Abstract
Purpose
Conducting a systematic review, meta-analysis and meta-regression regarding the prevalence and correlates of loneliness and social isolation amongst the community-dwelling and institutionalised oldest old (80 years and over).
Methods
Three electronic databases (PsycINFO, CINAHL and Medline) were searched, including studies from inception to January 5, 2023. An additional hand search was conducted by checking included studies’ references, and studies that cited included studies. We included observational studies describing the prevalence and (ideally) the correlates of loneliness, or social isolation, amongst individuals aged 80 years and over. Study design, operationalization of loneliness and social isolation, statistical analysis, characteristics of the sample and key findings were extracted. A random-effects meta-analysis was conducted.
Results
We included 22 studies. The estimated prevalence of severe loneliness was 27.1% (95% CI: 23.7–30.4%). The estimated prevalence of moderate loneliness equalled 32.1% (95% CI: 15.8–48.4%). Moreover, the estimated prevalence of social isolation was 33.6% (95% CI: 28.9–38.2%). There was heterogeneity between the studies. Egger tests suggest the absence of potential publication bias. Meta-regressions showed that the heterogeneity could neither be attributed to the assessment of loneliness nor to the continent where the study was conducted.
Conclusion
Loneliness and social isolation are important problems in the oldest old. In this age group, studies are required, in particular from regions outside Europe. Additionally, longitudinal studies are required to investigate the determinants of loneliness and social isolation amongst individuals aged 80 years and over. Studies using more sophisticated tools to quantify loneliness and social isolation are required.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
Introduction
Loneliness (perceived discrepancy between current and desired social relationships [1]) and social isolation (lack of social activities [2]) are major threats to morbidity and longevity. For example, a recent meta-analysis [3] showed that both loneliness and social isolation were significantly associated with a greater risk of all-cause mortality: the pooled effect size for loneliness was 1.14 (95% CI: 1.08 to 1.20, p < 0.001) and the pooled effect size for social isolation was 1.32 (95% CI: 1.26 to 1.39, p < 0.001). Moreover, another meta-analysis [4] showed that poor social relationships were associated with a 32% (pooled relative risk: 1.32, 95% CI: 1.04 to 1.68) increase in stroke risk and a 29% (pooled relative risk: 1.29, 95% CI: 1.04 to 1.59) increase in coronary heart disease risk.
Both greater loneliness and greater social isolation are associated with a greater likelihood of having mental health disorders. For example, a former meta-analysis showed that loneliness had large effects on mental health outcomes (depression, anxiety, general mental health and suicidality) [5]. Large negative effects of social isolation on mental health (particularly amongst older people) have also been demonstrated [6].
A recent meta-analysis showed a pooled loneliness prevalence of 28.6% (95% CI: 22.9 to 35.0%) and a pooled social isolation prevalence of 31.2% (95% CI: 20.2 to 44.9%) amongst older adults aged 65 years and over (based on 15 countries of four continents: North America, South America, Asia and Europe) during the COVID-19 pandemic [7]. Higher pooled prevalence rates for loneliness were also identified amongst older adults (compared to young adults) in eastern European countries [8]. Moreover, two former meta-analyses showed higher loneliness levels in eastern and southern European countries, compared to northern European countries [8, 9].
Critical life events take place (e.g. loss of friends and relatives or health deteriorations) in later life, which can contribute to loneliness and social isolation amongst the oldest old (individuals aged 80 years and over) [10,11,12]. Moreover, social distancing during the pandemic can also have contributed to increased levels of loneliness and social isolation [13, 14]. To date, nine studies (e.g. [15, 16]) have examined the prevalence—and occasionally the correlates—of loneliness and social isolation amongst the oldest old. However, there has been no systematic review of studies (including meta-analysis and meta-regression) that systematically synthesises the present evidence. Therefore, our aim was to address this knowledge gap (by focussing on community-dwelling and institutionalised individuals aged 80 and over).
Specifically, the aim of this systematic review, meta-analysis and meta-regression was to identify the prevalence and correlates of loneliness and social isolation in the oldest old. Such knowledge is of great importance, particularly in view of the growing number of individuals in this age bracket. Additionally, our work may identify correlates of loneliness and social isolation. Furthermore, our work may clarify potential knowledge gaps and may thus inspire upcoming studies. Moreover, pooling of studies is possible by performing a meta-analysis. This can help deliver a more accurate overview compared to individual studies. A meta-regression can also assist in separating the influence of significant moderating factors (such as region in which the study was conducted, or tool used to quantify loneliness or social isolation).
Methods
Our current work satisfied the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines [17]. A PRISMA checklist can be found in Additional file 1. Furthermore, our work has been registered in the International Prospective Register of Systematic Reviews (PROSPERO, registration number: CRD42022339013). No amendments were made. In January 2023, an electronic search was conducted (three databases: Medline, CINAHL, PsycINFO). In Table 1, our search strategy for Medline is displayed (for the other databases: please see Additional file 2). The suitability was assessed by two reviewers (AV, AH) based on two steps (1: title/abstract screening and 2: full-text screening afterwards). Moreover, we conducted a hand search (i.e. (1) we examined the references of included studies and (2) we examined studies that cited the included studies). Grey literature was not searched. When perspectives on inclusion of studies differed, we used discussions to resolve this (if needed, a third party (HHK) was used). The same procedure applied to assessment of study quality and extracting data.
Our inclusion criteria were:
-
Cross-sectional and longitudinal observational studies identifying the prevalence of loneliness and social isolation amongst the oldest old (80 years and over), covering both, community-dwelling and institutionalized individuals
-
Studies adequately assessing these constructs
-
Studies published in peer-reviewed journals (German or English language)
It should be noted that the appropriate assessment of the constructs was strongly guided by the criteria of the COSMIN guidelines [18].
In contrast, studies exclusively focussing on samples with a certain disorder (e.g. individuals with Parkinson’s disease) were excluded. No restrictions were applied with respect to the time and place of the studies.
A pretest of 100 titles/abstracts was performed before the final inclusion criteria were determined. However, our inclusion criteria remained unchanged. Data were extracted by one reviewer (AV) and cross-checked by another (AH). Study design, definition and operationalization of loneliness, social isolation, characteristics of the sample, statistical analysis and key outcomes were extracted as data. If data were missing, study authors were contacted.
The quality of the studies was assessed using the established Joanna Briggs Institute (JBI) standardised critical appraisal instrument for prevalence studies [19]. The score ranges from 0 to 9 (whereby higher values reflect higher study quality and a lower general risk of bias). Study quality was independently assessed by two reviewers (AV and AH). A cut-off score for excluding studies from meta-analysis was not applied.
With respect to the meta-analysis, in order to pool proportions across the included studies, we used random-effects models because heterogeneity across studies was expected. Following given recommendations, heterogeneity between studies was estimated using the I2 statistic, with I2 values between 25 and 50% considered as low, 50% and 75% as moderate and 75% or more as high heterogeneity [20]. The well-known Stata tool ‘metaprop’ [21] was used to conduct meta-analysis.
It should be noted that loneliness was grouped into “not lonely”, “moderately lonely” and “severely lonely” largely following the procedure proposed by Gardiner et al. [22]. For further details, please see the Additional file 3.
With regard to social isolation, the few single studies dealing with social isolation usually simply distinguished between the presence of social isolation and the absence of it. Therefore, we maintained this dichotomy for the meta-analysis. The detailed presentation of the dichotomization of social isolation in the single studies is provided in Table 2.
Regarding meta-regression, we used the ‘meta regress’ command. More precisely, we performed a random-effects meta-regression with restricted maximum likelihood. Knapp–Hartung adjustment was applied for the standard errors [23]. The coefficients were recalculated. The reason is that the coefficients in the meta-regressions were initially scaled as double arcsine values (rather than proportions) (following Lipsey and Wilson [24]). Meta-regressions were conducted to identify the heterogeneity sources [25].
To detect a potential publication bias, a funnel plot as well as the Egger test (p < 0.05 indicates publication bias) were conducted [26]. Stata 16.1 (College Station, TX, USA) was used in our current study.
Results
Study overview
A flow chart is given in Fig. 1 [17]. More precisely, this figure illustrates the flow of information across the various stages of our systematic review and meta-analysis.
After eliminating duplicates, a total of 6,906 studies underwent screening, specifically through the evaluation of titles and abstracts. During this initial phase, the most prevalent reason for exclusion was the absence of reported data on loneliness or social isolation prevalence amongst the oldest individuals. In the subsequent step, which involved assessing the full text of selected studies, some distinct reasons for exclusion were identified (e.g. not reporting prevalence data or not examining oldest old individuals). When different studies used the same dataset, we selected the study that used the most comprehensive dataset (see also: [27, 28]). Ultimately, our present systematic review incorporated a total of 22 studies [15, 16, 29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48], with all of these studies included in the meta-analysis. The hand search did not reveal additional studies. Table 2 describes important characteristics of the studies and key findings. Within the scope of our analysis, three studies investigated the factors associated with loneliness [16, 40, 43], and their adjusted results are presented in Table 2.
Data were from Europe (n = 12, two studies each from the Netherlands, Sweden, the United Kingdom, one study each from Belgium, Faroe Islands, Finland/Sweden, Germany, Greece, Portugal), South America (n = 1 from Brazil), Asia (n = 4, two studies from China, one study from Israel and one study from India), North America (n = 3, all from the United States) and Oceania (n = 2, both from New Zealand). In sum, 13 studies had a cross-sectional design, and nine studies had a longitudinal design. Large, representative surveys were used in 17 studies. Given the fact that data were collected amongst the oldest old, it has to be noted that twelve had a large sample size (with sample sizes in the higher three- or four-digit range, e.g. 600 or higher). The proportion of women ranged from about 60% to 80% in 15 studies and the average age, if reported, ranged from 80 to 90 years in nine studies. Overall, 15 studies used single item measures to quantify loneliness. The remaining four studies used different versions of the UCLA-tool and the De Jong Gierveld tool (11-item version). Three studies examining social isolation used the LSNS-6, whereas the remaining three studies used single-item measures and a self-developed tool (based on four activities). Additional file 4 displayed the frequency for the tools used to quantify loneliness and social isolation, respectively.
The studies were published between 1994 and 2022 and 14 out of the 22 studies were published in or after 2018 (3 times: 2018, 3 times: 2019, 3 times: 2021 and 5 times: 2022). Data collection took place during the first COVID-19 lockdown in one study [30]. In a second study, data collection took partly place during the COVID-19 pandemic (i.e. June to July 2020) [32]. Further details are shown in Table 2.
Correlates of loneliness
Three studies examined the correlates of loneliness [16, 40, 43]. All three studies found that living alone and the presence of depression are associated with a greater likelihood of loneliness [16, 40, 43]. Two (out of two) studies found that being widowed is also associated with a greater likelihood of loneliness [16, 40].
It may be worth noting that some other studies used loneliness as independent variable. They found that greater loneliness is associated with a greater likelihood of depression [29], lower quality of life [30], lower life satisfaction [39], and poor self-rated health [45].
Correlates of social isolation
The correlates of social isolation were not examined by any of the studies. In contrast, one study used social isolation as independent variable and found that the occurrence of social isolation was associated with an increased likelihood of experiencing self-care problems (OR: 1.92, 95% CI: 1.01–3.65), and pain/discomfort (OR: 2.01, 95% CI: 1.16–3.48) over time [34].
Meta-analysis and meta-regression
The estimated prevalence of severe loneliness was 27.1% (95% CI: 23.7–30.4%; Fig. 2). There was significant heterogeneity between studies (I2 = 98.7%, p < 0.001). The estimated prevalence of moderate loneliness equalled 32.1% (95% CI: 15.8–48.4%, Fig. 3; I2 = 98.6%, p < 0.001). Moreover, the estimated prevalence of social isolation was 33.6% (95% CI: 28.9–38.2%, p < 0.001; I2 = 88.7%, p < 0.001).
With regard to sex-stratified prevalences for loneliness: The estimated prevalence of severe loneliness was 33.6% amongst women (95% CI: 6.6–60.7%, I2 = 99.4, p < 0.001), whereas it was 22.7% amongst men (95% CI: 3.0%–42.4%, I2 = 99.0%, p < 0.001; see Additional file 5 for meta-analysis stratified by sex).
Furthermore, our meta-regression analysis indicated that the assessment tool for loneliness and the continent in which the study took place did not significantly influence the prevalence of loneliness (Table 3; a model that considers each category of the two variables individually can be found in Additional file 6). Please note: The coefficients of the regression reflect the predicted change in the logit given a 1-unit change in the moderator variable.
In a robustness check, we also added the risk of bias scores in a meta-regression (to compare findings from studies at lower and higher risk of bias) [49]. However, this factor also did not achieve statistical significance (p = 0.18). We refrained from doing a meta-regression analysis with prevalence of social isolation due to the small number of studies included. We therefore conducted meta-analysis with prevalence of social isolation for subgroups (by region; by tool used to quantify social isolation; by risk of bias). These findings can be found in Table 4.
The funnel plot (Fig. 4) suggested a potential asymmetry (for loneliness). However, the Egger test (p = 0.75) suggested no potential data asymmetry; this indicates the absence of potential publication bias (Fig. 5).
Quality assessment/risk of bias assessment
Table 5 shows the quality assessment/risk of bias evaluation. The scores varied from 2 to 8 (mean score: 6.4, SD: 1.6), indicating a moderate to good level, and a comparably low danger of bias. The most common limitation was that the response rate was not clearly displayed/unclear handling of low response rate (all studies).
Discussion
The goal of this present work was to identify the prevalence of loneliness and social isolation amongst the oldest old. High prevalence rates were identified. Heterogeneity was observed amongst the studies. The Egger tests indicated the absence of potential publication bias. Meta-regressions conducted to explore the sources of heterogeneity found that neither the assessment of loneliness nor the study continent could be attributed as significant factors contributing to the observed heterogeneity.
The prevalence of loneliness and social isolation amongst the oldest old was high compared to younger individuals. For example, Röhr found a prevalence of social isolation of 12.3% (95% CI: 11.6–13.0) based on the LSNS-6 amongst individuals aged 18 to 79 years in Leipzig Germany about ten years ago (data collection took place between Summer 2011 and Winter 2014) [50]. The prevalence of social isolation identified in this work is, for example, comparable to the prevalence amongst the frequently marginalised group of transgender individuals. This recent study identified a prevalence of 34.4% for objective social isolation based on the LSNS-6 amongst transgender individuals [51]. The data collection took place between April to October 2022. The overall high prevalence rates (for both loneliness and social isolation) of the oldest old may be attributed to the wide variety of mental and somatic disorders which are linked to this very high age [52]. Moreover, individuals aged 80 years and over have a high need for long-term care [53] (e.g. due to functional impairment). A high need for care is associated with higher loneliness levels [54]. Such individuals with a high care need may face difficulties coping with everyday life. For example, mobility restrictions could make it difficult to stay in contact with other people.
With regard to the meta-regressions, it was surprising for us that the loneliness prevalence neither varied by tool used nor the continent in which the study was conducted. This may suggest that loneliness is a general phenomenon amongst the oldest old, and may not be limited to areas or regions where, for example, individuals aged 80 years and over do not live directly or in the immediate vicinity of relatives, and where family cohesion is perhaps also differently pronounced.
Given a sufficient number of studies and data availability, future meta-analyses in this area could further explore other causes of heterogeneity. Those causes could be, for example, educational level, morbidity level or social factors (e.g. social engagement, owning a pet, grandchild care, private care receipt or spousal caregiving) [55,56,57,58]. Moreover, cultural differences, such as differences between individualistic and collectivistic societies, may be a source of heterogeneity and thus should be further explored [59].
It should be noted that the correlates of loneliness amongst the oldest old seem to be comparable to the identified correlates amongst individuals in old age [60]. For example, previous systematic reviews based on cross-sectional studies demonstrated the importance of marital status for both loneliness [60] and social isolation [61] amongst older adults.
However, great caution is required due to the overall very low number of studies investigating the correlates of loneliness amongst the oldest old. Indeed, in view of the limited number of studies examining the correlates of social isolation amongst the oldest old, it is not possible to compare it with prior findings in other age brackets (or other groups) [61].
With regards to study quality, the studies included in this meta-analysis generally exhibited a moderate to high level of methodological rigour. However, some common shortcomings were identified, such as a lack of description regarding response rates/unclear handling of low response rate. For example, a low response rate may reflect the fact that severely impaired individuals (e.g. functional or cognitive impairment) have a lower likelihood of participation; as is commonly found in cohort studies (e.g. [62]). Such impaired individuals often report higher loneliness and isolation scores compared to less impaired individuals. Thus, it may be the case that the prevalence rates reported in this work underestimates the true prevalence rates. Moreover, only very few longitudinal studies have been undertaken and even fewer have exploited the longitudinal data structure using, for example, FE estimates (which provide consistent estimates based on weak assumptions [63]). Overall, one should be very cautious about the causal interpretability of the findings, based on the available evidence.
Our systematic review and meta-analysis showed several gaps in present knowledge. There is a need for a greater number of longitudinal studies to identify the determinants of loneliness, and particularly social isolation amongst the oldest old population. In particular, we recommend the use of techniques to explore causal analysis relationships when dealing with observational data. Such techniques can include, for example, Mendelian randomisation [64, 65], matching approaches such as entropy balancing [66], cross-lagged panel models with fixed effects [67, 68], or difference-in-difference estimators [69].
Additionally, more studies based on more sophisticated tools (e.g. De Jong Gierveld tool or LSNS-6) are required. Moreover, additional studies from neglected geographic areas (particularly: Eastern Europe, South America (except for Brazil), Western Asia, South Asia, East Asia (except for China) and Africa) are required. Lastly, future studies should provide clear reporting of the response rate and should conduct a dropout analysis if necessary. Furthermore, studies are required examining the prevalence of loneliness and social isolation amongst the oldest in times of the challenging COVID-19 pandemic.
We would emphasise some strengths and shortcomings of our own work. It should be noted that this is the first systematic review/meta-analysis synthesising the prevalence and correlates of loneliness and social isolation exclusively amongst the oldest old. Additionally, important procedures were conducted by two reviewers independently. An additional hand search was performed. Furthermore, a meta-analysis and a meta-regression was conducted. A potential shortcoming is that we included solely peer-reviewed articles which may lead to the exclusion of appropriate studies. However, we decided to do so to assure a certain quality of the studies. Whilst most of the studies were published in the past few years, many more studies conducted during the COVID-19 pandemic and investigating individuals aged 80 years and over are needed. Furthermore, we restricted our search to three important databases. However, it may be the case that this choice may lead to an exclusion of studies that may be relevant. If available, other databases (e.g. Embase) should be included in the future research.
Conclusion
Loneliness and social isolation are important problems in the oldest old. In this age bracket, further studies are required from regions outside Europe. Additionally, longitudinal studies are required to investigate the determinants of loneliness and social isolation amongst individuals aged 80 years and over. Furthermore, studies using more pronounced tools to quantify loneliness and social isolation are required.
Data availability
Please see the Additional files 7 and 8.
References
Wenger GC, Davies R, Shahtahmasebi S, Scott A (1996) Social isolation and loneliness in old age: review and model refinement. Ageing Soc 16(3):333–358
Dahlberg L (2021) Loneliness during the COVID-19 pandemic. Aging Mental Health. https://doi.org/10.1080/13607863.2021.1875195
Wang F, Gao Y, Han Z, Yu Y, Long Z, Jiang X, Wu Y, Pei B, Cao Y, Ye J, Wang M, Zhao Y (2023) A systematic review and meta-analysis of 90 cohort studies of social isolation, loneliness and mortality. Nat Hum Behav 7(8):1307–1319. https://doi.org/10.1038/s41562-023-01617-6
Valtorta NK, Kanaan M, Gilbody S, Ronzi S, Hanratty B (2016) Loneliness and social isolation as risk factors for coronary heart disease and stroke: systematic review and meta-analysis of longitudinal observational studies. Heart 102(13):1009–1016
Park C, Majeed A, Gill H, Tamura J, Ho RC, Mansur RB, Nasri F, Lee Y, Rosenblat JD, Wong E, McIntyre RS (2020) The effect of loneliness on distinct health outcomes: a comprehensive review and meta-analysis. Psychiatry Res 294:113514. https://doi.org/10.1016/j.psychres.2020.113514
Rohde N, D’Ambrosio C, Tang KK, Rao P (2016) Estimating the mental health effects of social isolation. Appl Res Qual Life 11:853–869
Su Y, Rao W, Li M, Caron G, D’Arcy C, Meng X (2023) Prevalence of loneliness and social isolation among older adults during the COVID-19 pandemic: a systematic review and meta-analysis. Int Psychogeriatr 35(5):229–241. https://doi.org/10.1017/S1041610222000199
Surkalim DL, Luo M, Eres R, Gebel K, van Buskirk J, Bauman A, Ding D (2022) The prevalence of loneliness across 113 countries: systematic review and meta-analysis. BMJ 376:e067068
Chawla K, Kunonga TP, Stow D, Barker R, Craig D, Hanratty B (2021) Prevalence of loneliness amongst older people in high-income countries: a systematic review and meta-analysis. PLoS ONE 16(7):e0255088
Hajek A, Wolfram C, Spitzer M (2020) König H-H (2020) Association of vision problems with psychosocial factors among middle-aged and older individuals: Findings from a nationally representative study. Aging Mental Health 10(1080/13607863):1725806
Kristensen K, König HH, Hajek A (2019) The longitudinal association of multimorbidity on loneliness and network size: findings from a population-based study. Int J Geriatr Psychiatry 34(10):1490–1497
Petersen N, König H-H, Hajek A (2020) The link between falls, social isolation and loneliness: a systematic review. Arch Gerontol Geriatr 88:104020
Fuller HR, Huseth-Zosel A (2022) Older adults’ loneliness in early COVID-19 social distancing: Implications of rurality. J Gerontol Ser B 77(7):e100–e105
Williams SN, Armitage CJ, Tampe T, Dienes K (2020) Public perceptions and experiences of social distancing and social isolation during the COVID-19 pandemic: a UK-based focus group study. BMJ Open 10(7):e039334
Amaral AS, Afonso RM, Brandão D, Teixeira L, Ribeiro O (2021) Resilience in very advanced ages: a study with centenarians. Int J Aging Hum Dev 93(1):601–618
Brittain K, Kingston A, Davies K, Collerton J, Robinson LA, Kirkwood TB, Bond J, Jagger C (2017) An investigation into the patterns of loneliness and loss in the oldest old–Newcastle 85+ Study. Ageing Soc 37(1):39–62
Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, Shamseer L, Tetzlaff JM, Akl EA, Brennan SE, Chou R, Glanville J, Grimshaw JM, Hróbjartsson A, Lalu MM, Li T, Loder EW, Mayo-Wilson E, McDonald S, McGuinness LA, Stewart LA, Thomas J, Tricco AC, Welch VA, Whiting P, Moher D (2021) The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 372:n71. https://doi.org/10.1136/bmj.n71
Prinsen CA, Mokkink LB, Bouter LM, Alonso J, Patrick DL, De Vet HC, Terwee CB (2018) COSMIN guideline for systematic reviews of patient-reported outcome measures. Qual Life Res 27:1147–1157
Moola S, Munn Z, Tufanaru C, Aromataris E, Sears K, Sfetc R, Currie M, Lisy K, Qureshi R, Mattis PM (2017) Systematic reviews of etiology and risk. In: Aromataris E, Munn Z (eds) Joanna Briggs Institute Reviewer’s Manual. The Joanna Briggs Institute, Adelaide, pp 219–226
Higgins JP, Thompson SG, Deeks JJ, Altman DG (2003) Measuring inconsistency in meta-analyses. BMJ 327(7414):557–560
Nyaga VN, Arbyn M, Aerts M (2014) Metaprop: a Stata command to perform meta-analysis of binomial data. Archives of Public Health 72(1):1–10
Gardiner C, Laud P, Heaton T, Gott M (2020) What is the prevalence of loneliness amongst older people living in residential and nursing care homes? A systematic review and meta-analysis. Age Ageing 49(5):748–757
Hartung J, Knapp G (2001) On tests of the overall treatment effect in meta-analysis with normally distributed responses. Stat Med 20(12):1771–1782
Lipsey MW, Wilson DB (2001) Practical meta-analysis. SAGE publications Inc, Thousand Oaks
Thompson SG, Higgins JP (2002) How should meta-regression analyses be undertaken and interpreted? Stat Med 21(11):1559–1573
Egger M, Smith GD, Schneider M, Minder C (1997) Bias in meta-analysis detected by a simple, graphical test. BMJ 315(7109):629–634
Wood JA (2008) Methodology for dealing with duplicate study effects in a meta-analysis. Organ Res Methods 11(1):79–95
Von Elm E, Poglia G, Walder B, Tramer MR (2004) Different patterns of duplicate publication: an analysis of articles used in systematic reviews. JAMA 291(8):974–980
Chou KL, Chi I (2005) Prevalence and correlates of depression in Chinese oldest-old. Int J Geriatr Psychiatry 20(1):41–50. https://doi.org/10.1002/gps.1246
Costenoble A, De Baets S, Knoop V, Debain A, Bautmans I, Verté D, Gorus E, De Vriendt P (2022) The impact of covid-19 lockdown on the Quality of life, meaningful activities, and frailty in community-dwelling octogenarians: a study in Belgium. Aging Ment Health. https://doi.org/10.1080/13607863.2022.2145457
Dahlberg L, Andersson L, Lennartsson C (2018) Long-term predictors of loneliness in old age: results of a 20-year national study. Aging Ment Health 22(2):190–196. https://doi.org/10.1080/13607863.2016.1247425
Eliasen EH, Weihe P, Petersen MS (2022) The Faroese Septuagenarians cohort: a comparison of well-being before and during the COVID-19 pandemic among older home-dwelling Faroese. Scand J Public Health 50(1):136–143. https://doi.org/10.1177/14034948211056209
Ha JH, Hougham GW, Meltzer DO (2019) Risk of social isolation among older patients: what factors affect the availability of family, friends, and neighbors upon hospitalization? Clin Gerontol 42(1):60–69. https://doi.org/10.1080/07317115.2018.1447524
Hajek A, Brettschneider C, Mallon T, Kaduszkiewicz H, Oey A, Wiese B, Weyerer S, Werle J, Pentzek M, Fuchs A, Conrad I, Luppa M, Weeg D, Mösch E, Kleineidam L, Wagner M, Scherer M, Maier W, Riedel-Heller SG, König HH (2022) Social support and health-related quality of life among the oldest old—longitudinal evidence from the multicenter prospective AgeCoDe-AgeQualiDe study. Qual Life Res 31(6):1667–1676. https://doi.org/10.1007/s11136-021-03070-2
Holmén K, Ericsson K, Winblad B (1994) Loneliness and living conditions of the oldest old. Scand J Soc Med 22(1):15–19. https://doi.org/10.1177/140349489402200103
Jacobs JM, Cohen A, Ein-Mor E, Stessman J (2014) Gender differences in survival in old age. Rejuvenation Res 17(6):499–506. https://doi.org/10.1089/rej.2014.1587
Kim J, Angel JL, Rote SM (2022) A longitudinal study of cognitive and instrumental activities of daily living disablement among the oldest Mexican Americans. J Aging Health 34(2):196–205. https://doi.org/10.1177/08982643211037512
Kotian DB, Mathews M, Parsekar SS, Nair S, Binu VS, Subba SH (2018) Factors associated with social isolation among the older people in India. J Geriatr Psychiatry Neurol 31(5):271–278. https://doi.org/10.1177/0891988718796338
Lay-Yee R, Milne BJ, Wright-St Clair VA, Broad J, Wilkinson T, Connolly M, Teh R, Hayman K, Muru-Lanning M, Kerse N (2022) Prevalence of loneliness and its association with general and health-related measures of subjective well-being in a longitudinal bicultural cohort of older adults in advanced age living in New Zealand: LiLACS NZ. J Gerontol B Psychol Sci Soc Sci 77(10):1904–1915. https://doi.org/10.1093/geronb/gbac087
Leitch S, Glue P, Gray AR, Greco P, Barak Y (2018) Comparison of psychosocial variables associated with loneliness in centenarian vs elderly populations in New Zealand. JAMA Netw Open 1(6):e183880. https://doi.org/10.1001/jamanetworkopen.2018.3880
Liu G, Dupre ME, Gu D, Mair CA, Chen F (2012) Psychological well-being of the institutionalized and community-residing oldest old in China: the role of children. Soc Sci Med 75(10):1874–1882. https://doi.org/10.1016/j.socscimed.2012.07.019
Nóbrega JCL, Medeiros JB, da Silva Freitas JLG, Silva JMM, Simões RFM, Olinda R, de Ferreira Santos JL, Menezes TN, de Oliveira Duarte YA, Zatz M, Matheson D, Santos S (2022) Psychosocial aspects and support networks associated with disability in two longevous populations in Brazil: a cross-sectional study. BMC Geriatr 22(1):110. https://doi.org/10.1186/s12877-022-02810-4
Nyqvist F, Cattan M, Andersson L, Forsman AK, Gustafson Y (2013) Social capital and loneliness among the very old living at home and in institutional settings: a comparative study. J Aging Health 25(6):1013–1035. https://doi.org/10.1177/0898264313497508
Stek ML, Vinkers DJ, Gussekloo J, Beekman AT, van der Mast RC, Westendorp RG (2005) Is depression in old age fatal only when people feel lonely? Am J Psychiatry 162(1):178–180. https://doi.org/10.1176/appi.ajp.162.1.178
Tigani X, Artemiadis AK, Alexopoulos EC, Chrousos GP, Darviri C (2012) Self-rated health in centenarians: a nation-wide cross-sectional Greek study. Arch Gerontol Geriatr 54(3):e342-348. https://doi.org/10.1016/j.archger.2012.01.012
Timmermans EJ, Hoogendijk EO, Broese van Groenou MI, Comijs HC, van Schoor NM, Thomése FCF, Visser M, Deeg DJH, Huisman M (2019) Trends across 20 years in multiple indicators of functioning among older adults in the Netherlands. Eur J Public Health 29(6):1096–1102. https://doi.org/10.1093/eurpub/ckz065
Wang H, Zhao E, Fleming J, Dening T, Khaw KT, Brayne C (2019) Is loneliness associated with increased health and social care utilisation in the oldest old? Findings from a population-based longitudinal study. BMJ Open 9(5):e024645. https://doi.org/10.1136/bmjopen-2018-024645
Zaccaria D, Cavalli S, Masotti B, Gomes Da Rocha C, von Gunten A, Jopp DS (2022) Social isolation and loneliness among near-centenarians and centenarians: results from the fordham centenarian study. Int J Environ Res Public Health. https://doi.org/10.3390/ijerph19105940
Higgins JPT, Altman DG, Gøtzsche PC, Jüni P, Moher D, Oxman AD, Savović J, Schulz KF, Weeks L, Sterne JAC (2011) The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ 343:d5928. https://doi.org/10.1136/bmj.d5928
Röhr S, Wittmann F, Engel C, Enzenbach C, Witte AV, Villringer A, Löffler M, Riedel-Heller SG (2021) Social factors and the prevalence of social isolation in a population-based adult cohort. Soc Psychiatry Psychiatr Epidemiol 57(10):1959–1968
Hajek A, König H-H, Blessmann M, Grupp K (2023) Loneliness and social isolation among transgender and gender diverse people. Healthcare 11(10):1517
Formiga F, Ferrer A, Sanz H, Marengoni A, Alburquerque J, Pujol R, Members OS (2013) Patterns of comorbidity and multimorbidity in the oldest old: the Octabaix study. Eur J Intern Med 24(1):40–44
Hajek A, Brettschneider C, Ernst A, Posselt T, Mamone S, Wiese B, Weyerer S, Werle J, Pentzek M, Fuchs A, Stein J, Luck T, Bickel H, Mösch E, Kleineidam L, Heser K, Maier W, Scherer M, Riedel-Heller SG, König HH (2017) Einflussfaktoren auf die Pflegebedürftigkeit im Längsschnitt (Longitudinal Predictors of the Need for Care). Gesundheitswesen 79(02):73–79. https://doi.org/10.1055/s-0041-111841
Hajek A, Grupp K, Aarabi G, Gyasi RM, Freak-Poli R, Kretzler B, König HH (2023) Long-term care need, loneliness, and perceived social isolation during the COVID-19 pandemic: evidence from the German Ageing Survey. Aging Clin Exp Res 35(6):1377–1384. https://doi.org/10.1007/s40520-023-02411-0
Kretzler B, König H-H, Hajek A (2022) Pet ownership, loneliness, and social isolation: a systematic review. Soc Psychiatry Psychiatr Epidemiol 57(10):1935–1957
Quirke E, König H-H, Hajek A (2021) Extending understanding of grandchild care on feelings of loneliness and isolation in later life: a literature review. Z Gerontol Geriatr 54(5):513–516
Hajek A, Kretzler B, König H-H (2021) Informal caregiving, loneliness and social isolation: a systematic review. Int J Environ Res Public Health 18(22):12101
Hajek A, Brettschneider C, Mallon T, Ernst A, Mamone S, Wiese B, Weyerer S, Werle J, Pentzek M, Fuchs A (2017) The impact of social engagement on health-related quality of life and depressive symptoms in old age-evidence from a multicenter prospective cohort study in Germany. Health Qual Life Outcomes 15(1):1–8
Lykes VA, Kemmelmeier M (2014) What predicts loneliness? Cultural difference between individualistic and collectivistic societies in Europe. J Cross Cult Psychol 45(3):468–490
Dahlberg L, McKee KJ, Frank A, Naseer M (2022) A systematic review of longitudinal risk factors for loneliness in older adults. Aging Ment Health 26(2):225–249
Ejiri M, Kawai H, Ishii K, Oka K, Obuchi S (2021) Predictors of older adults’ objectively measured social isolation: a systematic review of observational studies. Arch Gerontol Geriatr 94:104357
Klaus D, Engstler H, Mahne K, Wolff JK, Simonson J, Wurm S, Tesch-Römer C (2017) Cohort profile: the German ageing survey (DEAS). Int J Epidemiol 46(4):1105–1105g
Brüderl J, Ludwig V (2015) Fixed-effects panel regression. The Sage handbook of regression analysis and causal inference 327–357
Wootton RE, Greenstone HS, Abdellaoui A, Denys D, Verweij KJ, Munafò MR, Treur JL (2021) Bidirectional effects between loneliness, smoking and alcohol use: evidence from a Mendelian randomization study. Addiction 116(2):400–406
Sanderson E, Glymour MM, Holmes MV, Kang H, Morrison J, Munafò MR, Palmer T, Schooling CM, Wallace C, Zhao Q (2022) Mendelian randomization. Nat Rev Methods Primers 2(1):6
Hainmueller J (2012) Entropy balancing for causal effects: a multivariate reweighting method to produce balanced samples in observational studies. Polit Anal 20(1):25–46
Leszczensky L, Wolbring T (2022) How to deal with reverse causality using panel data? Recommendations for researchers based on a simulation study. Sociol Methods Res 51(2):837–865
Allison PD, Williams R, Moral-Benito E (2017) Maximum likelihood for cross-lagged panel models with fixed effects. Socius 3:2378023117710578
Wooldridge JM (2021) Two-way fixed effects, the two-way mundlak regression, and difference-in-differences estimators. SSRN 3906345
Funding
Open Access funding enabled and organized by Projekt DEAL.
Author information
Authors and Affiliations
Contributions
The study concept was developed by André Hajek and Hans-Helmut König. The manuscript was drafted by André Hajek and critically revised by Alina Volkmar and Hans-Helmut König. The search strategy was developed by André Hajek and Hans-Helmut König. Study selection, data extraction and quality assessment were performed by André Hajek and Alina Volkmar, with Hans-Helmut König as a third party in case of disagreements. Meta-analysis and meta-regressions were performed by André Hajek, with critical assessment by Alina Volkmar and Hans-Helmut König. All authors have approved the final version of the manuscript.
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Institutional review board statement
Not applicable.
Informed consent
Not applicable.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Hajek, A., Volkmar, A. & König, HH. Prevalence and correlates of loneliness and social isolation in the oldest old: a systematic review, meta-analysis and meta-regression. Soc Psychiatry Psychiatr Epidemiol (2023). https://doi.org/10.1007/s00127-023-02602-0
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s00127-023-02602-0