The effectiveness of e-Interventions on reducing social isolation in older persons: A systematic review of systematic reviews

Abstract

As the older adult population group has been increasing in size, there has been evidence of growing social isolation and loneliness in their lives. The increased use of information communication technology and Internet-supported interventions has stimulated an interest in the benefits of e-Interventions for older people and specifically in having a role in increasing social networks and decreasing loneliness. A systematic review of e-Interventions to reduce loneliness in older people was conducted with the aim to synthesize high quality evidence on the effectiveness of e-Interventions to decrease social isolation/loneliness for older people living in community/residential care. A systematic search of 12 databases for reviews published between 2000–2017 was conducted using search term synonyms for older people, social isolation and interventions. Three independent researchers screened articles and two reviewers extracted data. The Revised-Assessment of Multiple Systematic Reviews was used to assess the quality of reviews. The final search identified 12 reviews, which included 22 unique primary research studies evaluating e-Interventions for social isolation or loneliness. The reviews were of moderate quality and the primary studies showed a lack of rigor. Loneliness was most frequently measured using the University California Los Angeles Loneliness Scale. Despite the limitations of the reviewed studies, there is inconsistent and weak evidence on using e-Interventions for loneliness in older people.

Keywords

E-Intervention loneliness older people social isolation systematic review

Introduction

Globally, older persons are a rapidly growing population group.¹ This growth has been accompanied by evidence of increased social isolation and loneliness, especially in older people living in residential care.²^–⁴ Social isolation and loneliness are often used interchangeably, with loneliness referring to the subjective feeling of being lonely, and social isolation referring to the objective paucity of social networks.⁵

There has been an exponential growth in the use of information communication technology and Internet-supported interventions⁶ in health. Research has shown evidence of improved access, delivery and management of healthcare⁷ including social healthcare.⁸ These interventions could offer the elderly access to increased communication pathways that can provide for social connectedness.⁹ Considering age-specific requirements for adoption of technology,^10,11 some evidence is emerging of behavioural intentions to use technology,⁹ presenting possibilities of therapeutic use.¹² Several systematic reviews have been published on the effectiveness of information communication technology and Internet-supported interventions for social isolation and loneliness, but the quality of these reviews, and the scope and heterogeneity and rigor of the primary studies included therein, are not known. To address this gap, a review question was developed: What is the level of evidence on the effectiveness of e-Interventions to reduce social isolation and loneliness in older people living in community/residential care? In this review, e-Interventions were defined as interventions that are delivered via Internet-supported,⁶ information communication technology (ICT) or other electronic technologies, with or without human support.¹³

Methods

An initial search identified several systematic reviews (SRs) of interventions for social isolation/loneliness. Though some reviews were found prior to 2000, these were either non-systematic or qualitative reviews.¹⁴ A protocol for a systematic review to address the review question was approved by the university’s Ethics Committee. Three independent reviewers, screened, selected the data and two reviewers extracted data during the review process. Kappa levels of agreement were calculated and disagreements were resolved through consensus and discussion.

The review was conducted in seven stages. See the Supplementary Material for details of search terms and criteria used.

Stage 1: A systematic comprehensive search was conducted in March 2017, using12 databases (Academic Search Complete, Centre for Reviews and Dissemination searching in the Database of Abstracts of Reviews of Effects (DARE), Health Technology and Assessment (HTA) database, Academic Search Complete, Cochrane Database of Systematic Reviews (CDSR), Cumulative Index of Nursing and Allied Health Literature (CINAHL), Joanna Briggs Institute (JBI), PubMed, Prospero, PsycINFO, Psych Articles, Sabinet, Science Direct and Scopus). The following search terms were used: (‘reviews’ and ‘2000–2017’ and synonyms for ‘cognitively intact older people’, ‘social isolation and loneliness’ and ‘interventions’). In addition, reference lists were scrutinized for additional reviews.

Stage 2: Relevant reviews were selected by screening the abstracts of publications found in Stage 1 using the Participants Interventions (Comparison) Outcome Study [PI(C)OS] inclusion and exclusion criteria (Table 1).

Stage 3: The selected reviews were assessed for quality using the Revised-Assessment of Multiple Systematic Reviews (R-AMSTAR) quality rating tool for systematic reviews.¹⁵ The 11-item R-AMSTAR allows for quantifiable assessment of systematic reviews providing a measurement of the methodological quality as required for conclusions about clinical evidence with a possible score of 44.¹⁵ A criterion cut-off point, reviews with scores<22 (an average of two marks per criteria which indicates that, on average, only two criteria for each of the domains were satisfied) was used to exclude reviews viewed a low quality.^15,16

Stage 4: Data from the selected reviews were extracted and the list of underlying primary studies was entered into Microsoft Excel.

Stage 5: The abstracts of the primary studies were reviewed using the PI(C)OS inclusion and exclusion criteria and assessment of study design criteria based on the Centre for Evidence-Based Medicine (CEBM) levels of evidence (LoEs) and grades or recommendations.¹⁷ The CEBM guidelines for study quality are: I: SR, II: randomized controlled trial (RCT), III-1: quasi-experimental studies (QE), III-2: cross sectional studies, and III-3: two group comparative surveys. All studies rated <III (case series, post-test or pre-test/post-test or surveys only) were excluded as comparison designs are inadequate to establish causation due to design problems, bias and confounders which can lead to erroneous conclusions.^18,19

Stage 6: A final list of included primary studies was selected and a table of studies per review was compiled to identify the individual unique studies across the reviews. The list of primary unique studies was then classified using an adapted Internet-supported interventions classification,⁶ comprised of three categories of e-Interventions, namely: (a) online activities (e.g. social media, computer/Internet use and training); (b) interpersonal communication (e.g. videoconferencing (VC) and Internet-based communication); (c) Internet-operated therapeutic software (e.g. robotics and gaming).

Stage 7: Data were extracted and synthesised from the review to address the review objectives.

Table 1.

Inclusion and exclusion criteria Participants Interventions (Comparison) Outcome Study [PI(C)OS].

Inclusion	Exclusion
Participants (P): Persons ≥60 years (or at least 80% of the sample) living in community or residential settings with no major neurocognitive impairments	Participants with major neurocognitive disorders, living in frail care and/or sight and hearing impaired.
Interventions (I): E-Interventions (interventions that employ any information communication technology (ICT) or Internet- supported⁶ intervention delivery mode with or without human support).	Smart devices for home or telehealth
Outcomes (O): Quantitative outcome data focusing on and/or social isolation/loneliness	Physical health outcomes/functional capabilities/lifestyle changes and NO impact on social isolation/loneliness
Study (S): Systematic review: Reviews that used systematic review methodology to review quantitative comparative studies.	Reviews without a systematic review process or reviews of qualitative studies only

Results

Search and selection

The initial search identified 719 publications. After exclusion of duplicates and irrelevant publications, 160 publications remained. The 160 publications were screened by three reviewers using the PI(C)OS criteria (Table 1). Twenty-six full text reviews were assessed for eligibility, resulting in 12 reviews which included primary studies with e-Interventions. An additional review²⁰ was identified at this stage and added, resulting in 13 reviews for quality assessment. One review²¹ was subsequently excluded (R-AMSTAR = 19) due to a lack of a priori systematic review processes (Figure 1). The levels of agreement between reviewers were good (Kappa = 0.93) and disagreements was resolved through consensus discussion.

Figure 1.

PRISMA flow diagram search process. R-AMSTAR: Revised-Assessment of Multiple Systematic Reviews. PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Characteristics of 12 selected reviews

Eight systematic reviews without meta-analysis,²²^–²⁹ three meta-analyses^14,20,30 and one integrative review³¹ were included. Five reviews were published in 2016.^23,26^–^28,31 Four reviews focused on e-Interventions targeting social isolation/loneliness in older people,^20,23,28,30 two^27,29 on e-Interventions for older people (i.e. not limited to social isolation/loneliness), and six on interventions for social isolation/loneliness in older people (i.e. not limited to ICTs).^14,22,24^–^26,31

Quality of 12 selected reviews

The selected reviews were of moderate quality (average R-AMSTAR rating = 29.3 (range 24–33) (Table 2). Three reviews conducted meta-analyses by selecting homogenous primary evaluation studies.^14,20,30 Methods for assessing the rigor of the primary studies varied and the scientific quality was often not considered in the recommendations of the review.

Table 2.

Characteristics of the 12 reviews with 22 e-Intervention studies.

Author	Review	Primary e-Intervention studies	E-Interventions	Outcomes
Reviews focused on e-Interventions for social isolation in older people (n = 4)
Chen and Schultz (2016)²³	Period: 2002–2015 # Databases: 9 # Studies: 25 # e-Intervention studies:12^33–^36,38^–⁴³ # Excluded: 13 (carer 1, qualitative 11, study design 1) R-AMSTAR = 29 Heterogeneity study approach QA: EPHPP (4 studies high quality)	n = 12 # Subjects: 35–5203 Age 55–105 Sex: Mixed Countries: Finland and Slovenia:1, Israel:1, Netherlands:3, Taiwan:2, USA:5 Settings: community, residential care	Online activities Computer/ Internet training: 5 Computer use: 3 Interpersonal communication VC: 2 Skype: 1 Internet-operated therapeutic software Wii: 1	Inconclusive impact on loneliness and <3 months (NS) Scale DJGLS: 4 UCLALS: 6 Other: 2 Self-report Hughes loneliness scale
Khosravi, Rezvani and Wiewiora, (2016)²⁸	Period: 2000–2015 # Databases:04 # Studies extracted: 35 # e-Interventions: 19 ³³^–^38,40^–^42,44^–⁵³ # Excluded: 16 (not loneliness 4, qualitative 2, duplicate study 4, study design 2 participants 4) R-AMSTAR 26 Including weak designs QA: Subjective 1-3 rating of quality and effectiveness (29)	n = 19 # Subjects: 12–3075 Age: 50–93 Sex: Mixed; 2 F studies Countries: Canada 1, Finland and Slovenia:1, Israel:1, Japan 1, Netherlands:4, Taiwan:1, Turkey:1, USA:9 Settings: Community; Residential care	Online activities Computer/Internet training: 8 Internet use: 3 SNS use: 3 Interpersonal communication VC: 1 Internet-operated therapeutic software Robotics: 2 Computer-generated exercise: 1 Videogames 1	70% of studies show changes in social isolation and loneliness (NS provided) Scale DJGLS: 5 UCLALS: 12 Other: 2 AKO scale Self-report
Bornemann (2014)²⁰ Meta-analysis	Period: 2004–2014 # Databases:15 # Studies: 7 # e-Intervention studies: 7^33–³⁹ R-AMSTAR 32 Heterogenous outcome measures QA: EPPCP risk of bias varied and unclear	n = 7 # Subjects: 12–60 Age: 71–82 Sex: Mixed Countries: Israel:1, Netherlands:2, USA:2 Settings: Community / day care centres	Online activities Computer/ internet training: 7	Overall mean weighted ES for loneliness (ICT) (n = 7) NS decreased loneliness (Z = 0.44; p = 0.37).
Choi et al. (2012)³⁰ Meta-analysis	Period: Jan 2000– July 2012 # Databases:5 English and 5 Korean # Studies: 6 # e-Intervention studies 5^33–³⁷ # Excluded: 1 (Depression) R-AMSTAR 32 Heterogenous outcome measures QA: Cochrane risk of bias varied and unclear	n = 5 # Subjects: 12–60 Age: 64–82 Sex: mixed Countries: Israel:1, Netherlands:2, USA:2 Settings: Community/day care centres	Online activities Computer/ internet training: 5	Overall mean weighted ES for loneliness (n = 5) decreased loneliness (Z = 2.085, p = 0.037) Scale DJGLS: 3 UCLALS: 2
Reviews focused on e-Interventions for older people (i.e. not limited to loneliness) (n = 2)
Morris et al. (2014)²⁹	Period: Jan 2000– Feb 2013 # Databases: 06 # studies:18 # e-Interventions: 3^33,^35,53 # Excluded: 15 (no loneliness 12, duplicate studies 2, <60 1) R-AMSTAR: 29 Heterogenous outcome: social connectedness QA: Downs and Black checklist including LoE II and III studies	n = 3 # Subjects: 12–191 Age: 66–82. Sex: Mixed; Countries: Netherlands:2, USA:1 Setting: Community Focus: Assistive technologies	Online activities Computer/ internet training: 2 Internet-operated therapeutic software Wii: 1	Emerging evidence ∼ to improve some dimensions of social connectedness: social support, loneliness, and social networks Scale DJGLS: 2 UCLALS:1
Khosravi, and Ghapanchi, (2016)²⁷	Period: 2000–2015 # Databases:4 # Studies: 41 # e-Interventions: 4^41,^46,48,49 # Excluded: 37 (qualitative 7, physical 23, QoL 3, duplicate study 2, study design 2) R-AMSTAR 26 Heterogeneity in studies, outcomes and small sample sizes robotics QA: Subjective 1-3 rating of quality and effectiveness (29)	n = 4 # Subjects: ND Age: 60 + Sex: ND Countries: inland/Slovenia:1, USA:3 Settings: Community; Residential care Focus: Smart technologies	Online activities Computer/ internet training: 2 Internet-operated therapeutic software Robotics: 1 Computer-generated exercise: 1	Online activities >robotics can impact on social isolation (NS provided) Scale UCLALS: 2 Other: 2 AKO scale Self-report
Reviews focused on interventions for social isolation in older people (n = 6)
Cattan et al. (2005)²²	Period: 1970–2002 # Databases:12 # Studies: 30 # e-Interventions: 6 (incl 1)³⁶ # Excluded: 5 (<2000) R-AMSTAR: 32 Non-English excl. Broad definition ‘aged’ Studies old QA: Own criteria	n = 1 # Subjects: 93 Mean age: 71 Sex: Mixed Country: USA Setting: Community	Online activities Computer and Internet training/use: 1	Trend to less loneliness (NS) Scale UCLALS: 1
Cohen-Mansfield and Perach (2015)²⁴	Period: 1996–2011 # Databases:6 # Studies: 34 # e-Interventions: 11 (8 incl)³³^–^{36,38,39,44,46} # Excluded: 3 (2 < 2000, 1 radio) R-AMSTAR 24 Methodological issues of studies and old QA: Not assessed	n = 8 # Subjects: 9–236 Age: 55–93 Sex: Mixed Countries: Israel: 1, Netherlands: 2, Taiwan: 2, USA: 3 Settings: Community; Residential care	Online activities Computer/internet training/use: 4 Interpersonal communication VC: 2 Internet-operated therapeutic software Computer generated exercise:1 Robotics: 1	Technology interventions to alleviate loneliness were effective in both one-on-one,^33,38 and group formats,³⁴ (NS provided) Scale DJGLS: 2 UCLALS: 6
Dickens et al. (2011)²⁵	Period: <May 2009; # Databases: 10 # Studies: 32 # e-Interventions: 5 (3 incl)^33,35,36 # Excluded: 2 (2 < 2000) R-AMSTAR 32 Heterogeneity QA: Cochrane bias of risk tools	n = 3 # Subjects: 29-236 Age: 66-72 Sex: Mixed; Countries: Netherlands:2, USA:1 Settings: Community; Residential care	Online activities Computer/internet training/use: 3	Effective but less than activities and support (NS provided) Scale DJGLS: 2 UCLALS: 1
Franck et al. (2016)²⁶	Search period: Jan 2009–Dec 2013 # Databases:8 # Studies: 6 # e-Interventions: 2 (1 incl)⁵³ # Excluded: 1 (radio) R-AMSTAR: 31 Weak quality evidence QA: Downs and Black	n = 1 # Subjects: 35 Age: 82 Sex: Mixed Country: USA Setting: Residential care	Internet-operated therapeutic software Videogames: 1	Older people playing Wii lowered loneliness pattern more than TV (NS) Scale UCLALS:1
Gardiner et al. (2016)³¹	Period: 2003–Jan 2016 # Databases: 6 # Studies extracted: 39 # e-Interventions: 7 (5 incl)^{33,39,41,44,54} # Excluded: 2 (qualitative 1, study design 1) R-AMSTAR: 33 Good quality studies QA: EPPCP	n = 5 # Subjects:12–85 Age: 57–85 Countries: Finland/Slovenia: 1, Netherlands: 1, Taiwan: 1, USA: 2 Settings: Community; Residential care	Online activities Computer/internet use/training: 3 Interpersonal communication VC:1 Internet-operated therapeutic software Robot:1	Some success in loneliness reduction but evidence weak (NS provided) Scale DJGLS: 1 UCLAS:2 Other: 2 Hughes loneliness scale Self-report
Masi et al. (2011)¹⁴ Meta-analysis	Period: 1970–September 2008 # Databases: 2 # Studies: 50 (28 = 60 + years) # e-Interventions: 10 4 incl)^33,34,36,44 # Excluded: 6 (<2000 3, participants 3) R-AMSTAR: 28 No duplicate extraction QA: Not stated	n = 4 # Subjects: 22–93 Age range: 66–93. Sex: Mixed Countries; Israel: 1, Netherlands: 1, USA: 2 Settings: Community; residential care	Interpersonal communication Computer/internet use/training: 3 Internet-operated therapeutic software Robotics 1	Comment on one group vs two group comparisons Scale DJGLS: 1 UCLALS: 3

DJGLS: de Jong Gierveld Loneliness Scale; EPHPP: Effective Public Health Practice Project tool; ND: not documented; NS: not significant; QA: quality assessment; UCLALS: University California Los Angeles loneliness scale; VC: videoconferencing.

Primary studies in reviews

The 12 reviews included 323 primary studies of interventions for social isolation/loneliness of which 174 were e-Intervention studies. Using the CEBM criteria of quality (I–III), 72 studies in the 12 reviews were selected for inclusion in this review, resulting in 22 unique e-Interventions primary studies.

More than half (12) of the 22 e-Interventions were conducted in the USA, followed by four in the Netherlands. The studies included cognitively intact older adults living in the community residential settings, though 12 studies were conducted in nursing homes. The most common e-Intervention reported was ‘online activities’ such as computer/Internet training and usage (n = 15). In terms of study design, six of the primary studies were RCTs, eight were QE studies and eight were cross-sectional studies. Outcomes for loneliness were measured using different versions of the University California Los Angeles Loneliness Scale (UCLALS) with the European studies tending to select versions of the de Jong Gierveld Loneliness Scale (DJGLS). Even though critiques have been levelled at the UCLALS for yielding greater effect sizes¹⁴ and being insensitive to change,²¹ it was the most frequently used measure for loneliness.

Impact of studies on social isolation

Two reviews provided a homogenous meta-analysis^20,30 on online activities, i.e. computer/Internet training in older people. The 2012 meta-analysis,³⁰ five studies,³³^–³⁷ reported a significant pooled decrease in loneliness (ES, Z = 2.085, p = 0.037, n = 5) (LoE I). In contrast, the 2014 review²⁰ pooled seven studies (the five from the previous review and two additional studies on Internet-supported communication – VC^38,39 and concluded a non-significant decrease in loneliness (Z = 0.44; p = 0.37). Though the five studies included in the two reviews were identical,³³^–³⁷ the reported pooled meta-analyses diverged. Due to the heterogeneity of the reviews and the diverging information from two of the meta-analyses, the findings from the relevant primary studies in the reviews are reported in Table 3.

Table 3.

Social isolation/loneliness outcomes for e-Intervention studies (n = 22).

Type of e-Intervention	Outcomes	Quality
Online activities: training and use		LoE I
Computer/Internet training and use^20,a	↔ Loneliness (UCLALS 3) (DJGLS 3) (ES z = 0.44, p = 0.037, n = 6) NS	SR of 3 RCTs and 3 QE
Computer/Internet training and use³⁰^,b	↓ Loneliness (UCLALS 2) (DJGLS 3) (ES z = 2.1, p = 0.037, n = 5 (n = 373)	SR of 3 RCTs and 2 QE
Online activities: training and use		LoE II
Training 9 h 6 × 11 wks Computers and Internet, Community (m = 71 years)^36,a^,b	↔ Loneliness (d = −1, p = 0.52) (UCLALS). (ES z = 2.1 p = 0.04)	RCT (I: 48\|C: 45)
ICT training (11 classes 2 wkly), rural (m = 71.9 years)^37,a^,b	↔ Loneliness (DJGLS). (ES z = 0.27 p = 0.826)	RCT (I: 45\|C: 38)
Training (4 h 2 wks); healthy^35,a^,b	↔ Loneliness (DJGLS). (ES z = 0.6 p = 0.570)	RCT (I: 62\|C: 61\|C: 56\|C: 41)
Online activities: training and use		LoE III
Training email, web, forums and virtual communities (15 wk 1–2 pw) daycare centres/nursing homes (m = 80.2 years)^34,a^,b	↓ Loneliness (UCLALS) (F(1,33) = 34.71; ES = 0.51 p < 0.001).	QE (I: 22\|C: 26) Small sample (LoE III-1)
Training (3 wks) Community (72.8 years)⁴¹	↓ Loneliness (Self-report) use of Skype and 5 emails pw) ↑ Increased networks (p = 0.025)	QE (I: 28\|C: 17) Small sample (LoE III-1)
Training (5 2 h lessons 15 wks). Community (m = 67 years)^33,a^,b	↓ Loneliness 2 years (p = 0.050) and 3 years (p = 0.02) ↓ Loneliness well-educated elderly 2 years (p = 0.04) and 3 years (p = 0.008) (ES z = 1.5 p = 0.144)	QE (I: 12\|C: 14) Small sample (LoE III-1)
Training (8 wks) and Internet to communicate FU 12 mths in AIC units (m = 84 years)⁴²	↓ Loneliness (UCLALS) ↓ isolation (β = 0.491, p < 0.001) 1 point ↑ online activity ∼ 0.236 ↓ loneliness (F = 4.67 p ≤ 0.001)	Xsection (n = 70) Small sample (LoE III-2)
Online activities: use		LoE III
Social network use. Community (m = 66.9 years)⁴⁰	↔ Loneliness (DJGLS) High use SNS (β = 0.111, p = 0.271) Low use SNS (β = 0.008, p = 0.956)	Comparison survey (High use 352: medium use 85) (LoE III-3)
Facebook use. Community (75 years)⁴⁵	↔ Loneliness	Comparison survey (FB: 352\|C: 85) (LoE III-3)
Computer use. Community (65+ years)⁵²	↔ Loneliness (DJGLS) (β = 0.100, NS) 65-75 years (β = 0.169, p < 0.05) 75+ years	Comparison survey (I: 637\|C: 210) (LoE III-3)
Internet use. Community.⁴⁷	↔ Communication decrease loneliness (UCLALS)	Comparison survey (n = 847) (LoE III-3)
Internet and email. Community.⁴³	↑ Networks (p < 0.001); no ES	Comparison survey (n = 809) (LoE III-3)
Internet use. Nursing home and hospital elderly.⁵¹	↓ Loneliness (t = −9.36, p < 0.001) (UCLALS)	Comparison survey (I: 185\|C: 165) (LoE III-3)
Internet use. Retirement housing.⁴⁸	↓ Loneliness (p < 0.001) (DJGLS)	Comparison survey (I: 462\|C: 199) (LoE III-3)
Internet use.⁵⁴	Social support (β = 0.096, t = 6.051, p < 0.001) ∼ ↓ loneliness (β = −0.139, t = −8.034, p < 0.001) (Hughes Loneliness Scale)	Comparison survey (n = 5203) (LoE III-3)
Interpersonal communication		Loe III
ICT VC (5 min 3 mths). Nursing homes (76 years)^39,a	↓ Loneliness (UCLALS) 1 week (β = −1.21, p = 0.002) 3 months (β = −2.84, p = 0.003)	QE (I: 21\|C: 28) (LoE III-1)
ICT VC (5 min 3 mths). Nursing homes (76 years)^38,a	↓ Loneliness (UCLALS) 3 months (β = −5.40, p < 0.001), 6 months (β = −6.47, p < 0.001), 12 months (β = −6.27, p = 0.001)	QE (I: 74\|C: 79) (LoE III-3)
Internet-operated therapeutic software		LoE III
Wii (1 h 10 wks) vs TV. Residential (82 years)⁵³	↓ Loneliness (UCLALS) (F(2,30) = 6.24 p < 0.005)	QE (I: 16\|C12\|C7) Small sample Selection bias (LoE III-1)
Relational agent exercise advisor (2 mths). Community. (74 years)⁴⁶	↔Loneliness (UCLALS) Control decrease (t(7) = 2.74, p < 0.05) Relational agent not (t(9) = 0.50, p = 0.63)	RCT (I: 11\|C: 10) Small sample Selection bias (LoE III-1)
Internet-operated therapeutic software (Robotics)
Robotics pet AIBO vs living dog (30 min 8 wks) Care facility elderly⁴⁴	↓ Loneliness (UCLALS) (F(2,35 = 37.3, p < 0.05)	QE (I: 13\|C: 12) Small sample Selection bias (LoE III-1)
Robotics pet AIBO (7 wks) Nursing home elderly⁴⁹	↓ Loneliness (3.3–1.1) (AOK) (p = 0.07)	QE (I: 6\|C: 6) Small sample Selection bias (LoE III-2)
Conversational agent (1 wk) Alone (>55 years)⁵⁰	↔ Loneliness decrease (UCLALS) (t(12) = −1.67, p = 0.13) and correlation with time (r = 0.7, p<0.05)	QE (Passive: 7\|Active: 7) Small sample Selection bias (LoE III-2)

↓

: significant decreased; ↑: significant increased; ↔: no significance; C: control; DJGLS: de Jong Gierveld Loneliness Scale; ES: effect size; I: intervention; LoE: level of evidence; mth: month; QE: quasi-experimental; pw: per week; RCT: randomized controlled trial; SR: systematic review; UCLALS: University California Los Angeles Loneliness Scale; wk: week.

Borneman SR; ^bChoi SR.

Internet/computer training

Seven studies evaluated computer/Internet training, three RCTs and four QE studies. The three RCTs³⁵^–³⁷ reported no significant decrease in loneliness measured by both the UCLALS and DJGLS (II). Some evidence of decreased loneliness was reported by the QE studies, including a sustained effect for up to three years. However, these studies were generally small and prone to bias (III-2). Use of computer/Internet was primarily assessed through cross-sectional studies (III-3), and evidence for associations for reduced loneliness was found and correlated to frequency of use (III-2).

Interpersonal communication

The two QE videoconferencing studies,^38,39 reported evidence of effectiveness in reducing loneliness (III-2).

Internet-operated therapeutics

Modern technologies such as robotic pets and video games showed mixed evidence of decreasing loneliness, but the QE studies were small and prone to selection bias (III-3) (Table 3). No mHealth studies were found.

Discussion

In the context of reported increases of loneliness in the growing older population,²^–⁴ the scope and rigor of studies conducted in this field is not clear. This systematic review of reviews confirms the heterogeneity of the publications in this field²⁰ evidenced by varying definitions of older people, social isolation/loneliness and e-Interventions/ICTs/Internet-supported interventions. Some consistency existed in the measurements of loneliness, with most of the studies using either the DJGLS or the UCLALS. The review supported the findings from Bornemann,²⁰ showing consistency of sensitivity to loneliness across instruments. However, only four reviews were found with a specific focus on e-Interventions for loneliness in older people.^20,23,28,30 The reviews were of moderate quality, with no reviews meeting all the R-AMSTAR criteria. The reviews also varied in scope, either with a specific focus on the effectiveness of a targeted e-Intervention such as computer training,^20,30 or overviews of the published studies in the field.^27,28 The criteria for inclusion of primary studies and quality assessment of these studies varied greatly with different tools used for rating rigor and bias. These variations would impact on the confidence in the conclusions of these reviews.

To assess the evidence of effectiveness of e-Interventions in this varied field, the interventions were classified into online activities, interpersonal, Internet-supported communications and Internet-supported therapeutics such as robotics, games and relational agents. The following conclusions emerged from the review.

Training and use of Internet/computer e-Interventions were not supported with conclusive evidence on the impact on loneliness. In examining the underlying studies, there appeared to be no definitive evidence of the effectiveness and sustainability of effectiveness. Level III-2 (cross-sectional studies) showed evidence of a correlation between use and loneliness, but the nature of the studies and the lack of an active intervention design, suggests further examination. Level III-2 studies, (QE two group evaluation studies), provided evidence of decreasing loneliness, but the low quality small-study designs suggest caution for future implementation. The level II (RCTs) studies showed no significant effect on decreasing loneliness and the two meta-analyses conducted in 2012 and 2014, including the same five core studies, with divergent pooled findings related to possible inclusion of different primary data in the meta-analyses.²⁰

The evidence for Internet-supported communication showed a significant reduction in loneliness, though this was mediated by self-efficacy²⁴ and frequency of use.⁴⁰ As e-Interventions may alleviate older people’s social isolation through different mechanisms, such as connecting to the outside world, gaining social support, and boosting self-confidence,²³ it would be important for studies to capture information on all these processes.

The emergence of the field of robotics showed some evidence of potential for decreasing loneliness, though the studies were small and biased. It would be important to explore the commonality between videogames, relational agents and robotics that allow for a decrease in social isolation/loneliness, as opposed to computer/Internet training and use, though no evidence of this was found.

This review confirmed the lack of high quality evidence in the field. In addition, despite the existence of guidelines to encourage the standardisation of systematic review reporting, the failure of researchers to adhere to such guidelines weakened the strength of the systematic reviews. The underlying studies lacked rigor in study design to address causation, and the heterogeneity led to challenges in generalizability of the primary research studies. Lastly, there is a need for research in mHealth interventions for social isolation as the grassroots uptake of mobile phones⁵⁵ may provide opportunities for increased uptake of e-Interventions to address loneliness in the older person.

Conclusion

In conclusion, due to the limitations of the reviewed primary studies and the reviews, insufficient quality evidence for the reduction of loneliness seems to exist to allow recommendations for e-Interventions for older people (Grade B).

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

References

United Nations Department of Economic and Social Affairs Population Division. World population ageing 2015, New York: United Nations, 2015.

Nyqvist

Cattan

Andersson

et al.

Social capital and loneliness among the very old living at home and in institutional settings: A comparative study. J Aging Health 2013; 25: 1013–1035.

Drageset

Kirkevold

Espehaug

. Loneliness and social support among nursing home residents without cognitive impairment: A questionnaire survey. Int J Nurs Stud 2011; 48: 611–619.

Chipps

Jarvis

. Social capital and mental well-being of older people residing in a residential care facility in Durban, South Africa. Aging Ment Health 2016; 20: 1264–1270.

Cattan

Kime

Bagnall

. The use of telephone befriending in low level support for socially isolated older people–an evaluation. Health Soc Care Community 2011; 19: 198–206.

Barak

Klein

Proudfoot

. Defining internet-supported therapeutic interventions. Ann Behav Med 2009; 38: 4–17.

Silva

Rodrigues

de la Torre Díez

et al.

Mobile-health: A review of current state in 2015. J Biomed Inform 2015; 56: 265–272.

Kwan A. mHealth solutions for improving mental health and illnesses in the aging process. White Paper Series on mHealth and Aging, mHealth Alliance, UN Foundation. 2013. Available at: www.mhealthknowledge.org/sites/default/files/7_mHA-Aging-Paper3_092713.pdf.

Chipps

Jarvis

. Technology-assisted communication in older persons in a residential care facility in South Africa. Information Development 2016. DOI: 10.1177/0266666916664388.

10.

Broady

Chan

Caputi

. Comparison of older and younger adults’ attitudes towards and abilities with computers: Implications for training and learning. Br J Educ Technol 2010; 41: 473–485.

11.

van Biljon

Renaud

. A qualitative study of the applicability of technology acceptance models to senior mobile phone users. Advances in conceptual modeling–Challenges and opportunities 2008, pp. 228–237.

12.

Cornwell

Waite

. Social disconnectedness, perceived isolation, and health among older adults. J Health Soc Behav 2009; 50: 31–48.

13.

Barak

Hen

Boniel-Nissim

et al.

A comprehensive review and a meta-analysis of the effectiveness of internet-based psychotherapeutic interventions. J Technol Hum Serv 2008; 26: 109–160.

14.

Masi

Chen

H-Y

Hawkley

et al.

A meta-analysis of interventions to reduce loneliness. Pers Soc Psychol Rev 2011; 15: 219–266.

15.

Kung

Chiappelli

Cajulis

et al.

From systematic reviews to clinical recommendations for evidence-based health care: Validation of revised assessment of multiple systematic reviews (R-AMSTAR) for grading of clinical relevance. Open Dent J 2010; 4: 84–84.

16.

Chipps

Brysiewicz

Mars

. Effectiveness and feasibility of telepsychiatry in resource constrained environments? A systematic review of the evidence. Afr J Psychiatry (Johannesbg) 2012; 15: 235–243.

17.

Centre for Evidence Based Medicine (CEBM) Oxford Centre for Evidence-based Medicine – levels of evidence (March 2009), http://www.cebm.net/oxford-centre-evidence-based-medicine-levels-evidence-march-2009/. (2009) Accessed 15 August 2017.

18.

Edinger

Cohen

. A large-scale analysis of the reasons given for excluding articles that are retrieved by literature search during systematic review. AMIA Annu Symp Proc 2013; 2013: 379–387.

19.

Möller

H-J

. Effectiveness studies: Advantages and disadvantages. Dialogues Clin Neurosci 2011; 13: 199–207.

20.

Bornemann R. The impact of information and communication technology (ICT) usage on social isolation including loneliness in older adults. A systematic review. Magdeburg/Stendal University of Applied Sciences, 2014 M.Sc Rehabilitation Psychology; Magdeburg /Stendal University of Applied Sciences. https://www.researchgate.net/publication/316740931_The_impact_of_information_and_communication_technology_ICT_usage_on_social_isolation_including_loneliness_in_older_adults_A_systematic_review.

21.

Hagan

Manktelow

Taylor

et al.

Reducing loneliness amongst older people: A systematic search and narrative review. Aging Ment Health 2014; 18: 683–693.

22.

Cattan

White

Bond

et al.

Preventing social isolation and loneliness among older people: A systematic review of health promotion interventions. Ageing Soc 2005; 25: 41–67.

23.

Chen

Y-RR

Schulz

. The effect of information communication technology interventions on reducing social isolation in the elderly: A systematic review. J Med Internet Res 2016; 18: e-18–e-18.

24.

Cohen-Mansfield

Perach

. Interventions for alleviating loneliness among older persons: A critical review. Am J Health Promot 2015; 29: e109–e125.

25.

Dickens

Richards

Greaves

et al.

Interventions targeting social isolation in older people: A systematic review. BMC Public Health 2011; 11: 647–647.

26.

Franck

Molyneux

Parkinson

. Systematic review of interventions addressing social isolation and depression in aged care clients. Qual Life Res 2016; 25: 1395–1407.

27.

Khosravi

Ghapanchi

. Investigating the effectiveness of technologies applied to assist seniors: A systematic literature review. Int J Med Inform 2016; 85: 17–26.

28.

Khosravi

Rezvani

Wiewiora

. The impact of technology on older adults’ social isolation. Comput Human Behav 2016; 63: 594–603.

29.

Morris

Adair

Ozanne

et al.

Smart technologies to enhance social connectedness in older people who live at home. Australas J Ageing 2014; 33: 142–152.

30.

Choi

Kong

Jung

. Computer and internet interventions for loneliness and depression in older adults: A meta-analysis. Healthc Inform Res 2012; 18: 191–198.

31.

Gardiner

Geldenhuys

Gott

. Interventions to reduce social isolation and loneliness among older people: An integrative review. Health Soc Care Community 2016, pp. 1–11. DOI: 10.1111/hsc.12367.

32.

Medical Advisory Secretariat. Social isolation in community-dwelling seniors: An evidence-based analysis. Ont Health Technol Assess Ser 2008; 8: 1–49.

33.

Fokkema

Knipscheer

Ozanne

et al.

Escape loneliness by going digital: A quantitative and qualitative evaluation of a Dutch experiment in using ECT to overcome loneliness among older adults. Aging & Mental Health 2007; 11(5): 496–504. DOI: 10.1080/13607860701366129.

34.

Shapira

Barak

Gal

et al.

Promoting older adults’ well-being through Internet training and use. Aging & Mental Health 2007; 11(5): 477–484. DOI:10.1080/13607860601086546.

35.

Slegers

Van Boxtel

Jolles

. Effects of computer training and Internet usage on the well-being and quality of life of older adults: A randomized, controlled study. J Gerontol B Psychol Sci Soc Sci 2008; 63: P176–P184.

36.

White

McConnell

Clipp

et al.

A randomized controlled trial of the psychosocial impact of providing Internet training and access to older adults. Aging Ment Health 2002; 6: 213–221.

37.

Woodward

Freddolino

Blaschke-Thompson

et al.

Technology and aging project: Training outcomes and efficacy from a randomized field trial. Ageing Int 2011; 36: 46–65.

38.

Tsai

H-H

Tsai

Y-F

. Changes in depressive symptoms, social support, and loneliness over 1 year after a minimum 3-month videoconference program for older nursing home residents. J Med Internet Res 2011; 13: e93–e93.

39.

Tsai

H-H

Tsai

Y-F

Wang

H-H

et al.

Videoconference program enhances social support, loneliness, and depressive status of elderly nursing home residents. Aging Ment Health 2010; 14: 947–954.

40.

Aarts

Peek

Wouters

. The relation between social network site usage and loneliness and mental health in community-dwelling older adults. Int J Geriatr Psychiatry 2015; 30: 942–949.

41.

Blažun

Saranto

Rissanen

. Impact of computer training courses on reduction of loneliness of older people in Finland and Slovenia. Comput Human Behav 2012; 28: 1202–1212.

42.

Cotten

Anderson

McCullough

. Impact of Internet use on loneliness and contact with others among older adults: Cross-sectional analysis. J Med Internet Res 2013; 15(2): e39–e39.

43.

Nahm

E-S

Resnick

Mills

. A model of computer-mediated social support among older adults. AMIA Annu Symp Proc 2003, pp. 948–948.

44.

Banks

Willoughby

Banks

. Animal-assisted therapy and loneliness in nursing homes: Use of robotic versus living dogs. J Am Med Dir Assoc 2008; 9: 173–177.

45.

Bell C, Fausset C, Farmer S, et al. Examining social media use among older adults. In: Proceedings of the 24th ACM conference on hypertext and social media. ACM, 2013, 1-3 May, Paris, France, ACM New York, NY, USA. 158-163. doi:10.1145/2481492.24815092013.

46.

Bickmore

Caruso

Clough-Gorr

et al.

‘It’s just like you talk to a friend’ relational agents for older adults. Interact Comput 2005; 17: 711–735.

47.

Erickson

Johnson

. Internet use and psychological wellness during late adulthood. Can J Aging 2011; 30: 197–209.

48.

James

Boyle

et al.

Internet use and decision making in community-based older adults. Front Psychol 2013; 4: 1–10.

49.

Kanamori M, Suzuki M, Oshiro H, et al. Pilot study on improvement of quality of life among elderly using a pet-type robot. In: Computational Intelligence in Robotics and Automation, 2003 Proceedings. IEEE, 2003, 16-20 July, Kobe, Japan. DOI: 10.1109/CIRA.2003.1222072.

50.

Ring L, Barry B, Totzke K, et al. Addressing loneliness and isolation in older adults: Proactive affective agents provide better support. In: International Conference on Affective Computing and Intelligent Interaction (ACII). IEEE, 2013, 2-5 September, Geneva, Switzerland, pp. 61-66. DOI: 10.1109/ACII.2013.17.

51.

Şar

Göktürk

Tura

et al.

Is the Internet use an effective method to cope with elderly loneliness and decrease loneliness symptom?

Procedia Soc Behav Sci 2012; 55: 1053–1059.

52.

Toepoel

. Ageing, leisure, and social connectedness: How could leisure help reduce social isolation of older people? Soc Indic Res 2013; 113: 355–372.

53.

Kahlbaugh

Sperandio

Carlson

et al.

Effects of playing Wii on well-being in the elderly: Physical activity, loneliness, and mood. Act Adapt Aging 2011; 35: 331–344.

54.

Heo

Chun

Lee

et al.

Internet use and well-being in older adults. Cyberpsychol Behav Soc Netw 2015; 18: 268–272.

55.

Pimmer

Brysiewicz

Linxen

et al.

Informal mobile learning in nurse education and practice in remote areas–a case study from rural South Africa. Nurse Educ Today 2014; 34: 1398–1404.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.21 MB