Evidence base
Publications register
What our guidance rests on. Each entry carries our summary, our note on what it means for practice, and a link to the original document.71 records.
2026
America Is Running Out of Places to Hang Out as Movie Theaters, Bars, Diners and Bowling Alleys Drop to All-Time Low
This Fortune report tracks the long-term decline of American "third places"—bowling centres, libraries, Elks lodges, and similar venues—citing steep drops in numbers and usage: bowling centres fell from about twelve thousand in the mid-1960s to under four thousand, library visits nearly halved between 2019 and 2022, and Elks lodge membership has dropped by half since 1976. It identifies rising costs, both of operating these venues and of attending them, as a primary driver of the decline.
What it means for practice
This is useful context for anyone designing community infrastructure or social connection interventions: it documents the erosion of informal, low-cost gathering spaces that programmes often assume still exist. Readers should treat the cost narrative as the reporter's explanatory frame rather than a tested causal finding, and look to dedicated research for mechanisms behind the decline.
Loneliness Strongly Linked to Poorer Mental Health and Wellbeing
A collaboration between the University of Bristol, Nesta, and Amsterdam UMC, published in Nature Communications, found that loneliness and social isolation are strongly linked to poorer mental health, reduced wellbeing, and worse general health.
What it means for practice
This adds to the evidence base establishing loneliness as a determinant of health outcomes, which matters for anyone building the case for social connection interventions to funders or health commissioners. The findings describe association, not mechanism or intervention effect, so they should be read as motivation for action rather than evidence of what works.
A Loneliness Epidemic Meets Uneven Social Infrastructure
This reporting links the decline of "third places" — informal community and social spaces — to broader loneliness prevalence data, and notes that this decline is not evenly distributed across geographies.
What it means for practice
Useful as framing context for why place-based social infrastructure matters, but it is journalism connecting existing datasets rather than new primary research; treat the geographic pattern as a lead for further investigation, not a finding on its own.
Americans of All Ages Are Spending Less Time Socializing
Axios reporting cites time-use data showing Americans' daily face-to-face socializing fell from 45 to about 35 minutes over twenty years, with only 30% of Americans socializing on an average day in 2024, down from 38% in 2014. The decline is steepest among 15-24 year-olds, whose daily socializing time dropped from roughly one hour to 35 minutes. Americans averaged 6 hours 40 minutes alone per day in 2024.
What it means for practice
This is a useful top-line indicator of eroding social contact time across age groups, but as reporting on time-use data it doesn't establish causes or what counts as "socializing." Programme designers should treat the youth figure as a signal worth cross-checking against dedicated loneliness or time-use surveys before building targeting decisions on it.
Longitudinal Association Between Social Isolation, Loneliness and Well-Being Among Community-Dwelling Older Adults in the United States
This longitudinal study tracks how social isolation and loneliness relate to well-being over time among community-dwelling older adults in the United States. The findings supplied indicate an association between these variables and well-being trajectories, though the specific measures, direction, and magnitude of the relationship are not detailed here.
What it means for practice
Longitudinal designs matter for this field because they can move beyond the cross-sectional snapshots that dominate loneliness research, but readers should seek the full paper to check how isolation and loneliness were measured separately and what confounders were controlled for before drawing programme conclusions.
At Best, Social Media Connections Unlikely to Make You Less Lonely
A study of adults aged 30-70 found that connecting online with strangers is associated with greater loneliness. Connecting online with people already known showed no association with increased loneliness, but also none with reduced loneliness.
What it means for practice
This is a useful check against the assumption that online contact substitutes for offline connection when designing loneliness interventions: the data suggest online interaction with existing contacts is at best neutral, and with strangers may be counterproductive. Programme designers relying on digital befriending or online community models should treat reduced loneliness as an outcome to test for, not assume.
Closeness of Social Media Contacts and Loneliness Among US Adults: A Nationally Representative Study
This nationally representative study of US adults examined whether the closeness of a person's social media contacts moderates their level of loneliness, rather than treating online contact as a single undifferentiated category.
What it means for practice
The finding supplied only establishes that closeness was examined as a moderator, not the direction or size of the effect; readers should treat this as a signal that "screen time" framings of loneliness are too blunt, and should look for the fuller results before drawing programme conclusions.
How Does Turning to AI for Companionship Predict Loneliness, and Vice Versa?
This 12-month longitudinal study followed more than 2,000 adults across four Western countries, examining the relationship between social chatbot use and loneliness. It found that increased chatbot use predicted increased loneliness over time, and framed the relationship as bidirectional: lonely people are drawn to chatbots, and that use appears to deepen loneliness further.
What it means for practice
The bidirectional design is what makes this worth reading closely; it moves past the simpler claim that lonely people simply seek out AI companionship and suggests a reinforcing loop that programme designers and clinicians recommending AI tools for isolation should not assume away.
Behavioral Activation and Mindfulness Interventions in Reducing Loneliness and Improving Well-Being in Older Adults: The HEAL-HOA Randomized Clinical Trial
A randomised trial of 1,151 older adults living in poverty, alone, and digitally excluded tested telephone-delivered behavioural activation and mindfulness against a befriending control. The intervention, delivered as eight 30-minute calls over one month by trained laypeople who were themselves older adults with lived experience of loneliness, produced significantly greater reductions in loneliness at 12 months than befriending.
What it means for practice
Befriending is the default intervention commissioned for loneliness in this population, so a trial in which it loses to a structured, peer-delivered alternative is a direct challenge to current commissioning defaults rather than an incremental finding. Programme managers choosing between models for isolated older adults should treat this as evidence that intervention structure and technique matter more than simple social contact.
Home Alone: Remote Work, Isolation, and Mental Health
This study finds that remote work increases time spent alone and worsens mental well-being, including higher use of mental health services and prescriptions. The shift to remote work accounts for roughly a third of the rise in isolation and mental distress observed between 2011-2019 and 2022-2024. Remote workers who live alone experienced an increase in days spent alone roughly ten times greater than that of remote workers who live with others.
What it means for practice
This gives employers and workplace programme designers a quantified link between remote work arrangements and mental health burden, not just a plausible association. The finding on employees living alone points to where intervention or monitoring should be targeted first, rather than treating remote-work risk as uniform across a workforce.
AI Companions Can Comfort Lonely Users but May Deepen Distress Over Time
Reporting on the AI companion literature finds that short-term use of AI companions can provide genuine comfort to lonely users, but that this effect diverges over time: heavy daily use correlates with increased loneliness and reduced human social engagement. The distinction the evidence points to is duration of use, not simply whether someone uses these tools at all.
What it means for practice
For programme managers weighing AI companion tools as a loneliness intervention, this reporting is a caution against treating early positive user feedback as proof of benefit; any pilot or rollout should track usage intensity over time and human contact levels, not just initial satisfaction.
Loneliness, Depression and Generalized Anxiety Across Eight Countries
A study across eight countries found that nearly half of young adults report experiencing loneliness. The association between loneliness and both depression and generalized anxiety persisted after adjusting for demographics and prior diagnoses.
What it means for practice
The cross-national scope and the adjustment for prior diagnoses strengthen the case that loneliness tracks with mental health outcomes independent of who is already more vulnerable, which matters for anyone designing youth-facing programmes or making the case to funders that loneliness is a target in its own right, not just a symptom of existing conditions.
Prevalence of Loneliness and Social Isolation by Demographic, Socioeconomic, and Other Factors: Estimates From the 2024 Household Pulse Survey
This study uses 2024 Household Pulse Survey data to break down the prevalence of loneliness and social isolation across demographic and socioeconomic factors. The publication, appearing in 2026, works from a large national dataset to show how rates vary across population groups.
What it means for practice
Use this as a reference for targeting rather than as evidence of programme effect: it tells you where loneliness concentrates by group, which is useful for prioritising outreach, but it does not test what reduces it. Pair it with intervention evaluations before committing resources.
Loneliness Around the World: Patterns, Predictors, and Well-Being Implications
This cross-national study examines patterns and predictors of loneliness across countries and links them to well-being outcomes. It positions loneliness as a variable that differs systematically across national contexts rather than a uniform individual experience.
What it means for practice
The finding as supplied is high-level and does not specify which countries, predictors, or effect sizes drive the pattern, so readers should treat this as a signal to seek the full study before drawing programme or policy conclusions.
AI Chatbots and Digital Companions Are Reshaping Emotional Connection
This APA Monitor review surveys emerging evidence on AI chatbots and digital companions as sources of emotional connection, finding the research base mixed rather than conclusive. It reports that effects on loneliness depend on who is using the tool, how the chatbot is designed, and how intensively it is used, rather than following a simple positive or negative pattern.
What it means for practice
This is a review, not primary research, so it should be read as a map of an unsettled field rather than a verdict on whether AI companionship helps or harms. Programme designers considering AI tools for isolation or loneliness work should treat the dependency on user characteristics and usage patterns as a caution against one-size-fits-all deployment.
New Directions in the Investigation of Loneliness and Social Connection Across Time Scales and Contexts
This review surveys emerging methodological approaches to measuring loneliness across different time scales and contexts, mapping how researchers are moving beyond single-point-in-time survey snapshots toward capturing variation over shorter and longer horizons.
What it means for practice
This is a methods-orientated survey rather than a findings report, so it won't give you effect sizes or intervention outcomes. It's useful for anyone designing a measurement or evaluation strategy who wants to know what tools exist before committing to a single loneliness metric.
Risk Factors of Loneliness in Community-Dwelling Socially Isolated Older Adults
This study examines community-dwelling older adults who are socially isolated and identifies which of them are most at risk of also experiencing loneliness, treating isolation and loneliness as distinct but related conditions.
What it means for practice
For programme managers designing outreach for older adults, this distinction matters: interventions aimed at increasing contact frequency will not necessarily address loneliness. Anyone commissioning social prescribing services for this group should check whether the risk factors named here align with their own referral criteria.
2025
Randomized Controlled Trial on the Impact of Befriending on Depression, Anxiety, Loneliness, and Social Support in Older People in Aged Care
A randomized controlled trial of befriending in residential aged care found reductions in UCLA Loneliness Scale scores of 2.39 points at 8 weeks and 2.71 points at 16 weeks compared with a control group. The trial also assessed depression, anxiety, and social support outcomes.
What it means for practice
This confirms befriending has a measurable effect on loneliness in aged care settings, but the finding needs to be read against the HEAL-HOA comparison, where befriending was outperformed by structured psychological approaches head-to-head. For programme managers choosing between interventions, this suggests befriending is a viable option where structured psychological approaches are not feasible, not necessarily the strongest available choice.
Remote Work and Loneliness: Evidence from a Nationally Representative Sample of Employed U.S. Adults
This study uses a nationally representative sample of employed U.S. adults to examine the association between remote work arrangements and loneliness. The findings, published in the Journal of Affective Disorders in December 2025, establish a national-level picture of how work location relates to reported loneliness among the employed population.
What it means for practice
Employers and workplace policymakers weighing remote or hybrid arrangements need evidence beyond individual case studies or small-sample surveys; a nationally representative dataset carries more weight for setting policy. Readers should note the summary as supplied does not specify the direction or strength of the association, so this entry should be read alongside the full study before drawing conclusions for programme design.
Prevalence of Loneliness States Among the U.S. Adult Population: Findings From the 2022 HINTS-6
Analysis of 2022 HINTS-6 survey data, published in 2025, finds that roughly 1 in 7 U.S. adults report severe loneliness and 1 in 5 report moderate loneliness, meaning more than 37% experience moderate-to-severe loneliness.
What it means for practice
This gives programme managers and funders a national baseline prevalence figure to size interventions against, but the underlying data are from 2022 and published with a three-year lag, so read it as a snapshot rather than a current trend measure.
Young Leaders Call for Urgent Action to Tackle Social Isolation and Loneliness at the Global Model WHO 2025
At the Global Model WHO 2025, youth delegates called on national governments to embed social connection explicitly within health strategies, framing loneliness and social isolation as a priority for formal policy attention rather than an incidental concern.
What it means for practice
This is an advocacy signal, not a research finding: it shows youth constituencies pushing loneliness up the political agenda, which matters for funders and programme designers watching where mandates and money may follow, but it establishes no new evidence on what works.
AI, Loneliness, and the Value of Human Connection
This commentary from George Mason University's College of Public Health frames AI companionship tools as carrying a substitution risk for human connection, rather than treating them as a straightforward remedy for loneliness. The piece argues for public health framing of the issue over a purely technological one.
What it means for practice
Useful as a counterweight to vendor and product framing of AI companionship, but it is a commentary, not a study — it offers a framing argument rather than evidence about outcomes, so treat it as a position to weigh against measured data.
Disconnected: The Escalating Challenge of Loneliness Among Adults 45-Plus
This AARP Public Policy Institute survey follows earlier waves studying loneliness among adults aged 45 and older, and finds loneliness has risen in this population since the prior 2018 survey. Because the survey instrument was kept consistent across waves, the 2025 results are directly comparable to the 2018 baseline.
What it means for practice
Trend comparability is unusual in loneliness research, where instruments often change between studies and undermine claims about change over time. That makes this a useful reference point for programme managers and funders tracking whether loneliness in the 45-plus population is worsening, though the findings supplied here describe direction of change only, not magnitude or cause.
As Community Spaces Disappear, New Research Warns of Health and Equity Risks
This piece reports on research examining the decline of "third places"—senior centres, churches, and other informal community spaces—for a general audience. It frames the disappearance of these spaces as a cause of loneliness, not merely a downstream effect of it, and connects this loss to broader health and equity concerns.
What it means for practice
This is journalistic coverage rather than the underlying study, so treat it as an accessible entry point into the argument rather than a source of primary evidence. Its value for practice is the framing: it positions physical infrastructure loss as an upstream driver worth intervening on, which matters for anyone designing programmes or making the case for funding community spaces.
Uneven Access to Essential Services and Amenities: Geographic Disparities in Third Place Availability Across the United States, 2010 to 2021
This study tracked twelve categories of third places—coffee shops, libraries, museums, recreation centres, and restaurants among them—across the US between 2010 and 2021. It found closures in every category between 2019 and 2021, with losses concentrated in census tracts with higher social vulnerability and in rural areas.
What it means for practice
This gives programme designers and funders a geographic evidence base for where connection infrastructure has thinned out most, useful for targeting resources rather than assuming uniform decline. It documents loss, not causation, so it should not be read as evidence that closures were driven by the pandemic itself rather than pre-existing vulnerability.
The Role of Social Prescribing in Alleviating Social Isolation and Loneliness in Older Adults: A Systematic Review Protocol
This protocol sets out a planned systematic review examining whether social prescribing reduces social isolation and loneliness in older adults. It notes that despite widespread adoption of social prescribing, evidence of effectiveness remains unclear, and that only one peer-reviewed randomised controlled trial exists in this area.
What it means for practice
This is a protocol, not a completed review, so it establishes a question and a gap rather than an answer. The finding that only one RCT exists is itself the most useful data point here: it should temper confidence in any current claims about social prescribing's impact on loneliness in older adults, pending the review's results.
WHO Director-General Opening Remarks at the Launch of the Commission on Social Connection Report
In opening remarks marking the launch of the WHO Commission on Social Connection's report (June 2025), the WHO Director-General frames social connection as a determinant of health, positioning it alongside other major public health priorities.
What it means for practice
This is a framing statement, not a findings document: it signals that social connection has WHO institutional backing as a health priority, which matters for advocacy and funding cases, but it establishes no evidence itself and should be read alongside the underlying Commission report for substance.
From Loneliness to Social Connection: Charting a Path to Healthier Societies
The WHO Commission on Social Connection reports that loneliness affects 1 in 6 people globally and is associated with an estimated 871,000 deaths annually. Prevalence is highest among young people and in low-income countries. The Commission calls on Member States to treat social connection as a public health priority.
What it means for practice
This is the reference point for framing loneliness as a population-level health issue rather than an individual wellbeing concern, useful for funders and policymakers building a mandate for action. It establishes scale and urgency but does not itself specify which interventions work; treat it as the case for action, not a delivery guide.
What Americans Say About Loneliness
This 2025 Harvard Gazette piece reports on contemporary American attitudes toward loneliness, covering how people perceive its causes. The finding supplied does not specify the survey methodology, sample, or the particular causes named.
What it means for practice
As a reporting piece rather than a primary study, this is useful for gauging public framing of loneliness but should not be treated as evidence of prevalence or cause; readers needing figures to plan interventions should look for the underlying survey data.
Singles in America 2025: The 14th Annual Study
The 2025 Singles in America study, based on a demographically representative sample of 5,001 U.S. singles aged 18 to 98, finds that 46% report feeling ready for a long-term relationship, a shift away from casual encounters. It also finds AI use in dating rose more than 300% since 2024, with nearly half of Gen Z singles reporting AI use in their dating life. Now in its 14th year, the study's main value lies in its time series rather than any single year's figures.
What it means for practice
This is commissioned by a dating company, so figures on relationship readiness and AI adoption should be read as industry-sponsored rather than independent research; useful for tracking directional trends over time, less so as a standalone evidentiary source.
Loneliness in America 2025
A 2025 survey by The Cigna Group and Evernorth Research Institute, based on fieldwork conducted with over 7,500 U.S. adults between 29 May and 14 June 2024, found that 57% of Americans report loneliness and 52% of workers report feeling lonely. Gen Z and Millennials report higher loneliness than Gen X and Baby Boomers, despite having greater technological connectedness.
What it means for practice
The gap between generational technology use and reported loneliness is worth attention for anyone designing workplace or community interventions, since it undercuts assumptions that digital connectivity alone reduces isolation. Note the lag between the 2024 fieldwork and 2025 publication when citing this alongside more recent data.
Social Connections and Loneliness in OECD Countries
This OECD flagship report presents cross-country data on social connectedness across OECD countries, finding that on average 10% of people feel unsupported, 8% report no close friends, and 6% experienced loneliness most or all of the past four weeks. In-person meeting frequency has declined over a longer period, while self-reported connection has worsened more recently, with men and young people -- previously considered lower risk -- showing some of the largest deteriorations. The report also examines social infrastructure and digitalisation as drivers, and discusses measurement implications, building on the OECD's How's Life? 2024 work.
What it means for practice
This is the most complete cross-country comparison of social connectedness currently available, useful as a baseline for anyone benchmarking national or local prevalence figures. Read the finding on men and young people carefully: it should prompt a check of whether your own targeting assumptions still hold, rather than being taken as settled cause-and-effect.
Social Connections Combat Loneliness and Promote Wellbeing Among College Students Coming Out of the COVID-19 Pandemic
This study looks at college students in the period following the COVID-19 pandemic to identify which types of social connection predicted wellbeing. It finds that the quality and regularity of contact were stronger predictors than the sheer number of connections students had.
What it means for practice
This adds to a growing body of evidence pushing back against "more contact is better" assumptions in loneliness interventions; programme designers working with young adults should weigh consistency and depth of interaction over volume of social touchpoints when building connection-focused activities.
2024
Understanding the Interplay Between Social Isolation, Age, and Loneliness During the COVID-19 Pandemic
This study examines how objective social isolation and subjective loneliness relate to one another during the COVID-19 pandemic, and shows that this relationship shifts across age groups rather than holding as a fixed pattern. The findings support treating isolation and loneliness as distinct constructs rather than interchangeable proxies for one another.
What it means for practice
Programmes that measure only isolation (contact frequency) or only loneliness (subjective scales) risk missing the group for whom the two diverge, and this paper gives an age-stratified basis for arguing that both need separate measurement and separate intervention design.
The Effects of Volunteering on Loneliness Among Lonely Older Adults: The HEAL-HOA Dual Randomised Controlled Trial
This randomised controlled trial tested whether prosocial engagement and volunteering reduce loneliness among lonely older adults in Hong Kong, comparing participants against a control group. It is one of the few RCTs in this area, in a field otherwise dominated by uncontrolled programme evaluations.
What it means for practice
Because it uses a controlled design, this trial carries more evidential weight than most volunteering-and-loneliness studies, and should be a reference point when assessing claims made from weaker, uncontrolled evaluations of similar programmes.
Over 1 in 5 People Worldwide Feel Lonely a Lot
Gallup's global survey finds that more than one in five adults worldwide report feeling lonely a lot of the previous day. The 2024 data confirm a pattern already documented elsewhere: younger adults report elevated loneliness relative to older age groups.
What it means for practice
The scale and consistency of this finding across a global sample makes age-targeted assumptions about loneliness (that it is primarily an older-adult problem) harder to sustain. Programme designers should read this alongside local prevalence data before assuming their population matches the global pattern.
Loneliness, Lack of Social and Emotional Support, and Mental Health Issues -- United States, 2022
This CDC MMWR report presents federal surveillance estimates of loneliness and lack of social and emotional support among U.S. adults, drawn from 2022 survey data (published June 2024). It links social disconnection to a range of health outcomes, including heart disease, stroke, dementia, type 2 diabetes, depression, anxiety, and premature mortality.
What it means for practice
This is a national baseline for prevalence, not a causal or intervention study — use it to size the problem or justify a programme's rationale, not to select what to do about it. Note the two-year lag between data collection and publication when comparing against more recent local figures.
1 in 5 Employees Worldwide Feel Lonely
Gallup's 2024 State of the Global Workplace survey finds that one in five employees worldwide report feeling lonely a lot of the previous day. Loneliness rates vary by work arrangement: 25% among fully remote employees, 21% among hybrid workers, and 16% among fully on-site staff. Employees under 35 report higher loneliness than older workers.
What it means for practice
This gives employers a concrete, comparative figure across work arrangements rather than a single loneliness estimate, which matters for decisions about remote and hybrid policy design. It is self-reported, single-day recall data, so it should inform rather than settle debates about causation between work arrangement and loneliness.
The Social Brain Hypothesis -- Thirty Years On
This review, authored by Dunbar himself, revisits thirty years of evidence for and against the social brain hypothesis, which links primate neocortex size to social group size. Rather than restating the headline claim, it weighs the accumulated support and the challenges raised against it.
What it means for practice
Useful as a corrective for anyone citing the "Dunbar's number" claim as settled fact: it comes from the theory's originator and surveys where the evidence has held up and where it hasn't, which matters for programme design work that leans on group-size heuristics.
Loneliness and Social Support Networks: Findings from the KFF Survey of Racism, Discrimination and Health
This 2024 KFF survey examines loneliness and social support by race and ethnicity, a breakdown most national surveys omit. It links reported experiences of discrimination to the size of people's social support networks.
What it means for practice
Most loneliness data treats populations as homogeneous; this survey gives programme designers and funders a rare disaggregated view, and its link between discrimination and network size is a useful prompt for targeting outreach rather than a causal claim to build interventions on alone.
Buffet's New App Tackles the Loneliness Epidemic by Connecting People in the Real World
TechCrunch reported on Buffet, a social matching app launched in Los Angeles in 2024 aimed at addressing loneliness. Users enter five interests and answer personality questions, and an algorithm matches them with another person and a venue for an in-person meeting. The app operates on a freemium model with plans for expansion beyond LA.
What it means for practice
This is trade-press coverage of a commercial product at launch, not an evaluation of whether algorithmic matching actually reduces loneliness or leads to sustained connection. Readers should treat it as a market signal rather than evidence of effectiveness.
The Evidence: Collaborative Review Documents on Adolescent Mental Health and Social Media
This is a set of open, collaboratively maintained review documents assembled by Jonathan Haidt and Jean Twenge, compiling studies on the relationship between social media use and adolescent mental health. The documents include dissenting evidence alongside studies supporting a causal link, rather than presenting a single settled position.
What it means for practice
Because the compilers are prominent advocates for the causal link between social media and adolescent mental health, this should be read as an argument-in-progress from interested parties rather than an independent synthesis; its value lies in surfacing dissenting studies you can check against their framing, not in resolving the debate.
A Cross-Sectional Investigation on Remote Working, Loneliness, Workplace Isolation, Well-Being and Perceived Social Support in Healthcare Workers
This cross-sectional study of healthcare workers examines remote working alongside loneliness, workplace isolation, well-being and perceived social support. It finds that workplace isolation and loneliness are distinct constructs with different correlates, and that perceived social support moderates the relationship between remote working and well-being.
What it means for practice
The separation of workplace isolation from loneliness matters for programme design: interventions aimed at reducing subjective loneliness may not address structural isolation from colleagues, and vice versa. Because the design is cross-sectional, it establishes association rather than the direction of effect between remote working, support and well-being.
Americans Don't Socialize Face-to-Face as Much as They Used To
This report draws on American Time Use Survey data to describe a decline in face-to-face socializing among Americans. It notes the decline predates the pandemic and has continued in the years since, rather than being a pandemic-specific effect.
What it means for practice
Useful as context-setting for programme design and community practice work, but this is a media report summarizing survey trends, not a primary analysis: readers needing figures or methodology should trace back to the underlying Time Use Survey data rather than cite this piece directly.
The Epidemic of Loneliness
This review examines how loneliness has come to be framed as an epidemic, and weighs the evidence used to support that framing.
What it means for practice
Before adopting "epidemic" language in programme or funding documents, it's worth understanding what that framing does and does not establish about scale, causation, or trend — this review is a useful check on that rhetoric rather than a source of new prevalence data.
2023
WHO Launches Commission to Foster Social Connection
In November 2023, the World Health Organization announced the establishment of a Commission on Social Connection, co-chaired by Vivek Murthy and Chido Mpemba. The announcement frames loneliness and social disconnection as a global public health priority, explicitly rejecting the framing of loneliness as a concern limited to high-income countries.
What it means for practice
This is an announcement, not a research output: it signals institutional attention and framing rather than evidence or findings. Read it as context for why social connection is gaining traction as a commissioning and policy priority, not as a source of data to cite.
How Strong Are the World's Social Connections?
Gallup's analysis breaks down social connectedness across countries and demographic groups, drawing on its global data. A key finding is that connectedness and loneliness should not be treated as simple inverses of one another; a person can score low on one without scoring high on the other.
What it means for practice
This matters for anyone designing measurement tools: if you're only tracking loneliness as a proxy for connection, you may be missing the people who report weak connectedness without describing themselves as lonely. Read this alongside programme-level evaluation work rather than as a substitute for it, since the analysis is descriptive and cross-national rather than diagnostic of what drives the gap.
The State of Social Connections
Gallup and Meta surveyed roughly 1,000 people aged 15 and older in each of 142 countries between June 2022 and February 2023 (China excluded, covering roughly 77% of the world's adult population). 72% of respondents, an estimated 3.2 billion people across the surveyed countries, reported feeling very or fairly socially connected.
What it means for practice
This is a baseline prevalence figure, not an explanation of what drives connection or disconnection, so it's most useful for framing scale rather than guiding intervention design. Its breadth across 142 countries makes it a reasonable reference point for comparing national or regional connection rates, provided you note the China exclusion when citing global totals.
Almost a Quarter of the World Feels Lonely
Gallup's 2023 survey of 142 countries found that 24% of people worldwide feel very or fairly lonely, with the highest rate among 19-29 year-olds (27%) and the lowest among adults 65 and older (17%). Nearly half of respondents (49%) reported not feeling lonely at all.
What it means for practice
This is a cross-national baseline, not a trend line: it establishes scale and the age gradient but says nothing about causes or change over time. Programme designers and funders should treat the youth-loneliness figure as a prompt to check local data before assuming it generalises to their population.
The State of Loneliness and Social Isolation Research: Current Knowledge and Future Directions
This review maps the current landscape of loneliness and social isolation research, identifying methodological gaps across the field. It highlights inconsistent measurement approaches as a key barrier preventing meaningful comparison of findings across studies.
What it means for practice
Before commissioning or comparing evaluation data on loneliness interventions, read this: it explains why cross-study comparisons in this field are often unreliable and flags measurement inconsistency as a structural problem, not just a quirk of individual studies.
Our Epidemic of Loneliness and Isolation: The U.S. Surgeon General Advisory on the Healing Effects of Social Connection and Community
This 2023 advisory from the U.S. Office of the Surgeon General reports that roughly half of U.S. adults experience loneliness, and states that the mortality risk associated with social disconnection is comparable to smoking up to 15 cigarettes a day. It sets out a six-pillar National Strategy to Advance Social Connection as a policy response.
What it means for practice
This is the reference point most funders and health systems now cite when justifying investment in connection-focused work, so it's worth reading in full rather than by headline stat. It establishes scale and framing at national policy level, but the six pillars are a strategic outline, not an implementation guide, so readers running programmes will need to look elsewhere for operational detail.
Changes in Social Isolation and Loneliness Prevalence During the COVID-19 Pandemic in Japan: The JACSIS 2020-2021 Study
This study tracked social isolation and loneliness prevalence in Japan across 2020 and 2021 using longitudinal panel data from the JACSIS survey, allowing it to assess change over the course of the pandemic rather than relying on a single-timepoint measurement. It provides a non-US national dataset for this period.
What it means for practice
The longitudinal design makes this more useful than cross-sectional pandemic studies for judging whether isolation and loneliness rose, fell, or held steady over time, and the Japanese setting gives programme managers a comparison point against the largely US- and UK-dominated evidence base. Read it alongside local prevalence data before assuming trends transfer directly.
2022
Editorial: The Impact of Social Isolation and Loneliness on Mental Health and Wellbeing
This editorial introduces a collection of fourteen papers on social isolation and loneliness gathered during the COVID-19 pandemic, framing their combined findings on the mental health and wellbeing consequences of social disconnection.
What it means for practice
As an editorial overview rather than primary research, this is best used as a map to the wider collection rather than as evidence in its own right; readers wanting specifics on measurement or intervention effects should follow through to the underlying papers.
Do People Perceive Benefits in the Use of Social Prescribing to Address Loneliness and/or Social Isolation? A Qualitative Meta-Synthesis
This qualitative meta-synthesis pulls together studies on how people experience social prescribing for loneliness and social isolation. Participants reported benefits going beyond simple social contact, describing restored meaningful participation and a sense of purpose. Across the synthesised studies, structured, purposeful group activity appeared more effective than contact alone.
What it means for practice
For anyone designing or commissioning social prescribing services, this is a signal to weight referral pathways toward structured, purpose-driven activities rather than generic social contact, though as a qualitative synthesis it speaks to perceived benefit rather than measured outcomes.
Social Isolation and Loneliness Increase the Risk of Death from Heart Attack, Stroke
This American Heart Association newsroom release summarises a scientific statement finding that social isolation and loneliness raise the risk of dying from heart attack and stroke. It flags older adults and socially vulnerable groups as facing elevated risk of isolation.
What it means for practice
This is a plain-language announcement, not the underlying scientific statement itself, so treat it as a citable summary for general audiences rather than a source of clinical detail or effect sizes. It's useful for making the case to health-systems stakeholders that loneliness belongs on the cardiovascular risk agenda, particularly for older and socially vulnerable populations.
Effects of Objective and Perceived Social Isolation on Cardiovascular and Brain Health: A Scientific Statement From the American Heart Association
This American Heart Association scientific statement (2022) reviews evidence linking social isolation and loneliness to cardiovascular outcomes, finding roughly a 30% increased risk of heart attack, stroke, or related death. It reports a 29% increased risk of heart attack and/or coronary death and a 32% increased risk of stroke, and notes worse prognosis—including recurrence and mortality—among those already living with coronary heart disease or stroke. The statement explicitly flags the lack of intervention evidence as the field's central gap.
What it means for practice
This is a strong evidentiary base establishing association and risk magnitude, but it does not tell you what works to reduce that risk—commissioners and clinicians reaching for this to justify a specific intervention should note the authors' own caveat that intervention evidence is largely absent. Read alongside programme evaluations rather than in place of them.
2021
Loneliness and Social Isolation as Risk Factors: The Power of Social Connection in Prevention
This review argues that social connection should be treated as a modifiable risk factor within preventive health frameworks, positioning it alongside diet, exercise, and smoking cessation as a target for population-level intervention.
What it means for practice
This is a framing argument rather than an intervention study: it makes the case for where social connection belongs in prevention frameworks, but does not establish which programmes or approaches actually reduce isolation. Useful for justifying investment in connection work to health-system audiences, less useful for choosing what to fund.
Understanding Loneliness: A Systematic Review of the Impact of Social Prescribing Initiatives on Loneliness
This systematic review examined nine studies of social prescribing initiatives and their effect on loneliness. All nine reported positive individual impacts on participants, and three of the nine also reported reductions in use of GP, emergency, social worker, or inpatient services.
What it means for practice
The consistency of positive individual outcomes across a small evidence base is encouraging, but the service-use reduction finding rests on only three studies, so it should not be treated as a reliable basis for cost or demand-reduction claims when commissioning social prescribing schemes.
Joint Message from the Loneliness Ministers Meeting
This joint statement records the first bilateral ministerial meeting between Japan and the UK dedicated specifically to loneliness policy, held in June 2021. It marks loneliness as a distinct cross-government policy area rather than a subset of health or welfare portfolios in either country.
What it means for practice
This is a diplomatic signal, not a research finding or programme evaluation: it tells you loneliness has reached ministerial-level attention in at least two governments, which is useful context for funding and commissioning arguments but establishes no evidence about what works.
The State of American Friendship: Change, Challenges, and Loss
This survey compares Americans' close-friendship networks in 2021 to 1990 benchmarks. The share reporting no close friends rose from 3% to 12% overall, with a steeper drop among men: those with six or more close friends fell from 55% to 27%, and men reporting no close friends rose fivefold to 15%.
What it means for practice
The 1990-to-2021 comparison gives this trend line more weight than a single-point survey, and the gendered split is useful for anyone designing outreach or peer-support programmes aimed at isolated men. Read it as evidence of a long decline rather than a pandemic-specific effect, since the drop predates 2020.
Dunbar's Number: Why My Theory That Humans Can Only Maintain 150 Friendships Has Withstood 30 Years of Scrutiny
Dunbar argues that humans can maintain around 150 stable social relationships, a figure derived from fitting the relationship between neocortex ratio and group size across primates and inserting the human value (yielding 147.8). He stresses that the layered structure around that number matters as much as the total: roughly 5 intimates, 15 close friends, 50 friends, and 150 meaningful contacts.
What it means for practice
This is the originator's own plain-language restatement of the theory, useful as a citable source rather than a secondary account, but it is a defence of the concept rather than new evidence, so readers should treat it as argument, not fresh data. The layered structure is arguably more actionable for programme design than the headline number itself.
Can Social Prescribing Foster Individual and Community Well-Being? A Systematic Review of the Evidence
This systematic review examines the evidence base for social prescribing's effects on individual and community well-being. It finds that increases in self-esteem and self-confidence are consistently reported as outcomes, but that the underlying trial evidence is limited and programmes vary widely in design.
What it means for practice
For anyone commissioning or building a social prescribing service, this is a reminder that the well-being case rests more on plausibility and consistent self-report gains than on robust comparative trials. Treat the heterogeneity finding as a flag: outcomes from one programme model may not transfer cleanly to another.
Japan Appoints Minister of Loneliness and Isolation
In February 2021 Japan's Cabinet Office announced the appointment of Tetsushi Sakamoto to a newly created ministerial post covering loneliness and isolation. The move made Japan the second country, after the UK, to establish a dedicated ministerial role on this issue.
What it means for practice
This is a government action record, not a programme or evaluation, but it marks loneliness as a matter warranting national coordination rather than local or clinical response alone. Readers working on funding or policy design should treat it as a signal of institutional priority, not as evidence of what such a ministry can achieve.
Loneliness in America: How the Pandemic Has Deepened an Epidemic of Loneliness
This 2021 survey from Harvard's Making Caring Common project found 36% of Americans reported serious loneliness, rising to 61% of young adults aged 18-25 and 51% of mothers with young children. 43% of young adults said their loneliness had increased since the pandemic began. About half of lonely young adults said no one had spent more than a few minutes recently asking how they were doing in a way that felt genuine.
What it means for practice
The finding on genuine check-ins points past prevalence statistics toward a specific, actionable gap: the quality of everyday contact, not just its frequency. Programme designers targeting young adults and new mothers should treat this as evidence for where to intervene, not as a measure of intervention effectiveness, since the data are self-reported and pandemic-specific.
2020
Social Isolation and Loneliness in Older Adults: Review and Commentary of a National Academies Report
This 2020 commentary reviews the National Academies consensus report on social isolation and loneliness in older adults, written for a clinician audience. It argues for routine assessment of isolation within clinical settings and discusses the practical requirements such assessment would involve.
What it means for practice
Useful as a bridge document between the underlying National Academies evidence and frontline clinical practice, since it translates a broad consensus report into what a clinician would actually need to change. It does not itself generate new evidence on isolation, so should be read alongside the source report rather than instead of it.
Social Isolation and Loneliness in Older Adults: Opportunities for the Health Care System
This 2020 consensus report from the National Academies finds that approximately one quarter of adults aged 65 and older are socially isolated, and calls on the health care system to routinely assess social isolation and loneliness among older patients.
What it means for practice
This report establishes the case for embedding social isolation assessment into routine clinical practice, but it stops short of specifying validated tools or workflows, so readers building a screening protocol will need to pair it with implementation guidance from elsewhere.
Loneliness and the Workplace: 2020 U.S. Report
Cigna's 2020 U.S. survey found 61% of adults report feeling lonely sometimes or always, a 7-point rise year over year, with loneliness concentrated among younger workers: 73% of 18-22 year-olds and over 80% of employed Gen Z respondents. Lonely workers reported missing work twice as often due to illness and five times as often due to stress.
What it means for practice
This is self-reported survey data collected pre-pandemic, so it should be read as a baseline rather than a current figure; the absence rates are the part most useful to employers building a case for intervention, since they tie loneliness directly to measurable workplace cost. It does not establish causation, only association, so treat the illness/stress absence multipliers as a planning signal rather than a guarantee of return on any specific programme.
2018
A Connected Society: A Strategy for Tackling Loneliness
This 2018 strategy from the UK Department for Digital, Culture, Media & Sport was the first national loneliness strategy published by any government. It committed to embedding loneliness measurement into Office for National Statistics data collection and funded social prescribing as a delivery mechanism.
What it means for practice
As a policy document rather than an evaluation, it establishes government commitment and infrastructure rather than evidence of impact; readers assessing social prescribing outcomes should look elsewhere for that, and treat this as the origin point of the measurement and funding architecture that later work builds on.
Loneliness and Social Connections: A National Survey of Adults 45 and Older
A 2018 AARP Foundation national survey of 3,020 U.S. adults aged 45 and older, using the 20-item UCLA Loneliness Scale, found that one in three respondents report being lonely. Network size and diversity, along with physical isolation, were the strongest predictors. Loneliness rates varied sharply with neighbour contact: 33% among those who have spoken to neighbours versus 61% among those who never have.
What it means for practice
The use of the standard UCLA scale makes this dataset directly comparable to academic loneliness research, which is unusual for a survey of this size. The neighbour-contact finding is a useful anchor point for anyone designing community-level interventions, though the survey establishes association, not causation.
2015
Loneliness and Social Isolation as Risk Factors for Mortality: A Meta-Analytic Review
This meta-analysis found that social isolation, loneliness, and living alone were each associated with increased odds of early mortality (odds ratios of 1.29, 1.26, and 1.32 respectively). These effects held after adjusting for existing health status, and social deficits showed stronger predictive power in samples with an average age under 65.
What it means for practice
This is one of the foundational quantitative anchors for arguing that social connection is a health determinant in its own right, not a proxy for poor health. The age finding is worth flagging to readers who assume loneliness interventions are primarily an older-adult concern: the evidence here points the other way.
2010
Social Relationships and Mortality Risk: A Meta-analytic Review
This 2010 meta-analysis pooled 148 studies covering 308,849 participants and found that stronger social relationships were associated with a 50% increased likelihood of survival over the follow-up periods studied. The authors report this effect size as comparable in magnitude to other well-established mortality risk factors.
What it means for practice
This is the reference point most social prescribing and connection-focused interventions cite to justify their premise; if you are building a case for funding or programme design, know that it establishes association across a large pooled sample, not causation, and it does not tell you which types of intervention reliably increase relationship strength.