A national survey of 2,042 people in Wales found that 54.7% of respondents experiencing both limited social contact and feelings of disconnection reported low life satisfaction. That compared with 5.6% of people classified as socially connected, according to research published on 13 October 2026 in Frontiers in Public Health.
The difference is striking, but it does not establish that isolation directly caused poorer well-being. The study analysed responses collected at one point in time, so the direction of the relationship cannot be determined.
Its important contribution is to separate two experiences that are often treated as interchangeable: having relatively few social contacts and feeling disconnected from other people. The results suggest that the combination may identify a particularly vulnerable group.
Researchers at Public Health Wales and Bangor University examined life satisfaction, feelings that life is worthwhile, happiness and anxiety. They found that people experiencing both forms of isolation generally reported the poorest outcomes across all four measures.
Why being alone and feeling lonely are not the same
Someone can spend considerable time alone and still feel supported and connected. Another person can interact with colleagues, neighbours or family members regularly while feeling that those relationships do not meet their emotional needs.
The researchers call the first dimension objective social isolation: a measurable shortage of social contact or participation. The second is subjective social isolation: the feeling of being disconnected from others, often discussed in relation to loneliness.
Distinguishing these experiences matters for public health policy. Programmes designed solely to increase the number of social activities may not necessarily address the emotional experience of loneliness, while counselling or peer support may not solve practical barriers to meeting other people.
The study sought to determine whether each type of isolation, and their combination, had different associations with well-being in a population covering a broad age range rather than only older adults.
How the national survey was organised
The researchers analysed the February 2025 wave of the Time to Talk Public Health panel, a survey programme established by Public Health Wales to inform policy and practice. Residents aged 16 and older could participate through online, telephone or face-to-face methods.
The original survey involved 2,137 respondents. After excluding responses with missing or unusable information needed for the analysis, the researchers retained 2,042 participants. This distinction is important because the article’s headline findings refer to the analytical sample.
The panel used recruitment and quota targets designed to improve demographic coverage. Survey organisers also conducted in-person interviews and targeted recruitment to reach groups that can be underrepresented, including younger people and men.
Among the final analytical sample, 54.8% were women, 52.8% were aged 16 to 49 and 39.6% reported a disability under the study’s definition. Most respondents, 95.9%, were recorded as white, an important consideration for generalising to minority populations.
The researchers used statistical tests to compare groups and logistic regression to examine associations after accounting for selected demographic factors. Such adjustments can reduce some confounding but cannot remove all differences between people or establish causality.
How social isolation was measured
Objective isolation was assessed using three questions. Respondents received one point each for living alone, seeing friends or family less than monthly, and taking part in no weekly social activities from a specified list.
People with a score of at least two out of three were classified as objectively isolated. This rule offers a consistent definition for analysis, but it cannot capture every meaningful aspect of someone’s relationships.
Subjective isolation was assessed by asking how connected respondents felt to other people generally. Those answering that they felt connected only a little or not at all were classified as subjectively isolated.
The team then divided respondents into four mutually exclusive groups: neither form of isolation, objective isolation only, subjective isolation only, and both objective and subjective isolation.
Well-being was measured using four established personal well-being questions covering life satisfaction, the sense that life is worthwhile, happiness and anxiety. These were self-reported ratings rather than clinical diagnoses.
Two-thirds were classified as connected
Of the 2,042 respondents, 66.9% were in the connected group. A further 20.8% experienced subjective isolation without meeting the study’s threshold for objective isolation.
Only 4.9% were classified as objectively isolated without feeling disconnected, while 7.3% experienced both forms. These four categories add to approximately 100%, with small differences attributable to rounding.
The pattern illustrates why counting social contacts alone may miss a large group. People reporting subjective isolation were substantially more common than those who had objectively few connections without feeling isolated.
The researchers found statistically significant differences in isolation categories by age and disability status. These relationships should be understood as patterns within the survey, not as proof that age or disability directly produces isolation.
Low life satisfaction was especially concentrated
The clearest contrast concerned life satisfaction. Among people experiencing both objective and subjective isolation, 54.7% reported low life satisfaction. Among respondents who were neither objectively nor subjectively isolated, the figure was 5.6%.
That is a difference of 49.1 percentage points between the two groups. It does not mean that becoming isolated raises an individual’s probability of low life satisfaction by exactly that amount, because the comparison is observational.
Across the full sample, 15.1% reported low life satisfaction. Other poor well-being outcomes were also present: 10.7% had a low sense that life was worthwhile, 16.3% reported low happiness and 32.7% reported high anxiety.
In the unadjusted comparisons, respondents experiencing both forms of isolation consistently had the highest proportions of adverse outcomes. The researchers also used adjusted analyses to examine whether this overall pattern remained when selected demographic characteristics were taken into account.
The central message is not that everyone with few social contacts experiences poor well-being. Instead, subjective disconnection appears especially important, and combining it with objectively limited contact may identify greater vulnerability.
Middle-aged adults stood out
The share experiencing both types of isolation varied considerably by age. Among respondents aged 50 to 69, 12.2% were in the combined-isolation group, compared with just 1.1% of those aged 16 to 29.
The corresponding figures were 7.5% for ages 30 to 49 and 6.2% for those aged 70 and older. This is noteworthy because public discussion often frames social isolation primarily as a problem of old age.
The survey instead suggests that some middle-aged adults may be particularly exposed to the combination of limited contacts and subjective disconnection. Responsibilities involving work, caring or changes in household circumstances might be relevant, although the study does not demonstrate which mechanisms explain the pattern.
Among 16- to 29-year-olds, subjective isolation alone was much more common than combined isolation. The findings therefore caution against assuming that a younger person who regularly meets others necessarily feels socially connected.
Disability was linked to a larger difference
Among respondents reporting a disability, 13.3% experienced both objective and subjective isolation. The comparable figure among people without a reported disability was 3.4%.
The survey also found that 25.1% of people with disabilities experienced subjective isolation alone, compared with 18.0% of those without disabilities. The differences point to the importance of accessibility and inclusive opportunities for connection.
People with disabilities were also more likely to report low life satisfaction: 26.8%, compared with 7.4% among those without disabilities. These statistics describe overlapping vulnerabilities rather than proving one factor causes another.
Physical access, transport, financial constraints, discrimination and health-related limitations could plausibly affect social participation. However, this particular analysis cannot quantify the contribution of each possible barrier.
What the findings could mean for support services
The authors argue that public health responses should consider both the quantity of people’s connections and their subjective experience of those relationships. A single question about whether someone lives alone would miss much of the issue.
Health and community services could ask whether individuals have opportunities to meet others and whether those opportunities provide a meaningful sense of belonging. Different answers may point to different forms of assistance.
For someone with limited social participation, accessible transport, community programmes or practical support may help create opportunities for contact. For someone who is frequently around people yet feels disconnected, improving relationship quality or offering appropriate psychosocial support may be more relevant.
These are plausible applications of the research, not interventions tested by the survey. The study did not randomly assign participants to programmes or measure whether any particular support service improved outcomes.
A careful approach would avoid treating solitude itself as a problem. Some people prefer substantial time alone and do not experience it as distressing. The concern is the combination of limited connection and poorer well-being, especially when people want more support.
Why the study cannot establish cause and effect
Because the survey was cross-sectional, it measured isolation and well-being during the same period. People may experience poorer well-being because they feel disconnected, but low mood or anxiety could also make social participation more difficult.
Other factors may influence both conditions. Financial insecurity, chronic illness, bereavement, housing circumstances and access to transport could affect relationships and personal well-being simultaneously.
Although the researchers adjusted for selected demographic characteristics, their analysis cannot eliminate all unmeasured confounding. Reported percentages should therefore not be interpreted as the effects of changing social contact.
All isolation and well-being measures relied on self-report. Respondents may interpret social connection differently, and the threshold used to classify objective isolation is only one of several possible definitions.
The sample was designed for broad national coverage but was not a simple random sample of every resident in Wales. Some groups remained relatively small, and complete-case exclusions could introduce additional differences.
These limitations do not erase the observed associations. They define what can be responsibly concluded: people experiencing both forms of isolation reported worse well-being, but the study cannot determine why or what intervention would work best.
What future research should examine
Longitudinal studies could follow people over time to see whether changes in social contact precede changes in life satisfaction, happiness or anxiety. This would offer stronger evidence about direction, although it would still require careful analysis of confounding.
Intervention trials could compare approaches focused on increasing contact, improving the quality of relationships or combining both. Researchers could then measure not only attendance or participation but also meaningful changes in subjective connection and well-being.
Further work should examine people with disabilities and those in middle age, given the higher proportions of combined isolation observed in these groups. Larger and more diverse samples could also support more reliable analysis across ethnic and socioeconomic groups.
The study reinforces a simple but consequential distinction: being around other people and feeling connected to them are not the same. Public health programmes that recognise both dimensions may be better equipped to identify people who need support.
Source Information
Study title: The relationship between objective and subjective social isolation and well-being: a national cross-sectional survey in Wales.
Authors: Carys Dale, Sara Wood, Catherine A. Sharp and Karen Hughes.
Journal: Frontiers in Public Health, volume 14, Public Mental Health.
Publication date: 13 October 2026.
Study design: Cross-sectional analysis of the February 2025 Time to Talk Public Health panel survey in Wales, with statistical comparisons and adjusted logistic regression.
Participants: 2,042 residents aged 16 or older included in the final analysis, from 2,137 survey respondents.
Key results: 66.9% connected, 20.8% subjective isolation only, 4.9% objective isolation only and 7.3% both; low life satisfaction was 54.7% in the combined-isolation group versus 5.6% among connected respondents.
Original DOI: 10.3389/fpubh.2026.1918382.
