key: cord-0260706-94zqk6lu authors: Shah, S. G. S.; Nogueras, D.; van Woerden, H. C.; Kiparoglou, V. title: Are digital technology interventions effective to reduce loneliness in older adults? A systematic review and meta-analysis date: 2020-09-01 journal: nan DOI: 10.1101/2020.08.27.20183012 sha: 3daf9687792f07fd36f54d8045e53089a3a77a26 doc_id: 260706 cord_uid: 94zqk6lu Objective: To review the latest literature on the effectiveness of DTIs in reducing loneliness in (older) adults. Data Sources: Electronic searches in PubMed, Medline, CINAHL, EMBASE and Web of Science covering publication period from 1 January 2010 to 31 July 2019. Subjects: Adult men and women Design: Systematic review and meta-analysis Main Outcome Measure: Loneliness. Study Selection: Primary studies that used DTIs for tackling loneliness in adults (aged [≥]18 years) with follow-up measurements at least three months or more and publication in the English language. Data Extraction and Synthesis: Two researchers independently screened articles and extracted data on several variables: participants, interventions, comparators and outcomes. Data was extracted on the primary outcome i.e. loneliness measured at the baseline and follow-up measurements at three, four, six and twelve months after the intervention. Results: Six studies were selected from 4939 articles screened. Selected studies included 5 clinical trials (4 RCTs and 1 quasi experimental study) and one before and after study, which enrolled 646 participants (men =154 (24%), women =427 (66%), no gender information =65 (10%) with average age between 73 and 78 years (SD 6-11). Five clinical trials were included in the meta-analysis and standardised mean differences (SMD) were calculated for each trial and pooled across studies using a random effects model. The overall effect estimates were not statistically significant in follow-up measurements at three months (SMD= 0.02, 95% CI= -0.36, 0.40; P=0.92), four months (SMDs= -1.11, 95% CI= -2.60, 0.38; P=0.14) and six months (SMD= -0.11, 95% CI= -0.54, 0.32; P=0.61). The quality of evidence was very low to moderate in these trials. Conclusions: There is insufficient evidence to make conclusions that DTIs are effective in reducing loneliness in older adults. Future research may consider RCTs with larger sample sizes and longer duration of interventions and follow-up. loneliness and attempting to deal with isolation due to social distancing in the epidemic situations such as the COVID-19 pandemic (3) . Loneliness refers to subjective feelings of an individual because of a perceived discrepancy between actual and the desired social relationships (13, 14) . While loneliness affects people of all ages, (14, 15) older and vulnerable people are affected more (10, 16, 17) . It is associated with the social and physical environment (16) (17) (18) . Loneliness enhances the risk of poor mental and physical health (13, (19) (20) (21) (22) , dementia (23) , premature mortality and allcause mortality (20) particularly in older adults (22) . Loneliness could be addressed through a range of social (24) and technological interventions (25) such as digital applications (apps), online social networks, and social robots (26) . However, there is limited evidence on the effectiveness of technological interventions for loneliness (27) . While several previously published reviews report that these interventions are effective in reducing loneliness (28) (29) (30) (31) (32) , some of these studies are weak and have a high risk of bias (33) . Some use very selected technological interventions and cover literature published over a very short time period i.e. between January 2010 and January 2013 (29) . These findings suggest the need for further research (30, 31) to assess and identify the latest digital technological interventions that are effective in reducing loneliness (27, 32) . We therefore appraise the latest evidence to assess the effectiveness of digital technology interventions to reduce loneliness, which is imperative not only from the perspectives of patients and their families but also from the perspectives of other stakeholders such as public health, health and social care providers, and health insurers (34). Our main objective was to assess the effectiveness of digital technology interventions (DTIs) to reduce loneliness in adults. The secondary objective was to identify DTIs that are used to reduce loneliness in adults. We registered our systematic review and meta-analysis with the PROSPERO database on 10 th June 2019 (Registration ID CRD42019131524) (35) and we published our protocol (34). We defined a digital technology intervention as an intervention that applies digital technology i.e. the technology, equipment and applications that process information in the form of numeric codes, usually a binary code (36). We selected studies that met our predefined eligibility Criteria (34) . These included interventional studies (randomised and non-randomised) on loneliness using DTIs. The DTIs included the use of computers, computer tablets, iPads, Internet, online videos, communication, chatting, social groups, meetings, conferences and messages, sensors, (social) robots, (smart) phones, social media tools and the World Wide Web. A minimum Intervention duration and follow-up period was set at three months. Participants included both male and female adults (aged 18 years and more) living in different settings (i.e. residential dwellings including private residences and care / nursing homes in any country. The studies were limited to journal articles in the English language published from 1 January 2010 to 31 July 2019. We did electronic searches in PubMed, Medline, EMBASE, CINAHL and Web of Science covering the publication period from 1 January 2010 to 31 July 2019. We used an a priori list of keywords prepared in our preliminary literature searches (34) . The keywords were of two categories: medical condition / problem (i.e. loneliness, lonely, isolation, aloneness, disconnect*, solitude, singleness*, lonesomeness, solitariness, and remoteness).and intervention / technology (i.e. digital, technolog*, sensor*, robot*, internet, social media, *phone*, online, iPad*, tablet*, computer*, electronic, web, video, and videoconference), as reported in our protocol (34). First, we searched the keywords in the 'subject headings' such as Medical Subject Headings (MeSH) major terms in the PubMed or equivalent in other databases (Appendix-1). Thereafter, we searched for keywords in the 'title' and 'abstract' fields in the selected databases using three Boolean operators: 'OR', 'AND' and 'NOT'. In addition, we hand searched through reference lists of shortlisted articles. We wrote emails to the authors of two studies for full copies of their research articles (37, 38) , which were gratefully provided to us. We contacted the authors of two further studies for missing / additional data (39, 40) . We had good response from the authors of both studies and data were thankfully provided for one study (40) . We sought support from an expert librarian for running literature searches. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020. . https://doi.org/10.1101/2020.08.27.20183012 doi: medRxiv preprint Searches retrieved 4939 articles that included 965 duplicate articles, which were removed (Fig. 1) . Two researchers (SGSS and DN) independently screened titles of 3974 articles and read abstracts of 442 articles, which led to exclusion of 3876 articles.. Full text of the remaining 98 articles was read independently by three researchers (SGSS, DN and VK). When recommendations differed between reviewers at the title, abstract and full text review stages, another reviewer (HCvW) reviewed these articles and his recommendations to either include or exclude an article were final. Finally, we excluded 92 articles and included six articles in the data-extraction and narrative synthesis, and included five studies in the metaanalysis (Fig. 1) . is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020. . https://doi.org/10.1101/2020.08.27.20183012 doi: medRxiv preprint For data collection, we used a priori data extraction template (Tables 1-2) , which comprised a number of columns: Author(s), year and country of study; study aim/objectives; research design; settings; participants' characteristics (age, gender and ethnicity); health/medical condition; sampling method and size; participant attrition (numbers / %), research method(s) / data collection tool(s); intervention(s) (e.g. type/tool of digital technology), comparator(s).(e.g. alternative intervention, placebo or care as usual), intervention duration (weeks / months),.measurement stages (e.g. baseline and follow-up -weeks/months after the baseline), outcomes/results/findings (e.g. loneliness scores (including statistics e.g. mean values, standard deviations, standard errors and confidence intervals) and authors' conclusion(s) (34) . SGSS and DN independently extracted data from all included studies (n=6) using the data extraction template (Tables 1-2) and resolved discrepancies in extracted data with discussion. We compared and contrasted data extraction forms; thus, we avoided bias and reduced errors in the data extraction process (41) . We extracted aggregated data at the study level as much as possible with respect to the intervention, which is imperative for the reproducibility of effective interventions (42, 43) . Following suggestions for reporting data once from studies with duplicate and multiple publications, (41) we extracted, and report, data only once (44) from a research study with multiple publications (44, 45). We report both a narrative synthesis (narrative summary) and a statistical (quantitative) synthesis (meta-analysis) of our review study as suggested for reporting of a systematic review on the effectiveness (46) . In the narrative synthesis, we include all six studies and report their characteristics including the study design, settings, sample sizes, data collection methods, participants, intervention(s), comparator(s), outcome measurements and study conclusion(s).. In the meta-analysis, we included five studies and pooled extracted data on loneliness measured by continuous loneliness scales. In meta-analysis, the standardised mean difference (SMD) as a summary statistic for reporting continuous data has been suggested for studies that assess the same outcome but use different scales for measuring the outcome (47) . In RevMan, the SMD is the effect size known as Hedges' (adjusted) g, which is akin to Cohen's d and it includes an adjustment for small sample size bias (47) . In our review, the main outcome i.e. loneliness, was measured using different loneliness scales which included the UCLA loneliness scale (score range: 20-80) (48, 49) and the de Jong Gierveld loneliness scale (score range.0-11) (50). We calculated the standardised mean values (SMDs) from extracted data i.e. loneliness mean scores with standard deviations (SD) and sample sizes in the intervention and control groups during the follow-up measurements at three, four and six months after the intervention. We conducted meta-analysis in the Review Manager (RevMan), version 5.3.5 software (51) . In the meta-analysis, we ran the random effects model as the statistical model because we hypothesised that the true effect sizes between studies would vary (52, 53) due to differences in the methodological and clinical characteristics between studies (54) such as . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020. . https://doi.org/10.1101/2020.08.27.20183012 doi: medRxiv preprint differences in the sample sizes, participants, interventions types and duration, and the follow-up measurement times. The Cochrane guidelines (47) suggest that different study designs should not be combined in the meta-analysis because it can increase heterogeneity and studies with repeated measurements at different follow-up periods cannot be combined without a unit-of-analysis error (47) . We therefore conducted meta-analyses for similar study designs i.e. clinical trials but separated them based on the follow-up measurement periods. Thus, we undertook a separate meta-analysis for each follow-up measurement at three, four and six months after the intervention. In addition, we ran meta-analyses when there were at least two or more studies for the same outcome or the same follow-up period (55) . We therefore did not run a meta-analysis for follow-up measurements at 12 months reported in two studies because they involved different study designs i.e. an RCT with intervention and control groups (40) and a pre and post intervention study with only intervention group(56) to avoid increase in the heterogeneity (47) . Thus, we also avoided the overestimation of the effect of intervention in the absence of a control group (57) in the pre and post intervention study (56). In addition, we did not conduct sensitivity analyses due to the small number of studies in meta-analyses at each follow-up point (55) . We had only one pre-post study and we did not run metaanalysis for this study because meta-analysis cannot be run with only one study (47). We assessed the quality of research (Table 3 ) by applying the GRADE (Grading of Recommendations Assessment, Development and Evaluation) approach (58). We assessed risk of bias focusing on five domains: evaluation of sequence generation, allocation concealment, blinding (outcome assessors), incomplete data, selective outcome reporting, and assessing other biases using the Cochrane method (47) . In a meta-analysis, publication bias can be assessed by graphical method using Funnel plots (47, 59) and statistical methods such as the Egger's test (47) ; however, the both methods require at least 10 studies in the meta-analysis (47) . When the number of studies is less, the Egger's test has low power and it fails to differentiate chance from real asymmetry. (47) . Similarly, assessing publication bias by Funnel plots with fewer studies would be of very limited usefulness because it would be difficult to spot the publication bias. We had three studies maximum in a meta-analysis (Fig. 2) ; hence, we could not check the publication bias by either method. We checked heterogeneity i.e. variation in study outcomes / effect sizes between studies by the Cochran's Q ( 2 ) Test with a significance level of ρ < 0.05 (25, 60) . We used I 2 statistics for determining the magnitude of heterogeneity (i.e. the proportion of variance in the true effect sizes) between studies (25) . We considered I 2 of ≤25%, between 25% and 50% and >50% as low (55) , moderate (25, 61) and high heterogeneity between studies, respectively (19, 61). We report the findings of our meta-analyses using the SMDs with 95% confidence intervals (CI) as a statistical summary with the forest plots (47) . is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020. . https://doi.org/10.1101/2020.08.27.20183012 doi: medRxiv preprint Participants, intervention(s), comparator(s), outcome measurements and study conclusion(s) are presented in Table 1 . Characteristics of studies including the study designs, settings, sample sizes, data collection methods are given in Table 2 . The database search of PubMed, Medline, CINAHL, EMBASE, and Web of Science generated a total of 4939 papers ( Fig. 1 ) of which six studies met the eligibility Criteria. All six studies are included in the narrative synthesis and five studies are included in the metaanalysis. The total number of participants enrolled in all six included studies was 646 (mean = 108, SD = 102, median = 77, IQR= 32 to 130). Studies varied in total sample sizes (mean 108, SD 102, range 30-300) and the sample sizes of intervention groups and control groups also varied at both the baseline and the follow-up measurements ( Table 1 ). The attrition rate also varied between studies (range 7-35%, mean 19%, SD 10%). Participants' average age was between 73 and 78 years (SD 6-11). Total enrolled participants included 427 women (66%) and 154 men (24%) while no information about the gender was available for 65 (10%) participants. Studies varied by gender of participants i.e. female (mean= 66%, SD =16%, range = 46-81%) and male (mean =25%, SD =9%, range = 19-42%). Only two studies reported on participants' ethnicity, which included white (54%) and non-whites (46%) in the US study (40) and mostly Asian Indians (no numbers reported) in the South African study (62). The characteristics of included studies (n=6) are given in Tables 2. Of six studies included, four studies were randomised clinical trials (RCTs) (40, 44, 62, 63), one non-randomised clinical trial (NRCT) (64) and one pre-and post-test (before and after) study with intervention group only (no control group) (56) ( Table 1) . Four studies were conducted in developed countries i.e. the Netherlands (56), UK (63), US Czaja et al., 2018) and Sweden (44) . Two studies were undertaken in developing countries i.e. Taiwan (64) and South Africa (62) . The settings included living in independent housing in the community (40), living in ordinary housing without any home care services (44) , receiving care in their own home/supported housing in the community ("domiciliary care"), or residential care in care homes (63), residential care facilities for older people (62), nursing home (64) and elderly home care (56). Participants were selected by random sampling in four (60%) studies (40, 44, 62, 63) and the other two (20%) studies used purposive (64) and convenience (56) sampling each. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020. . https://doi.org/10.1101/2020.08.27.20183012 doi: medRxiv preprint Digital technology interventions included social internet-based activities (SIBAs) such as Skype and Facebook (44), videoconferencing (64) , customized computer platform with simplified touch-screen interface (63), personal reminder information and social management system (40), WhatsApp group (62) and video/voice network (56). The duration of intervention was three months in four studies (44, (62) (63) (64) and 12 months in two studies (40, 56). The main outcome measure i.e. loneliness was measured at the baseline and multiple follow-up times, which included at three months in three studies (44, 62, 64) , four months in two studies (62, 63) , six months in two studies (40, 44) and 12 months in two studies (40, 56). Loneliness measurement tools were the UCLA loneliness scale (48, 49) used in four studies (40, 44, 63, 64) and the De Jong-Gierveld loneliness questionnaire (50, 65) used in two studies (56, 62). Table 2 presents loneliness scores measured in the intervention and control groups, if any, at the baseline and follow-ups. Narrative synthesis showed that there was reduction in loneliness in the intervention groups at the follow-ups compared to at the baseline ( Table 2) . A statistical summary of the loneliness measurements in the intervention and control groups at the follow-ups is reported in the meta-analysis section. . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020. . We ran three meta-analyses one each for follow-up measurements at three, four and six months involving three, two and two studies respectively. Three studies (44, 62, 64) involving 106 participants with follow-up measurement at three months were entered into a meta-analysis, which showed a very small reduction in loneliness in favour of the control (SMD = 0.02, 95%CI -0.36, 0.40) but it was not statistically significant (Z =0.10, P=0.92). The heterogeneity between studies was not statistically significant (τ 2 =0.00,  2 = 0.10, P=0.95, I 2 =0%) ( Fig. 2 -panel A) . Two studies (62, 63) involving 105 participants with four month follow-up were entered into a meta-analysis, which revealed a large reduction in loneliness in favour of the intervention (SMD = -1.11, 95%CI -2.60, 0.38) but was not statistically significant (Z=1.46; P=0.14).There was statistically significant high heterogeneity between studies (τ 2 =1.03,  2 = 8.84, P=0.003, I 2 =88%) ( Fig. 2 -panel B ). A meta-analysis involving 2 studies (40, 44) with 208 participants with six month follow-up showed a very small reduction in loneliness in favour of the intervention (SMD = -0.11, 95%CI -0.54, 0.32) but it was not statistically significant (Z=0.51, P=0.61). There was moderate heterogeneity between studies but not statistically significant (τ 2 =0.05,  2 = 1.58, P=0.21, I 2 =37%) ( Fig. 2-panel C) . . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020. Risk of bias assessment is presented in Figs. 3-4 . A high risk of bias was noted more in the attrition bias and other bias; unclear risk of bias was detected in blinding of outcome assessment, allocation concealment and blinding of participants and personnel while a low risk of bias was observed especially in the random sequence generation and selective reporting (Fig. 3 ). In addition, most studies reported only within-group changes and not between-group comparisons of change, which may suggest a weak quality of the reporting of results and the analysis in these studies ( Table 2) . is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020. . https://doi.org/10.1101/2020.08.27.20183012 doi: medRxiv preprint We assessed the quality of evidence as moderate, very low and moderate in meta-analyses involving three (44, 62, 64) , two (62, 63) and two studies(40, 44) with follow-up at three, four and six months respectively ( Table 3 -GRADE quality of evidence). . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020. To provide the answer whether digital technology is effective in reducing loneliness in adults, we appraised peer reviewed empirical research involving the application of DTIs in adults with loneliness. Our systematic review includes a narrative summary of six studies, which reported reduction in loneliness in the intervention groups at the follow-up compared to the baseline (Table 2) . However, our meta-analyses of five clinical trials with follow-up measurements at three, four and six months showed no statistically significant pooled effect estimates as SMDsthe preferred method for summarizing effects on continuous outcomes such as loneliness. Although not statistically significant, the summary effect size at four months follow-up was better compared to the effect size at three and six months follow-up (Fig. 2) . Our meta-analysis also revealed that CIs of the summary effects of two studies i.e. (44) . and (64) were very wide and the SMDs from these studies were more in favour of control group rather than the intervention group (Fig. 2) . Thus, the wide widths of CIs of the summary effects in these studies leave uncertainty about the beneficial effect or otherwise of DTI on measures of loneliness. Overall, findings of our meta-analysis provide insufficient evidence to make conclusions that DTIs are effective in reducing loneliness in (older) adults. The quality of evidence of included studies was very low to moderate (Table 3) and there was a high heterogeneity between studies (62, 63). All included studies had a high proportion of female participants. Most notably, the total number of participants was low (44, 62) and the sample sizes were reduced further due to a high attrition rate in some studies (40, 63). The types and methods of DTIs varied between studies, which were conducted in diverse settings and different countries (Tables 1-2) . These factors could have contributed in pooled estimates being not statistically significant. There a few published meta-analyses on technological interventions for tackling loneliness covering literature published up to 2009 (25) and 2011 (28) whereas our review and metaanalysis includes the latest evidence published between 1January 2010 and 31July 2019. We did not replicate the findings of earlier meta-analyses that reported evidence suggesting technological interventions resulting in decreased loneliness (25, 28) . A meta-analysis by (28) reported statistically significant evidence suggesting internet and computers reduce loneliness. However, (28) focused on older adults with depression and included the Internet and computers only as technological interventions whereas we included all types of digital technology interventions and adults of all age groups (≥ 18 years). In addition, studies (n=5) included in the meta-analysis by (28) had different follow-up periods (3-6 months) but they did not report which follow-up measurements they included in their meta-analysis. We ran different meta-analyses for measurements at different follow-up periods, i.e. three, four and six months, as suggested by the Cochrane guidelines (47) . In addition, a meta-analysis by Masi et al., 2011 also reported that technological interventions reduce loneliness, which was more in pre-post studies and non-randomised studies compared to RCTs. However, they included studies with technology and non-. CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020. . https://doi.org/10.1101/2020.08.27.20183012 doi: medRxiv preprint technology based interventions (25) whereas we focused on studies with DTIs only. Masi et al., 2011 did not report how they used measurements at different follow-up periods while we avoided combining follow-up measurements at different times as suggested (47) . Nonetheless, Masi et al concluded that the technology is yet to be capitalised on for loneliness (25) , which is evident from our findings that show that DTIs do not reduce loneliness, especially in adults. Interestingly, our findings provide new insights about loneliness in older adults. Despite our inclusion criteria of age (18 years and above), the selected studies more commonly involved older people (averagely 70+ years old) and there was no statistically significant reduction in loneliness. Our findings might suggest that DTIs do not reduce loneliness in adults, which is opposite to a commonly held view that digital technology can solve the problem of loneliness, especially in older people. However, the digital technologies are tools and means to social connectedness, which may help in lessening loneliness for a little while because the effects of DTIs are short-lived and do not last beyond six months of the intervention (66). In addition, digital technologies might provide digital social connection but in fact they reduce social connectedness in real life (67) . Digital technologies are only tools that extend opportunities for connecting with others but do not provide real human interaction (68) and cannot replace human contact (40). Limitations of our systematic reviews include a small number of studies (n=6) with heterogeneous sets of results, the selection of English language peer reviewed literature published from January 1, 2010 to July 31, 2019 and the intervention duration minimum of three months, which could have resulted in the inclusion of a low number of studies and possible exclusion of potential studies that would have provided useful evidence. In addition, we could not run sub-group and meta-regression analyses due to very limited number of studies and lack of data on loneliness by participants' demographic characteristics. In addition, our review might be narrow because we excluded some studies (39, (69) (70) (71) (72) (73) (74) (75) , which met the technology criterion such as the use of robots, sensors, digital speakers and apps but failed to meet other selection criteria. Thus, our meta-analysis was limited to studies mostly about social interactions using digital tools. Moreover, another limitation could be using a meta-analysis based on only the follow-up data. For example, study by Tsai et al., 2010 in the 3 months follow-up meta-analysis had an SMD of 0.06 with a CI -0.8 to +0.65 (p=0.03) (Fig. 2) , which may suggest that these studies may have had higher power to show a difference compared to baseline loneliness. As recommendations for future research, we suggest that researchers involved in trials may agree on a common measure of loneliness and consider reporting of results in a standardised way that would allow pooling of baseline-adjusted estimates of the treatment effect rather than differences in follow-up means. We did not find sufficient evidence supporting the effectiveness of digital technology intervention to reduce loneliness in (older) adults. Our findings may suggest further research . CC-BY 4.0 International license It is made available under a perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020. . https://doi.org/10.1101/2020.08.27.20183012 doi: medRxiv preprint involving RCTs (44) with larger sample sizes with longer duration of interventions and followup measurement periods. Future research may apply inclusive research designs using a combination of digital applications including robots, sensors and social connecting applications by involving adults in the age group 45-65 years as this segment of the population is more likely to be more technology savvy and digital interventions might be more effective. Future research might also target ethnic, racial, and sexual orientation minority communities where loneliness is common (11) . is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020. . is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020. is the author/funder, who has granted medRxiv a license to display the preprint in (which was not certified by peer review) preprint The copyright holder for this this version posted September 1, 2020 Mindfulness training reduces loneliness and increases social contact in a randomized controlled trial The Mental Health Consequences of COVID-19 and Physical Distancing: The Need for Prevention and Early Intervention Multidisciplinary research priorities for the COVID-19 pandemic: a call for action for mental health science Online first World Health Organization: Mental health and COVID-19 outbreak Loneliness is pervasive and rising, particularly among the young -Daily chart A longitudinal analysis of loneliness among older people in Great Britain Murthy V: Work and the loneliness epidemic Interventions for alleviating loneliness among older persons: a critical review Battling the Modern Behavioral Epidemic of Loneliness: Suggestions for Research and Interventions Loneliness among older adults in Latin America, China, and India: prevalence, correlates and association with mortality Loneliness, life satisfaction and psychological distress among out-of-school adolescents in a Nigerian urban city The Scottish Government: A Connected Scotland: our strategy for tackling social isolation and loneliness and building stronger social connections Qualter P: Loneliness in children and adolescents: What do schools and teachers need to know and how can they help? Pastor Care Educ Correlates and predictors of loneliness in olderadults: a review of quantitative results informed by qualitative insights Risk factors for loneliness in patients with cancer: a systematic literature review and meta-analysis Loneliness and social support of older people in China: a systematic literature review Association of loneliness with all-cause mortality: A meta-analysis The potential public health relevance of social isolation and loneliness: Prevalence, epidemiology, and risk factors Loneliness and social isolation as risk factors for mortality: a meta-analytic review A systematic literature review on the association between loneliness and coping strategies Does loneliness contribute to mild cognitive impairment and dementia? A systematic review and meta-analysis of longitudinal studies A meta-analysis of interventions to reduce loneliness The effectiveness of social robots for older adults: A systematic review and meta-analysis of randomized controlled studies Medical Advisory Secretariat: Social isolation in community-dwelling seniors: an evidence-based analysis Computer and internet interventions for loneliness and depression in older adults: a meta-analysis Smart-home technologies to assist older people to live well at home Smart technologies to enhance social connectedness in older people who live at home Robotics to enable older adults to remain living at home Reducing loneliness amongst older people: a systematic search and narrative review Effectiveness and feasibility of virtual reality and gaming system use at home by older adults for enabling physical activity to improve health-related domains: a systematic review Effectiveness of digital technology interventions to reduce loneliness in adults: a protocol for a systematic review and meta-analysis The effectiveness of digital technology interventions to reduce loneliness in adult people: a systematic review and meta-analysis What are digital technologies? R: A Window Toward the World: Older Adults' Experiences of Becoming in Health and Developing as Human Beings Through Interacting With Others Using Real Video Communication Design of a Mobile Social Community Platform for Older Chinese People in Urban Areas A Smart-Home System to Unobtrusively and Continuously Assess Loneliness in Older Adults Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015: elaboration and explanation JBI's systematic reviews: data extraction and synthesis What kind of systematic review should I conduct? A proposed typology and guidance for systematic reviewers in the medical and health sciences Effects of a social internet-based intervention programme for older adults: An explorative randomised crossover study Participation in social internet-based activities: Five seniors' intervention processes Chapter 3: Systematic reviews of effectiveness [Internet], inAromataris E, Munn Z, editorsJoanna Briggs Institute Reviewer's Manual Cochrane Handbook for Systematic Reviews of Interventions The Cochrane Collaboration UCLA Loneliness Scale (Version 3): Reliability, Validity, and Factor Structure The revised UCLA Loneliness Scale: Concurrent and discriminant validity evidence A 6-item scale for overall, emotional, and social loneliness: Confirmatory tests on survey data The Cochrane Collaboration: Review Manager (RevMan) Version 5.3.5 Introduction to meta-analysis A basic introduction to fixed-effect and randomeffects models for meta-analysis Systematic reviews and meta-analyses of randomized trials: principles and pitfalls Psychological therapies (Internet-delivered) for the management of chronic pain in adults Implementation of CareTV in care for the elderly: the effects on feelings of loneliness and safety and future challenges Experimental and quasi-experimental designs for evaluating guideline implementation strategies Rating up the quality of evidence Meta-analyses: how to read a funnel plot Quantifying heterogeneity in a meta-analysis Measuring inconsistency in meta-analyses An Evaluation of a Low-Intensity Cognitive Behavioral Therapy mHealth-Supported Intervention to Reduce Loneliness in Older People Activating and Guiding the Engagement of Seniors With Online Social Networking: Experimental Findings From the AGES 2.0 Project [Internet Videoconference program enhances social support, loneliness, and depressive status of elderly nursing home residents The De Jong Gierveld short scales for emotional and social loneliness: tested on data from 7 countries in the UN generations and gender surveys The Effect of Information Communication Technology Interventions on Reducing Social Isolation in the Elderly: A Systematic Review Digitally connected, socially disconnected: The effects of relying on technology rather than other people Loneliness and Social Connections: A National Survey of Adults 45 and Older Impact of computer training courses on reduction of loneliness of older people in Finland and Slovenia Older family carers in rural areas: experiences from using caregiver support services based on Information and Communication Technology (ICT) The Benefits of Social Technology Use Among Older Adults Are Mediated by Reduced Loneliness Video Conferencing: An Intervention for Emotional Loneliness in Long-Term Care Can Digital Technology Enhance Social Connectedness Among Older Adults? A Feasibility Study OR sensores[Title/Abstract] OR sensorevery[Title/Abstract] OR sensoreverywhere[Title/Abstract] OR sensorex[Title/Abstract] OR sensorexhibited[Title/Abstract] OR sensorfor[Title/Abstract] OR sensorfret[Title/Abstract] OR sensorg[Title/Abstract] OR sensorgraft[Title/Abstract] OR sensorgram[Title/Abstract] OR sensorgrams[Title/Abstract] OR sensorgraphy[Title/Abstract] OR sensorhis[Title/Abstract] OR sensori[Title/Abstract] OR sensoria[Title/Abstract] OR sensoriactuators[Title/Abstract] OR sensoriais[Title/Abstract] OR sensorial[Title/Abstract] OR sensorial'[Title/Abstract] OR sensoriales[Title/Abstract] OR sensorialist[Title/Abstract] OR sensorialite[Title/Abstract] OR sensoriality[Title/Abstract] OR sensorialized[Title/Abstract] OR sensorially[Title/Abstract] OR sensorials[Title/Abstract] OR sensorialy[Title/Abstract] OR sensoribehavioural[Title/Abstract] OR sensoric[Title/Abstract] OR sensorica[Title/Abstract] OR sensorical[Title/Abstract] OR sensorically[Title/Abstract] OR sensoricortical[Title/Abstract] OR sensoricphysiological[Title/Abstract] OR sensorics[Title/Abstract] OR sensoridiscriminative[Title/Abstract] OR sensoriel[Title/Abstract] OR sensorielle[Title/Abstract] OR sensorielles[Title/Abstract] OR sensoriels[Title/Abstract] OR sensories[Title/Abstract] OR sensorif[Title/Abstract] OR sensorigenesis[Title/Abstract] OR sensorii[Title/Abstract] OR sensoril[Title/Abstract] OR sensorilimbic[Title/Abstract] OR sensorily[Title/Abstract] OR sensorimetry[Title/Abstract] OR sensorimortor[Title/Abstract] OR sensorimoter[Title/Abstract] OR sensorimoteur[Title/Abstract] OR sensorimoteurs[Title/Abstract] OR sensorimotility[Title/Abstract] OR sensorimotor[Title/Abstract] OR sensorimotor'[Title/Abstract] OR sensorimotoraffective[Title/Abstract] OR sensorimotoraxonal[Title/Abstract] OR sensorimotorcontrol[Title/Abstract] OR sensorimotorcortical[Title/Abstract] OR sensorimotori[Title/Abstract] OR sensorimotorial[Title/Abstract] OR sensorimotoric[Title/Abstract] OR sensorimotorrhythm[Title/Abstract] OR sensorimotorsystem[Title/Abstract] OR sensorimotory[Title/Abstract] OR sensorimototor[Title/Abstract] OR sensorimotrice[Title/Abstract] OR sensorimotrices[Title/Abstract] OR sensorimotricity[Title/Abstract] OR sensorin[Title/Abstract] OR sensorincludes[Title/Abstract] OR sensorineal[Title/Abstract] OR sensorineral[Title/Abstract] OR sensorineraul[Title/Abstract] OR sensorinerual[Title/Abstract] OR sensorinerural[Title/Abstract] OR sensoriness OR phonetician's[Title/Abstract] OR phoneticians[Title/Abstract] OR phoneticization[Title/Abstract] OR phoneticized[Title/Abstract] OR phonetics[Title/Abstract] OR phonetics'[Title/Abstract] OR phonetik[Title/Abstract] OR phonetion[Title/Abstract] OR phonetique[Title/Abstract] OR phonetisation[Title/Abstract] OR phonetogram[Title/Abstract] OR phonetograms[Title/Abstract] OR phonetograph[Title/Abstract] OR phonetographic[Title/Abstract] OR phonetography[Title/Abstract] OR phonetometric[Title/Abstract] OR phonetoxin[Title/Abstract] OR phonetype[Title/Abstract] OR phonetypical[Title/Abstract] OR phoneus[Title/Abstract] OR phoneuse[Title/Abstract] OR phoneutria[Title/Abstract] OR phoneutriatoxin[Title/Abstract] OR phoneutrism[Title/Abstract] OR phoney[Title/Abstract] OR phoneyusa[Title/Abstract])) OR (telephone[Title/Abstract] OR telephone digital* or technol* or sensor* or robot* or internet* or social media or smartphone* or smart phone* or telephone* or phone* or online or ipad* or computer* or electronic* or Web) ti,ab. 3. (lonely or social isolation).ti,ab digital* or technol* or sensor* or robot* or internet* or social media or smartphone* or smart phone* or telephone* or phone* or online or ipad* or computer* or electronic* or Web) 14 to (english language and humans and yr="2010 -Current" and (clinical trial, all or evaluation studies or journal article or meta analysis or observational study or The authors acknowledge support from Liz Callow of Bodleian Health Care Libraries, University of Oxford for running literature searches. We gratefully thank to Dr Alexandra Farrow, Brunel University London for checking the manuscript.