypersensitive heart disease
  • I13: Hypersensitive heart and kidney disease
  • I20–I25: Ischemic heart disease
  • I26–I28: Pulmonary heart disease and diseases of pulmonary circulation
  • I30–I5A: Other forms of heart disease
  • I60–I69: Cerebrovascular diseases
  • I70–I79: Diseases of arteries, arterioles, and capillaries

Predictor (Independent Variable)

Loneliness is defined as the perception of social isolation or the experience of being lonely (7). Social isolation is defined as a measure of the size of an individual’s relationship network, diversity, and frequency of contact. The evaluation of both loneliness and social isolation is heterogeneous across studies. To be included in this review, studies were required to measure one of the three aspects of social health: loneliness; social isolation, and/or living alone. The exposure may be measured by a single item or multi-item instruments.

Outcome Measure (Dependent Variable)

Only studies that provided data on mortality (all-cause or cause-specific) were included in this review.

Literature Search

An extensive search of the literature was conducted between January and November 2021. A variety of strategies were applied to identify potentially relevant studies. The reference lists of pertinent systematic reviews or meta-analysis were scrutinized (17,43). An extensive full-text search using the search terms was completed. Thesaurus and free-text terms (e.g., loneliness, lonely, social* isolat*, living alone) were tailored for each database. The search strategy included thesaurus and free terms for the population of study (e.g., cardiovascular disease, stroke, coronary heart disease), the exposure of study (e.g., loneliness, lonely, social* isolat*), and the outcome of study (e.g., mortality, all-cause, longevity). The search strategy was tailored to each database including MeSH, terms EMTREE, and APA Psychological Index. See Supplemental Digital Content, https://links.lww.com/PSYMED/A881, for the full electronic strategy used to search MEDLINE on Ovid.

Data Collection and Extraction

All titles and abstracts were initially screened by the first author (R.M.L.), and 30% were independently screened by a second review author (S.D). Full texts were obtained and screened for articles that seemed to meet the inclusion criteria. All articles chosen for full-text review were independently screened by the two authors against the inclusion criteria. Where necessary, a third author resolved disagreements (P.S.O.). Data coding was performed by one rater (R.M.L.), and 50% of the extracted data was cross-checked by a second (S.D.) to reduce the likelihood of human error in coding. Study authors were contacted to clarify extracted data when necessary. Data extracted included verifiable characteristics of the studies: a) the number of participants and their composition by age and sex; b) CVD diagnoses; c) cause of mortality; d) length of follow-up; e) measure of living alone, social isolation, and/or loneliness; and f) covariates included in the statistical model. For each study, the reported effect size was extracted. When researchers reported multiple effect sizes across different levels of social isolation, social integration, or perceived social isolation (high versus moderate, moderate versus low), we extracted values with the greatest contrast (high versus low) (7,52). When a study reported multiple effect sizes across time, data were extracted from the longest follow-up period.

Risk of Bias Assessment

Two review authors independently assessed bias risk for each of the included studies based on recommendations of the Cochrane Handbook for Systematic Reviews using the Newcastle-Ottawa Scale that was developed for assessing the quality of observational epidemiological studies (53). Any differences in the assessment of quality between the first and second authors were resolved via discussion between these authors, and if they remained unresolved, they were referred to a third author (one study was referred to a third author). A standard validated criterion for what constitutes a high-quality study has not yet been established for the Newcastle-Ottawa Scale. In keeping with previous research, however, a score of >7 was considered to be a high-quality study (54). Detailed results are provided in Supplemental Digital Content, https://links.lww.com/PSYMED/A881, for risk of bias.

Data Analysis

The effect of the exposure variable (loneliness, social isolation, and living alone) on the risk of mortality was included in the meta-analysis as hazard ratios (HRs). When possible, we conducted meta-analyses of basic models with a few covariates and of fully adjusted models with the most covariates. Similarly, where studies reported effects as odds ratios (ORs) or relative risks, unadjusted values were first extracted (where reported) followed by values from the final model. In cases where the continuous measure of social isolation was in the opposite direction (i.e., higher scores reflected more social contacts and less social isolation), the effect sizes were reversed. Only studies for which an effect estimate and CI were available or could be calculated were included in the meta-analysis.

All statistical analyses were done with the Comprehensive Meta-Analysis (CMA) version 2.0 software. The reported summary statistics were calculated as random-effects models because heterogeneity across studies was expected (55). Data were categorized by exposure type (loneliness, social isolation, living alone). Heterogeneity between studies was assessed by examining forest plots of studies and through I2 statistics. Higher I2 values indicate greater variability between studies than would be expected (range, 0%–100%). Previous researchers have proposed that I2 values ≥75% are indicative of high heterogeneity (56). Where 10 or more studies were identified, we used funnel plots and the Egger test to assess small-sample bias, which is an indicator of possible publication bias (57).

RESULTS

Study Selection

Figure 1 summarizes the search selection process. All results were downloaded and stored in the reference database program EndNote. From the total 4082 references, 1986 remained after 2096 duplicates were removed. Of the 1986 references, 1929 articles were excluded by title and abstract (Figure 1). The resulting 57 articles were retrieved for a detailed assessment and evaluation. We excluded 22 of these articles. Primary reasons for exclusion were as follows: different outcome (n = 8), different predictor variable (n = 4), publication type (n = 4), definition of social isolation or loneliness not outlined previously (n = 3), study population, and participants who were not living with CVD at baseline (n = 2). Studies were excluded based on the first exclusion criteria met; however, some studies met more than one of the exclusion criteria. Most studies were excluded because the outcome was cardiovascular events broadly (58–65).

F1
FIGURE 1:
PRISMA flowchart. PRISMA = Preferred Reporting Items for Systematic Review and Meta-Analysis; CVD = cardiovascular disease. Adapted from Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372:n71. doi: 10.1136/bmj.n71.

Description of Studies

A total of 35 studies were identified for inclusion in this review. Detailed summaries of each study are available within Supplemental Digital Content, https://links.lww.com/PSYMED/A881. Sixteen studies originated in Europe (46,49,66–79), 10 in North America (48,51,80–87), 4 in Asia (29,88–90), and 5 across multiple regions (30,47,50,91,92). All studies were published in English.

All 35 studies provided information on clinical characteristics at baseline. Twelve studies included patients with acute myocardial infarction/myocardial infarction (AMI/MI) (48,51,69,71,74,79,81,82,87–89,91), five with heart failure (30,49,76,83,84), four with stroke (66,68,75,85), four with coronary artery disease (47,50,80,86), three with broad CVD (29,46,67), two with acute coronary syndrome (70,73), one with patients awaiting heart transplantation surgery (77), one included patients who had undergone coronary artery bypass grafting (78), one looked at patients with arterial hypertension (72), and one with atherothrombosis (92). Periods of follow-up ranged between 6 months (68) and 16 years (71). Sample sizes included in analyses ranged from 100 (68) to 44,573 (92). Thirty-two studies included both men and women, two were restricted to women only (74,86), and one was restricted to men only (87). Twenty-six studies provided mean ages of study participants (mean = 65.61 years). Three studies provided data on median age ranging from 56 to 73 years (73,78,92). Four studies provided age ranges that were <60–75+, 60–69, 36–82, and >65 years (29,69,87,91). The age range of the remaining study was unclear (70).

Thirteen studies were stratified by exposure type: socially isolated versus not socially isolated, and living alone versus not living alone. Two studies stratified results by sex (46,88). Loneliness was an exposure variable in 4 studies (29,46,78,84), social isolation was an exposure variable in 11 studies (29,67,69,72,76,77,80–82,86,87), and living alone was an exposure variable in 22 studies (30,46,48–51,66,68,70,71,73–75,79,83,85,88–92). One study compared loneliness and social isolation (29), and one compared loneliness and living alone (46) as predictors of all-cause mortality.

The outcome measure was all-cause mortality and/or cause-specific mortality (cardiac or vascular) in all 35 studies. Most studies controlled for potential confounders. There was variation across studies regarding the number and type of variables that were controlled for. Covariates that were included in some form in statistical models included age (k = 33), sex/gender (k = 20), education status (k = 13), body mass index (k = 15), cardiovascular diagnosis (k = 26), comorbidities more broadly (k = 14), adherence to medical treatments (k = 10), alcohol consumption (k = 7), smoking (k = 18), diabetes (k = 16), working status (k = 6), marital status (k = 4), biomarkers (k = 6), depression/antidepressant use (k = 6), and so on. Table S1, Supplemental Digital Content, https://links.lww.com/PSYMED/A881, provides a full detailed overview of each covariate used in each statistical model.

Narrative Synthesis

All 35 studies were included in the narrative synthesis. As with previous reviews (31), to aid the synthesis of these data, findings are grouped according to the exposure variable: loneliness, social isolation, and living alone.

Loneliness

Four studies analyzed the risk of all-cause mortality among individuals living with CVD as function of loneliness. Loneliness was assessed with a single item in three studies (“do you feel lonely?” “have you felt lonely in the last week?” or “I feel lonely”). One study used items from the adapted UCLA scale (84). Loneliness was treated as a categorical variable in two studies (46,78); two studies analyzed loneliness as a continuous variable (29,84). Three studies found a significant effect for loneliness in both unadjusted and adjusted analyses (46,78,84). One study found no significant association for loneliness after adjusting for potential confounders including social isolation (29). These results suggest some evidence for the negative association of feeling lonely and mortality risk among individuals with preexisting CVD.

Social Isolation

Eleven studies analyzed the risk of mortality among individuals living with CVD as a function of social isolation, the results of which seem somewhat mixed. A majority of studies found a significant association between social isolation/social network/number of social contacts and mortality (all-cause or cause-specific [cardiovascular] mortality) after adjusting for potential confounders (29,67,72,80,81). One study found an unadjusted effect (87), and one study found an adjusted effect for social isolation. A number of studies found no effect for social isolation/social network/number of social contacts and mortality (76,77,82,86). Most studies measured social isolation by a person’s social network, although there was variation in the assessment of social isolation across studies. All studies focused on a structural measure of the number of social contacts a person had or the size of their social network. Methods of assessment ranged from one item that asked participants for their number of social contacts (67) to six items that assessed the size of a person’s social network (29). This included marital status and living alone. A number of studies also used standardized measures including the Mannheim Social Support Interview (93), which included a measure of social network size (80). One study (81) used the Berkman Social Network Index (94), one study (72) used the Blake-McKay method to assess participants social network size (95), one study used the Social Network Questionnaire (82), and one study used Cohen’s Social Network Questionnaire (86). Social isolation was treated as a categorical variable (e.g., socially isolated versus not socially isolated/low social network versus high social network) in five studies (69,72,81,82,87); six studies analyzed social isolation as a continuous variable (29,67,76,77,80,86).

Living Alone

Twenty-two studies analyzed the risk of mortality among individuals living with CVD as a function of living alone. The majority of these studies found a significant effect for living alone and mortality (all-cause or cause-specific) among individuals living with CVD (47–49,68,71,74,79,92). However, some studies found no association after multivariate adjustment (50,51,66,85,89,91). Interestingly, one study found that after adjustments, those living alone had a lower risk of all-cause death than those living with others (30). Two studies stratified outcomes by sex, one found an unadjusted effect for living alone and all-cause mortality among men (46), and one found that this effect was maintained among men after adjustments (75). Two studies stratified the analysis by age: ≤65 versus >65 (90), 45–65, 66–80, and >80 years (92). Both studies found that the incidence of all-cause mortality was higher among younger population who lived alone, whereas this association was not observed among the older population.

Meta-Analysis

Twenty-eight studies provided sufficient data to be included in the meta-analysis. Effect sizes across studies had been calculated by researchers using a variety of methods, with some researchers reporting unadjusted values only, adjusted values only, or both using a variety of covariates.

Loneliness

Because of the small number of studies that assessed mortality as a function of loneliness among people living with CVD, as well as the differences across studies, a meta-analysis was not carried out because the literature was too sparse to support meta-analytic conclusions. The measurement of loneliness differed across studies with three studies using a single-item measurement and one study using an adapted version of the UCLA Loneliness Scale. For example, loneliness was a categorical variable, measured with a single item in two studies, and one of these studies stratified the results by sex. In addition, two studies measured loneliness as a continuous variable, one study used a single-item measure (effects reported as relative risk) and the other used a four-item measure (effects reported as HR).

Social Isolation

Separate analyses were conducted for the categorical (group comparisons) and continuous measurement of social isolation. A significant effect was observed for studies that used a categorical measurement (socially isolated or not socially isolated, high versus low social network) of social isolation on all-cause mortality (k = 3, n = 2648, HR = 1.49, 95% confidence interval [CI] = 1.08–2.04). Similarly, for studies that used a continuous measurement of social isolation, there was a significant association with all-cause mortality (k = 5, n = 2388, HR = 1.11, 95% CI = 1.02–1.20). Heterogeneity between studies ranged from moderate (Q = 2.91, p = .23, I2 = 31.31%) for categorical measurement of social isolation to substantial (Q = 8.23, p = .08, I2 = 51.37%) for studies with a continuous measure of social isolation. Overall, these results suggested that social isolation was associated with increased risk of all-cause mortality among individuals living with CVD.

Living Alone

Fifteen studies analyzed the risk of all-cause mortality among individuals with CVD as a function of living alone, reported as HRs. The effect of living alone on all-cause mortality in the unadjusted analysis was significant (k = 15, n = 80,475, HR = 1.46, 95% CI = 1.28–1.66). Even after covariates were controlled for, the effect of living alone on all-cause mortality remained significant (k = 15, n = 80,243, HR = 1.48, 95% CI = 1.20–1.83). Heterogeneity between studies was high in both the unadjusted (Q = 75.45, p < .01, I2 = 81.45%) and adjusted analyses (Q = 81.72, p < .01, I2 = 82.87%). Five studies analyzed the risk of all-cause mortality among individuals with CVD as a function of living alone as an OR. Two studies reported unadjusted OR values that did not find a significant association for living alone (n = 10,148, OR = 1.76, 95% CI = 0.46–6.8, p = .41). Five studies reported adjusted OR values, and similarly, no effect was found for living alone in the adjusted analyses (k = 5, n = 24,304, OR = 1.06, 95% CI = 0.84–1.34, p = .61). Heterogeneity among these studies was high (Q = 10.43, p = .4, I2 = 90.42%; Q = 12.08, p = .6, I2 = 66.91%, respectively).

Cause-specific mortality was an outcome in four studies included in the meta-analysis, with CVD identified as the cause of death. The effect of living alone on cause-specific mortality was significant in the unadjusted analyses (k = 4, n = 54,428, HR = 1.31, 95% CI = 1.08–1.60); however, no effect was found for living alone in the adjusted analyses (k = 6, n = 56,237, HR =1.28, 95% CI = 0.99–1.66, p = .062). Heterogeneity between studies was high (Q = 11.05, p = .01, I2 = 72.86; Q = 26.63, p = .06, I2 = 81.22%, respectively). Despite variability across studies, the overall results reflect the negative association of living alone on risk of all-cause mortality for people living with CVD, whereas there seems to be a tentative association with CVD-specific morality.

Meta-Regression

There were 13 studies to examine reasons for heterogeneity (I2 = 83%) for the association between living alone and all-cause mortality (effect sizes reported as HRs from the fully adjusted data) (7). We examined categorical (world region, quality assessment rating) and continuous variables (length of follow-up, percentage of female, date of publication, number of covariates) using mixed-effects regression analyses. We found two significant moderators: the associations were stronger in studies with longer follow-up (p = .003) and in studies from Europe (p = .031). Interestingly, a majority of European studies were completed in Nordic countries (k = 6). There was also a trend for year of publication, with more recent studies reporting weaker effects (p = .056; Figure 2).

F2
FIGURE 2:
Meta-regression for living alone, adjusted HRs for risk of all-cause mortality by lengths of follow-up (A), geographical region (B), and year of publication (C). HR = hazard ratio.

Sensitivity Analysis

A “leave-one-out” analysis was run for robustness because of the wide range of values in the meta-analysis for living alone. This analysis indicated that results remained significant (p < .001) and were not dependent on any one study.

Given the heterogeneity of the conditions included, spanning a broad range of CVD-related conditions including stroke and myocardial infarction, a sensitivity analysis was completed. The CVD conditions were stratified by the AMI/MI condition in the unadjusted and adjusted meta-analyses for living alone. Both analyses indicated that results were not dependent on the AMI/MI condition (p = .46, p = .21).

Quality Assessment of Included Studies

The complete quality assessment of included studies is outlined in Supplemental Digital Content, https://links.lww.com/PSYMED/A881. In summary, 32 studies included both sexes, whereas two studies included female participants only (74,86) and one included male participants only (87). Eleven studies had reliable methods for ascertaining the exposure variable (loneliness, social isolation, living alone), which involved structured interviews and record linkage to national registers regarding living arrangements (47,50,51,66,67,73,75,80,86,87). The remaining 24 studies used self-report measures for the predictor variable. Twenty-eight studies derived mortality from national databases or linkage to medical records. Six studies did not state how the outcome was ascertained (31,37,40,52,56,61). In one study, patients’ relatives reported if a death had occurred (56). Thirty-one studies included adjusted analyses, one study included unadjusted analyses (33) only, one study provided crude results only (34), and it was unclear what adjustments were made in a final study (87).

Publication Bias

Fifteen studies reported an HR for living alone and all-cause mortality, and thus, publication bias could be evaluated with the Egger regression test. A visual assessment of the funnel plot for the pooled unadjusted analyses highlighted asymmetry, with larger effects in studies with larger standard error. Egger regression test (β = 2.04, p = .007) was suggestive of small-study effects and potential publication bias that warranted further investigation. Therefore, the trim-and-fill method was applied (96), which estimated four missing studies. After the trim-and-fill approach was applied, the association between living alone and all-cause mortality remained significant (pooled HR = 1.29, 95% CI = 1.13–1.48). Similarly, for the adjusted analyses, the likelihood of publication bias was investigated. Egger regression test suggested small-study effects and potential publication bias (β = 2.24, p = .045). Visual investigation of the funnel plot indicated asymmetry, with larger effects in studies with larger standard error. Thus, the trim-and-fill method (96) was applied and estimated one missing study. After the trim-and-fill method was applied, the adjusted association between living alone and all-cause mortality remained significant (pooled HR = 1.44, 95% CI = 1.16–1.78), indicating that the results of this meta-analysis were robust. See Figures S1 and S2, Supplemental Digital Content, https://links.lww.com/PSYMED/A881, for graphical representations of funnel plots.

DISCUSSION

We examined the association between three aspects of social health: loneliness, social isolation, and living alone and mortality risk among individuals with established CVD. The overall results from the narrative synthesis and meta-analysis suggest that these aspects of social health are associated with an increased risk of mortality among those with established CVD. Importantly, our results indicated notable heterogeneity between studies. This variation may have been due, in part, to different methods for measuring loneliness or social isolation (6), different covariates included in the analyses, and differences in study methodologies, including sample size, population being studied (i.e., type of CVD diagnosis), and follow-up length (91). This review offers a valuable synthesis of the existing prognostic literature. Both narrative syntheses of studies and meta-analytic data support the notion that loneliness, social isolation, and living alone are important social health factors to consider within the context of mortality risk in those with established CVD.

This was the first review of the association between these three aspects of social health and a single outcome of mortality among people with established CVD. As previously highlighted, prior reviews have focused on aspects of social health and CVD incidence (31) or on mortality risk among community samples, including both healthy community samples and those with an established medical condition (only six of the studies included in this review were included in our review) (7); have included studies focusing on incidence and progression with social support (43); or have focused specifically on examining one aspect of social health such as living alone (44) or number of social roles (17) and cardiovascular outcomes. There were two differences between prior reviews and the current study that extended the current cardiovascular evidence base. First, the current review specifically focused on studies with individuals with established CVD. This cohort represents a high-risk group because research indicates that the cardiovascular system is vulnerable to the impact of social health factors, including loneliness and social isolation (37). Second, we examined effect sizes sorted by three aspects of social health among this high-risk population cohort. As previously highlighted, existing reviews have addressed the importance of examining these aspects of social health simultaneously in relation to CVD (7), and because of the historical overlap of these social health factors, there exists a limited understanding of how they individually are associated with cardiovascular health (97). A previous meta-analytic review found that all three aspects comparatively were associated with an increased mortality risk among community samples (7). Our review thus sought to investigate whether loneliness, social isolation, and living alone are differentially associated with mortality specifically among those with CVD. However, because of the limited number of studies assessing the association between loneliness and mortality among this cohort, similar meta-analytical conclusions could not be drawn from our review, highlighting gaps in the existing literature.

Despite our limited ability to quantitatively compare evaluative (loneliness) and structural (social isolation and living alone) aspects of social health, the overall synthesis suggests that future efforts to ameliorate the risk of mortality among those living with CVD should consider all three aspects of social health. Although exact causal mechanisms remain unclear (33), there are several potential mechanisms that may explain our observed findings. The pathways that explain the association between these social health factors and mortality in those with CVD are likely to include numerous biobehavioral mechanisms (8,34), which reflects the complex interplay of these factors on health processes across the life span. Poorer biological profiles have been suggested for these aspects of social health including overactivation of the hypothalamic-pituitary-adrenal axis, the main producer of cortisol (61,98), atypical physiological reactivity to acute stress (99), and a possible link to systemic inflammation (100). Behavioral pathways include engagement in health risk behaviors, such as tobacco use among individuals who are lonely or more socially isolated (101,102), whereas living alone may lead to a delay in obtaining acute care (70) and has been associated with rehospitalization after a cardiovascular event (30).

We were able to investigate the extent to which variability in effect sizes could be explained by certain study characteristics for studies that analyzed the risk of mortality as a function of living alone. We found that the associations were stronger in studies with longer follow-up. Findings from a study with a 16-year follow-up suggested that survival between those living alone and those living with a partner only began to differ after 4 years (71). This finding is contrasted with a previous review that found that associations between social isolation and poorer CVD progression were observed when the length of follow-up was shorter (≤5 years), but it was mixed for longer follow-up (>5 years) (17). More research is needed to better understand the long-term adverse effects of living alone as compared with social isolation on CVD progression and mortality risk. In addition, we found stronger associations in studies from Europe, and more specifically among Nordic countries, where the risk may be greatest because of the increasingly high proportion of people living alone. For example, Sweden reports 50% of households as single households, the highest in Europe (103). Although cultural differences between individualistic and collectivist societies in the experience of loneliness and social isolation have been suggested (104), and those living in individualistic societies are more likely to live alone, we were limited in making such comparisons by the available data. There was also a trend for year of publication, with more recent studies reporting weaker effects (p = .056) and may be consistent with the frequently observed pattern of early studies reporting larger effects (often referred to as the winner’s curse).

Strengths and Limitations

There are several strengths of this review, which include clear definitions and inclusion criteria and rigorous use of standard reporting instruments (105). Our review covered a broad range of CVD conditions. Synthesizing results from studies whose populations consisted of individuals with any form of CVD that measured loneliness, social isolation, or living alone allowed us to access a broader pool of research. In addition, our review distinguished between evaluative (loneliness) and structural (social isolation, leaving alone) variables regarding their associations with risk of mortality, which is important in providing information on potential future interventions (7). Limitations must also be noted. Publication bias was noted in the association between living alone and all-cause mortality. A further limitation of our review with implications for the generalizability and scope of the evidence is that we only included studies published in English language and did not include any “gray literature.” In addition, the studies that focused on living alone only recorded living arrangements at baseline and did not describe how long each participant was living alone or account for the change in living arrangements throughout the follow-up period (50). A further limitation relates to the number of studies examining loneliness, which limited our ability to conduct a meta-analysis and to quantitatively compare associations across all three aspects of social health. Results from our review highlighted that living alone is the aspect of social health that studies have largely focused on among those with established CVD. The disparity between the number of studies focusing on living alone and social isolation compared with loneliness was similarly found in a recent scoping review among population studies where social isolation was the exposure in 72% of studies (6).

Overall findings from our review may have been impacted by heterogeneity across studies, and further analysis to investigate this was not possible for studies focusing on loneliness or social isolation. For example, there is a lack of consistency in the definition and measurement of loneliness and social isolation, which limits comparisons between studies (6). Loneliness has been suggested to encompass two dimensions including emotional and social loneliness (106), which may also be relevant within the context of mortality risk (107). However, each of the studies included in this review examined loneliness as a unidimensional construct. In addition, the majority of studies used a single-item measurement that, despite its wide use in the literature, may be more prone to social desirability bias (108). Similarly, social isolation has been proposed to be a multifaceted construct, reflecting more than simply the number of social contacts a person has, and it remains unclear which aspects of social isolation may or may not drive the observed effects on mortality risk. Lastly, given the broad range of CVD conditions that were included, representing a broad range of levels of mobility, limitations in functioning, and expected longevity, results of the meta-analyses for living alone were stratified according to AMI/MI (yes/no), revealing that the results of our analyses were not dependent on this condition. Nonetheless, the broad range of conditions likely contributed to the heterogeneity in the findings, and it was not possible to stratify the results for loneliness and social isolation according to CVD type.

Future Research

Further rigorous research examining loneliness and social isolation is needed to gain more robust estimates of the observed associations and to better understand the observed heterogeneity across studies, perhaps through identification of potential contextual and methodological moderators (e.g., for whom and within which context does social health impact clinical outcomes among individuals with CVD). This may then allow for future research to make comparisons between those with CVD and those without. For example, over recent decades, there have been developments within the cardiovascular field with regard to prevention and intervention including the use of lipid-lowering treatments. Research now suggests that cardiovascular risk may result from superficial erosion of plaques over time (109). Thus, future research examining the pathways through which aspects of social health impact mortality outcomes among those with CVD may consider the possible pathways through which lipid/inflammatory factors are regulated. Future research may also benefit from analysis of cause-specific mortality outcomes. There were a limited number of studies that examined cause-specific (cardiovascular) mortality outcomes. Recent research has highlighted that limiting analyses to the heterogeneous outcome of all-cause mortality may mask potential mechanisms to account for between social health factors and mortality (110). However, those cause-specific mortality outcomes outlined in this review should be tentatively interpreted given the reliability of cause-specific mortality reports, particularly given the comorbidity typically evident at the end of life (111).

CONCLUSIONS

Supporting public health concerns, our review finds that in those with established CVD, loneliness, social isolation, and living alone are associated with premature mortality. Loneliness, social isolation, and living alone have gained significant attention over recent years, making this review timely. Additional empirical research investigating the associations between loneliness and mortality among those living with CVD is required so that meta-analytic data can be drawn and further comparisons between these three aspects of social health can be made. In addition, in light of the evidence of publication bias among studies that investigate the effect of living alone on all-cause mortality as well as the large methodological differences across studies, there is a need for more rigorous research so that stronger conclusions can be drawn.

Source of Funding and Conflicts of Interest: All authors confirm that there are no conflicts of interest and no source of funding to declare.

REFERENCES

1. Buecker S, Maes M, Denissen JJ, Luhmann M. Loneliness and the Big Five personality traits: a meta–analysis. Eur J Pers 2020;34:8–28.
2. Hawkley LC, Capitanio JP. Perceived social isolation, evolutionary fitness and health outcomes: a lifespan approach. Philos Trans R Soc Lond B Biol Sci 2015;370:20140114.
3. Hawkley LC, Thisted RA, Masi CM, Cacioppo JT. Loneliness predicts increased blood pressure: 5-year cross-lagged analyses in middle-aged and older adults. Psychol Aging 2010;25:132–41.
4. Smith RW, Barnes I, Green J, Reeves GK, Beral V, Floud S. Social isolation and risk of heart disease and stroke: analysis of two large UK prospective studies. Lancet Public Health 2021;6:e232–9.
5. Paul E, Bu F, Fancourt D. Loneliness and risk for cardiovascular disease: mechanisms and future directions. Curr Cardiol Rep 2021;23:1–7.
6. Courtin E, Knapp M. Social isolation, loneliness and health in old age: a scoping review. Health Soc Care Community 2017;25:799–812.
7. Holt-Lunstad J, Smith TB, Barker M, Harris T, Stephenson D. Loneliness and social isolation as risk factors for mortality: a meta-analytic review. Perspect Psychol Sci 2015;10:227–37.
8. Hodgson S, Watts I, Fraser S, Roderick P, Dambha-Miller H. Loneliness, social isolation, cardiovascular disease and mortality: a synthesis of the literature and conceptual framework. J R Soc Med 2020;113:185–92.
9. Mund M, Freuding MM, Möbius K, Horn N, Neyer FJ. The stability and change of loneliness across the life span: a meta-analysis of longitudinal studies. Pers Soc Psychol Rev 2020;24:24–52.
10. McHugh J, Kenny RA, Lawlor BA, Steptoe A, Kee F. The discrepancy between social isolation and loneliness as a clinically meaningful metric: findings from the Irish and English longitudinal studies of ageing (TILDA and ELSA). Int J Geriatr Psychiatry 2017;32:664–74.
11. Coyle CE, Dugan E. Social isolation, loneliness and health among older adults. J Aging Health 2012;24:1346–63.
12. Perlman D, Peplau LA. Toward a social psychology of loneliness. Pers Relatsh 1981;3:31–56.
13. Luanaigh CO, Lawlor BA. Loneliness and the health of older people. Int J Geriatr Psychiatry 2008;23:1213–21.
14. Cacioppo S, Grippo AJ, London S, Goossens L, Cacioppo JT. Loneliness. Perspect Psychol Sci 2015;10:238–49.
15. de Jong Gierveld J, Van Tilburg T, Dykstra PA. Loneliness and social isolation. In: Vangelisti AL, Perlman D, editors. Cambridge Handbook of Personal Relationships. Cambridge, United Kingdom: Cambridge University Press; 2006:485–500.
16. Russell D, Peplau LA, Cutrona CE. The revised UCLA Loneliness Scale: concurrent and discriminant validity evidence. J Pers Soc Psychol 1980;39:472–80.
17. Chin B, Cohen S. Review of the association between number of social roles and cardiovascular disease: graded or threshold effect?Psychosom Med 2020;82:471–86.
18. Steptoe A, Shankar A, Demakakos P, Wardle J. Social isolation, loneliness, and all-cause mortality in older men and women. Proc Natl Acad Sci U S A 2013;110:5797–801.
19. Holt-Lunstad J, Steptoe A. Social isolation: an underappreciated determinant of physical health. Curr Opin Psychol 2022;43:232–7.
20. Smith KJ, Victor C. Typologies of loneliness, living alone and social isolation, and their associations with physical and mental health. Ageing Soc 2019;39:1709–30.
21. 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;21:140–51.
22. Xiu-Ying H, Qian C, Xiao-Dong P, Xue-Mei Z, Chang-Quan H. Living arrangements and risk for late life depression: a meta-analysis of published literature. Int J Psychiatry Med 2012;43:19–34.
23. 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;47:392–7.
24. Maltby J, Hunt SA, Ohinata A, Palmer E, Conroy S. Frailty and social isolation: comparing the relationship between frailty and unidimensional and multifactorial models of social isolation. J Aging Health 2020;32:1297–308.
25. Holwerda TJ, Deeg DJ, Beekman AT, van Tilburg TG, Stek ML, Jonker C, et al. Feelings of loneliness, but not social isolation, predict dementia onset: results from the Amsterdam Study of the Elderly (AMSTEL). J Neurol Neurosurg Psychiatry 2014;85:135–42.
26. Cacioppo JT, Cacioppo S. Older adults reporting social isolation or loneliness show poorer cognitive function 4 years later. Evid Based Nurs 2014;17:59–60.
27. Luchetti M, Ledermann T, Aschwanden D, Nikitin J, O’Súilleabháin PS, Stephan Y, et al. Actor and partner effect of loneliness on episodic memory and verbal fluency: a dyadic multilevel analysis of romantic couples across 28 countries. J Gerontol B Psychol Sci Soc Sci 2022.
28. Rico-Uribe LA, Caballero FF, Martín-María N, Cabello M, Ayuso-Mateos JL, Miret M. Association of loneliness with all-cause mortality: a meta-analysis. PloS one 2018;13:e0190033.
29. Yu B, Steptoe A, Chen L-J, Chen Y-H, Lin C-H, Ku P-W. Social isolation, loneliness, and all-cause mortality in patients with cardiovascular disease: a 10-year follow-up study. Psychosom Med 2020;82:208–14.
30. Zhu W, Wu Y, Zhou Y, Liang W, Xue R, Wu Z, et al. Living alone and clinical outcomes in patients with heart failure with preserved ejection fraction. Psychosom Med 2021;83:470–6.
31. Valtorta NK, Kanaan M, Gilbody S, Ronzi S, Hanratty B. Loneliness and social isolation as risk factors for coronary heart disease and stroke: systematic review and meta-analysis of longitudinal observational studies. Heart 2016;102:1009–16.
32. Li H, Xia N. The role of oxidative stress in cardiovascular disease caused by social isolation and loneliness. Redox Biol 2020;37:101585.
33. Holt-Lunstad J, Smith TB. Loneliness and social isolation as risk factors for CVD: implications for evidence-based patient care and scientific inquiry. Heart 2016;102:987–9.
34. Kivimäki M, Steptoe A. Effects of stress on the development and progression of cardiovascular disease. Nat Rev Cardiol 2018;15:215–29.
35. Shankar A, McMunn A, Banks J, Steptoe A. Loneliness, social isolation, and behavioral and biological health indicators in older adults. Health Psychol 2011;30:377–85.
36. Eng PM, Rimm EB, Fitzmaurice G, Kawachi I. Social ties and change in social ties in relation to subsequent total and cause-specific mortality and coronary heart disease incidence in men. Am J Epidemiol 2002;155:700–9.
37. Hawkley L. Social isolation, loneliness, and health. In: Lobel J, Smith PS, editors. Solitary Confinement: Effects, Practices, and Pathways Toward Reform vol. 185. New York, NY: Oxford Academic; 2019.
38. Hawkley LC, Cacioppo JT. Loneliness matters: a theoretical and empirical review of consequences and mechanisms. Ann Behav Med 2010;40:218–27.
39. Rutledge T, Redwine LS, Linke SE, Mills PJ. A meta-analysis of mental health treatments and cardiac rehabilitation for improving clinical outcomes and depression among patients with coronary heart disease. Psychosom Med 2013;75:335–49.
40. Mahon NE, Yarcheski A, Yarcheski TJ. Social support and positive health practices in young adults: loneliness as a mediating variable. Clin Nurs Res 1998;7:292–308.
41. Walker D, Beauchene RE. The relationship of loneliness, social isolation, and physical health to dietary adequacy of independently living elderly. J Am Diet Assoc 1991;91:300–4.
42. Udell JA, Steg G, Scirica BM, Smith SC, Ohman EM, Eagle KA, et al. Living alone and cardiovascular events in stable outpatients at risk of or with atherothrombosis: insight from the reach registry. Circulation 2011;124.
43. Barth J, Schneider S, Von Känel R. Lack of social support in the etiology and the prognosis of coronary heart disease: a systematic review and meta-analysis. Psychosom Med 2010;72:229–38.
44. Gan T, Yang J, Jiang L, Gao Y. Living alone and cardiovascular outcomes: a meta-analysis of 11 cohort studies. Psychol Health Med 2021;1–13.
45. Mookadam F, Arthur HM. Social support and its relationship to morbidity and mortality after acute myocardial infarction: systematic overview. Arch Intern Med 2004;164:1514–8.
46. Christensen AV, Juel K, Ekholm O, Thrysøe L, Thorup CB, Borregaard B, et al. Significantly increased risk of all-cause mortality among cardiac patients feeling lonely. Heart 2020;106:140–6.
47. Hagström E, Norlund F, Stebbins A, Armstrong PW, Chiswell K, Granger CB, et al. Psychosocial stress and major cardiovascular events in patients with stable coronary heart disease. J Intern Med 2018;283:83–92.
48. Case RB, Moss AJ, Case N, McDermott M, Eberly S. Living alone after myocardial infarction: impact on prognosis. JAMA: Jama 1992;267:515–9.
49. Schockmel M, Agrinier N, Jourdain P, Alla F, Eicher JC, Coulon P, et al. Socioeconomic factors and mortality in diastolic heart failure. Eur J Clin Invest 2014;44:372–83.
50. Gandhi S, Goodman SG, Greenlaw N, Ford I, McSkimming P, Ferrari R, et al. Living alone and cardiovascular disease outcomes. Heart 2019;105:1087–95.
51. Bucholz EM, Rathore SS, Gosch K, Schoenfeld A, Jones PG, Buchanan DM, et al. Effect of living alone on patient outcomes after hospitalization for acute myocardial infarction. Am J Cardiol 2011;108:943–8.
52. Holt-Lunstad J, Smith TB, Layton JB. Social relationships and mortality risk: a meta-analytic review. PLoS Med 2010;7:e1000316.
53. Wells GA, Shea B, O’Connell D, Peterson J, Welch V, Losos M, et al. The Newcastle-Ottawa Scale (NOS) for Assessing the Quality of Nonrandomised Studies in Meta-analyses. Oxford; 2000. Available at: https://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed January 2021.
54. Thomas ET, Guppy M, Straus SE, Bell KJ, Glasziou P. Rate of normal lung function decline in ageing adults: a systematic review of prospective cohort studies. BMJ Open 2019;9:e028150.
55. Dalemans RJP, De Witte LP, Beurskens AJHM, Van Den Heuvel WJA, Wade DT. An investigation into the social participation of stroke survivors with aphasia. Disabil Rehabil 2010;32:1678–85.
56. Deeks JJ, Higgins J, Altman DG, Green S. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 (updated March 2011). The Cochrane Collaboration. 2011;2. Available at: https://training.cochrane.org/handbook. Accessed January 2021.
57. Sterne JA, Sutton AJ, Ioannidis JP, Terrin N, Jones DR, Lau J, et al. Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials. BMJ 2011;343:d4002.
58. Akhtar N, Kate M, Kamran S, Singh R, Bhutta Z, Saqqur M, et al. Sex-specific differences in short-term and long-term outcomes in acute stroke patients from Qatar. Eur Neurol 2020;83:154–61.
59. Bell D, Turpin K. Quality of life at three months following admission to intensive and coronary care units. Clin Intensive Care 1994;5:276–81.
60. Boden-Albala B, Litwak E, Elkind M, Rundek T, Sacco RL. Social isolation and outcomes post stroke. Neurol 2005;64:1888–92.
61. Bu F, Zaninotto P, Fancourt D. Longitudinal associations between loneliness, social isolation and cardiovascular events. Heart 2020;106:1394–9.
62. Horsten M, Mittleman MA, Wamala SP, Schenck-Gustafsson K, Orth-Gomér K. Depressive symptoms and lack of social integration in relation to prognosis of CHD in middle-aged women. The Stockholm Female Coronary Risk Study. Eur Heart J 2000;21:1072–80.
63. Nielsen KM, Faergeman O, Larsen ML, Foldspang A. Danish singles have a twofold risk of acute coronary syndrome: data from a cohort of 138 290 persons. J Epidemiol Community Health 2006;60:721–8.
64. Roijers J, Sunamura M, Utens EM, Dulfer K, Ter Hoeve N, van Geffen M, et al. Marital quality and loneliness as predictors for subjective health status in cardiac rehabilitation patients following percutaneous coronary intervention. Eur J Prev Cardiol 2016;23:1245–51.
65. Sokoreli I, Pauws SC, Steyerberg EW, de Vries GJ, Riistama JM, Tesanovic A, et al. Prognostic value of psychosocial factors for first and recurrent hospitalizations and mortality in heart failure patients: insights from the OPERA-HF study. Eur J Heart Fail 2018;20:689–96.
66. Appelros P, Nydevik I, Viitanen M. Poor outcome after first-ever stroke. Stroke 2003;34:122–6.
67. Birket-Smith M, Hansen BH, Hanash JA, Hansen JF, Rasmussen A. Mental disorders and general well-being in cardiology outpatients—6-year survival. J Psychosom Res 2009;67:5–10.
68. Börü UT, Öztürk ZE, Taşdemir M, Sur H. Living alone following first-ever stroke: a prospective study in Turkey identifying the risk factors and evaluating their effects. N Z Med J 2007;120:U2559.
69. Jenkinson CM, Madeley RJ, Mitchell JRA, Turner ID. The influence of psychosocial factors on survival after myocardial infarction. Public Health 1993;107:305–17.
70. Lammintausta A, Airaksinen JK, Immonen-Räihä P, Torppa J, Kesäniemi AY, Ketonen M, et al. Prognosis of acute coronary events is worse in patients living alone: the FINAMI myocardial infarction register. Eur J Prev Cardiol 2014;21:989–96.
71. Mard S, Nielsen FE. Single living is associated with increased risk of death in patients with heart failure. Eur J Heart Fail Suppl 2011;10:S52–3.
72. Menéndez-Villalva C, Gamarra-Mondelo MT, Alonso-Fachado A, Naveira-Castelo A, Montes-Martínez A. Social network, presence of cardiovascular events and mortality in hypertensive patients. J Hum Hypertens 2015;29:417–23.
73. Nielsen KM, Larsen ML, Foldspang A, Faergeman O. Living alone and atypical clinical presentation are associated with higher mortality in patients with all components of the acute coronary syndrome. Eur J Cardiovasc Prev Rehabil 2007;14:152–4.
74. Norekvål TM, Fridlund B, Rokne B, Segadal L, Wentzel-Larsen T, Nordrehaug JE. Patient-reported outcomes as predictors of 10-year survival in women after acute myocardial infarction. Health Qual Life Outcomes 2010;8:1–10.
75. Redfors P, Isaksén D, Lappas G, Blomstrand C, Rosengren A, Jood K, et al. Living alone predicts mortality in patients with ischemic stroke before 70 years of age: a long-term prospective follow-up study. BMC Neurol 2016;16:80.
76. Rodríguez-Artalejo F, Guallar-Castillón P, Herrera MC, Otero CM, Chiva MO, Ochoa CC, et al. Social network as a predictor of hospital readmission and mortality among older patients with heart failure. J Card Fail 2006;12:621–7.
77. Spaderna H, Zittermann A, Reichenspurner H, Ziegler C, Smits J, Weidner G. Role of depression and social isolation at time of waitlisting for survival 8 years after heart transplantation. J Am Heart Assoc 2017;6:e007016.
78. Herlitz J, Wiklund I, Caidahl K, Hartford M, Haglid M, Karlsson BW, et al. The feeling of loneliness prior to coronary artery bypass grafting might be a predictor of short- and long-term postoperative mortality. Eur J Vasc Endovasc Surg 1998;16:120–5.
79. Vujcic I, Vlajinac H, Dubljanin E, Vasiljevic Z, Matanovic D, Maksimovic J, et al. Long-term prognostic significance of living alone and other risk factors in patients with acute myocardial infarction. Ir J Med Sci 2015;184:153–8.
80. Brummett BH, Barefoot JC, Siegler IC, Clapp-Channing NE, Lytle BL, Bosworth HB, et al. Characteristics of socially isolated patients with coronary artery disease who are at elevated risk for mortality. Psychosom Med 2001;63:267–72.
81. Kreibig SD, Whooley MA, Gross JJ. Social integration and mortality in patients with coronary heart disease: findings from the Heart and Soul Study. Psychosom Med 2014;76:659–68.
82. Lett HS, Blumenthal JA, Babyak MA, Catellier DJ, Carney RM, Berkman LF, et al. Social support and prognosis in patients at increased psychosocial risk recovering from myocardial infarction. Health Psychol 2007;26:418–27.
83. Lu MLR, Davila CD, Shah M, Wheeler DS, Ziccardi MR, Banerji S, et al. Marital status and living condition as predictors of mortality and readmissions among African Americans with heart failure. Int J Cardiol 2016;222:313–8.
84. Manemann SM, Chamberlain AM, Roger VL, Griffin JM, Boyd CM, Cudjoe TKM, et al. Perceived social isolation and outcomes in patients with heart failure. J Am Heart Assoc 2018;7:e008069.
85. Reeves MJ, Prager M, Fang J, Stamplecoski M, Kapral MK. Impact of living alone on the care and outcomes of patients with acute stroke. Stroke 2014;45:3083–5.
86. Rutledge T, Kenkre TS, Thompson DV, Bittner VA, Whittaker K, Eastwood J-A, et al. Psychosocial predictors of long-term mortality among women with suspected myocardial ischemia: the NHLBI-sponsored Women’s Ischemia Syndrome Evaluation. J Behav Med 2016;39:687–93.
87. Ruberman W, Weinblatt E, Goldberg JD, Chaudhary BS. Psychosocial influences on mortality after myocardial infarction. N Engl J Med 1984;311:552–9.
88. Kitamura T, Sakata Y, Nakatani D, Suna S, Usami M, Matsumoto S, et al. Living alone and risk of cardiovascular events following discharge after acute myocardial infarction in Japan. J Cardiol 2013;62:257–62.
89. Nakatsuma K, Shiomi H, Watanabe H, Morimoto T, Taniguchi T, Toyota T, et al. Comparison of long-term mortality after acute myocardial infarction treated by percutaneous coronary intervention in patients living alone versus not living alone at the time of hospitalization. Am J Cardiol 2014;114:522–7.
90. Takeuchi M, Ogita M, Wada H, Takahashi D, Nozaki Y, Nishio R, et al. Comparison of long-term mortality between patients living alone vs patients living with others with acute coronary syndrome treated with percutaneous coronary intervention: reply. Eur Heart J Qual Care Clin Outcomes 2020;6:340.
91. O’Shea JC, Wilcox RG, Skene AM, Stebbins AL, Granger CB, Armstrong PW, et al. Comparison of outcomes of patients with myocardial infarction when living alone versus those not living alone. Am J Cardiol 2002;90:1374–7.
92. Udell JA, Steg PG, Scirica BM, Smith SC Jr., Ohman EM, Eagle KA, et al. Living alone and cardiovascular risk in outpatients at risk of or with atherothrombosis. Arch Intern Med 2012;172:1086–95.
93. Veiel H. The Mannheim interview on social support. Soc Psychiatry Psychiatr Epidemiol 1990;25:250–9.
94. Berkman LF, Syme SL. Social networks, host resistance, and mortality: a nine-year follow-up study of Alameda County residents. Am J Epidemiol 1979;109:186–204.
95. Blake RL, McKay DA. A single-item measure of social supports as a predictor of morbidity. J Fam Pract 1986;22:82–4.
96. Duval S, Tweedie R. A nonparametric “trim and fill” method of accounting for publication bias in meta-analysis. J Am Stat Assoc 2000;95:89–98.
97. Freak-Poli R, Ryan J, Neumann JT, Tonkin A, Reid CM, Woods RL, et al. Social isolation, social support and loneliness as predictors of cardiovascular disease incidence and mortality. BMC Geriatr 2021;21:1–14.
98. Liu GD, Wang H, Huang ZG. Social isolation, chronic diseases and prevention in old age. J Am Geriatr Soc 2019;67:S663.
99. Brown EG, Gallagher S, Creaven AM. Loneliness and acute stress reactivity: a systematic review of psychophysiological studies. Psychophysiology 2018;55:e13031.
100. Smith KJ, Gavey S, RIddell NE, Kontari P, Victor C. The association between loneliness, social isolation and inflammation: a systematic review and meta-analysis. Neurosci Biobehav Rev 2020;112:519–41.
101. Leigh-Hunt N, Bagguley D, Bash K, Turner V, Turnbull S, Valtorta N, et al. An overview of systematic reviews on the public health consequences of social isolation and loneliness. Public Health 2017;152:157–71.
102. Ikeda T, Cable N, Saito M, Koyama S, Tsuji T, Noguchi T, et al. Association between social isolation and smoking in Japan and England. J Epidemiol 2021;31:523–9.
103. Ng N, Santosa A, Weinehall L, Malmberg G. Living alone and mortality among older people in Västerbotten County in Sweden: a survey and register-based longitudinal study. BMC Geriatr 2020;20:1–7.
104. 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 Dif 2021;169:110066.
105. Malcolm M, Frost H, Cowie J. Loneliness and social isolation causal association with health-related lifestyle risk in older adults: a systematic review and meta-analysis protocol. Syst Rev 2019;8:48.
106. Weiss RS. Loneliness: The Experience of Emotional and Social Isolation. The MIT Press; 1973.
107. O’Súilleabháin PS, Gallagher S, Steptoe A. Loneliness, living alone, and all-cause mortality: the role of emotional and social loneliness in the elderly during 19 years of follow-up. Psychosom Med 2019;81:521–6.
108. Dykstra PA. Older adult loneliness: myths and realities. Eur J Ageing 2009;6:91–100.
109. Quillard T, Franck G, Mawson T, Folco E, Libby P. Mechanisms of erosion of atherosclerotic plaques. Curr Opin Lipidol 2017;28:434–41.
110. Ward M, May P, Normand C, Kenny RA, Nolan A. Mortality risk associated with combinations of loneliness and social isolation. Findings from the Irish Longitudinal Study on Ageing (TILDA). Age Ageing 2021;50:1329–35.
111. Gerstorf D, Smith J, Baltes PB. A systemic-wholistic approach to differential aging: longitudinal findings from the Berlin Aging Study. Psychol Aging 2006;21:645–63.
Keywords:

loneliness; social isolation; living alone; cardiovascular disease; mortality; AMI/MI = acute myocardial infarction/myocardial infarction; CI = confidence interval; CVD = cardiovascular disease; HR = hazard ratio; OR = odds ratio; UCLA = University of California Los Angeles

Supplemental Digital Content

Copyright © 2022 by the American Psychosomatic Society