A Life Has Company

Connection, belonging, and the possibilities of a day

The Arising · Relationships & health · November 2026 · John Griffin, MD

A ride to an appointment. A shared meal. Someone who notices you have not been around. A conversation in which you do not have to explain everything from the beginning.

Much of the practical work of living happens between people. Yet health is often described as a list of things an individual should do: exercise, eat well, sleep, keep the numbers in range. The people and circumstances that make those activities possible can disappear from the picture.

Connection deserves attention in its own right. Its meaning does not depend on turning friendship into a longevity technique.

A large study of UK Biobank participants adds evidence that social circumstances and health are linked. It also shows why a careful reading matters. The researchers estimated differences in years alive without certain recorded illnesses. They did not discover how many years a new friendship will add to someone's life. [1]

Being alone and feeling alone

Loneliness and social isolation are related, but they are not the same experience.

Loneliness concerns how someone experiences their relationships. Isolation concerns aspects of their social circumstances, such as contact with other people or participation in activities. A person can spend much of the day alone and feel content. Someone surrounded by people can still feel unseen or unable to confide in anyone.

The study used particular questionnaire definitions. Its loneliness category required both frequent feelings of loneliness and confiding in someone less than once a month. Its isolation category required at least two of three circumstances: living alone, seeing friends or family less than monthly, and not taking part in a weekly social activity. [1]

Those definitions help researchers make comparisons. They are not a complete description of a person's relationships, and they are not a diagnosis.

Someone who does not meet the study's loneliness definition may still need support. Someone who meets its isolation definition may not want a busier social calendar. A useful conversation starts with the person's experience, not only the category.

What the researchers followed

The analysis included 277,489 adults who were initially free of seven selected major disease categories. Researchers compared groups reporting neither loneliness nor isolation, one of the two, or both. The group reporting both contained 2,271 people, about 0.8% of the total. [1]

The physical outcomes were type 2 diabetes, cardiovascular disease, chronic respiratory disease, neurodegenerative disease and cancer excluding non-melanoma skin cancer. The mental-health outcomes were recorded depression and anxiety. “Disease-free” in this study therefore means free of those selected outcomes, not free of every illness, symptom or disability. [1]

New diagnoses were identified through hospital inpatient records. That approach can miss illness managed in the community without a hospital admission. The study's measures are shaped by what the records capture, not just by what people experience.

Median follow-up was about 13.4 years for disease outcomes and 13.8 years for deaths. The researchers then used a mathematical model to estimate disease-free life expectancy from age 50 to 100. They did not observe every participant for 50 years. [1]

What the estimates showed

Compared with people reporting neither loneliness nor isolation, those reporting both had fewer modeled years alive without the selected physical diseases: 2.4 fewer years for women and 1.7 fewer for men, starting from age 50. [1]

For the selected mental-health outcomes, the estimated differences were 3.7 years for women and 5.8 years for men. These are separate comparisons. The physical and mental differences should not be added together. [1]

The estimates also had uncertainty. For example, the 2.4-year physical-disease difference for women had a reported range from 1.1 to 3.9 years. The corresponding range for men was 0.8 to 2.7 years. The mental-health ranges were 2.0 to 5.5 years for women and 4.5 to 7.1 years for men. [1]

Researchers call these 95% confidence intervals. They describe statistical uncertainty around the model's group estimates under its assumptions. They are not a range of years that will be lost by a particular person, and they do not account for every possible source of bias.

The useful message is that the study found a substantial association worth taking seriously. The numbers are not a countdown attached to anyone's social life.

What “years free of disorder” does not mean

The wording can sound more definite than the model really is.

The model allowed people to move from being free of the selected diseases to a recorded diagnosis or death, and from diagnosis to death. It therefore accounts for death before diagnosis when estimating years alive without those diagnoses. It did not include a return from diagnosis to a disease-free state. For depression and anxiety, that is especially important: the estimate concerns years alive without a first recorded diagnosis, not years someone is inevitably destined to spend continuously unwell. [1]

People's symptoms can improve, recur or respond to treatment. A model designed this way does not describe that full experience.

Nor does a group estimate tell one person what will happen next. Averages can identify patterns across many lives while remaining a poor script for any single life.

Association is not a prescription

The researchers did not randomly assign people to be lonely or isolated. They observed circumstances and later health records.

Resources, health, mobility, personality and life events can influence both social experience and illness. A health problem may make it harder to leave home before it appears in a hospital record. Financial pressure can affect housing, opportunities to meet people and access to support. Bereavement can alter many parts of life at once.

The analysis adjusted for measured differences including age, ethnicity, deprivation, education and income. Adjustment means trying to account for those differences statistically. It cannot guarantee that every important difference has been measured or that all remaining comparisons represent cause and effect. [1]

Loneliness, isolation and lifestyle were assessed at the beginning. They can change over time. UK Biobank participants also differ from the population as a whole, which limits how directly the exact estimates apply in every community. [1]

The study therefore does not show that joining a club, scheduling a phone call or sending a message will produce a specified number of healthy years. Those actions may matter for many human reasons. Their effects cannot be calculated from this study's headline numbers.

What about healthy habits?

The researchers also examined a baseline lifestyle score incorporating smoking, body mass index, physical activity, diet and sleep. Higher scores were associated with more modeled disease-free years, including among people classified as lonely. [1]

The isolation-related gap was smaller across healthier-score groups. The analogous narrowing for loneliness was not statistically significant. That does not mean healthy habits have no value for someone who feels lonely. It means the study did not establish the same clear pattern of a smaller loneliness-related gap. [1]

Nor did the study test what happens when a particular person changes habits. Comparing groups with different starting scores is not equivalent to running a program and measuring the results.

There was no separate lifestyle analysis for the group reporting both loneliness and isolation because that group was too small. The researchers examined loneliness and isolation separately while accounting for the other measure. [1]

The practical lesson is to resist a single formula. Feeling disconnected and having few contacts can overlap, but one response may not address both.

Make participation possible

When connection becomes harder, a useful starting question is not “Why aren't you socializing more?” It is “What is getting in the way of something you would like to do?”

Transport may be unreliable. A room may be too noisy to follow the conversation. Fatigue may make a long gathering impossible. Stairs, cost, caregiving responsibilities or unfamiliar technology may turn an invitation into work. A familiar role may have disappeared after retirement, illness or bereavement.

These are examples to explore, not assumptions about any particular person. The study did not test a program that fixes them or measure how many years such changes would add.

But they help make the conversation concrete. A quieter place may be more useful than another invitation. A dependable ride may matter more than encouragement. A shorter visit may make participation possible without leaving someone exhausted.

The condition of a body does not completely describe the possibilities of a day. Surroundings and dependable help can change what is available.

The person across the table is not an intervention

Relationships involve mutual care, choice, history and sometimes difficulty. They cannot be reduced to a recommended dose of contact.

More interaction is not always better. A person may need rest, privacy or distance from a relationship that feels unsafe or draining. Respecting those needs is part of connection, not a failure to pursue it.

A useful aim may be one dependable person to talk with, a valued role, a place where someone feels welcome, or the chance to contribute. The right answer belongs to the people involved.

Research can help identify a pattern that deserves attention. It cannot decide what belonging should look like in an individual life.

Three ordinary situations, three different questions

Someone wants to return to a group but has stopped attending. Before adding reminders, ask what changed. Perhaps the meeting time conflicts with care responsibilities or the journey has become difficult. The useful next step might be a different time or practical help with getting there.

Someone has plenty of contact but feels alone. A fuller calendar may miss the point. The question may be whether there is anyone they can speak honestly with, whether a recent loss has been recognized, or whether the relationships feel meaningful to them.

Someone enjoys living alone and is content with a small circle. A study category should not become a demand to change their personality or preferred life. It can still be useful to ask whether support is available when needed, without treating solitude as a problem to correct.

These are illustrative situations, not study participants or tested treatments. They show why the response should fit the person rather than a score.

What belongs in a useful service?

A clinical or educational service can make room for questions about support, belonging and obstacles to participation. The answers should inform a conversation, not label someone as responsible for an unfavorable risk estimate.

For Arising Health, a useful design direction is to include goals such as getting back to a group, caring for someone, contributing to a shared activity or taking part without exhaustion. These are proposals for what a service could pay attention to, not claims that a particular service has already improved health.

Its benefit would need evaluation. Did the person reach a goal they valued? Was the help accessible and welcome? Did the effort create extra burden? Were people with fewer resources able to use it?

A platform becomes more useful when its measurements remain connected to the person's purpose.

Who supported the research?

The paper reports funding from a Fudan University startup grant and the US National Institutes of Health, including its heart, lung and blood and diabetes, digestive and kidney disease institutes. The authors declared no competing interests. [1]

These disclosures provide context. The central limits remain those of the observational design, the measurements and the model, regardless of who funded the work.

A larger picture of health

Longevity is not simply a longer interval in isolation. Its meaning includes who is there, what we can contribute and whether the day contains something worth returning to.

The study offers evidence for taking social circumstances seriously, while leaving the causal questions open. It should not become another reason to blame someone whose life has become smaller or harder.

Sometimes the most useful next step is modest: notice an obstacle, ask what matters, and make one valued activity easier to reach. No promise of extra years is needed to recognize why that might be worthwhile.

A life has company in many forms. The opportunity is to help people find the forms that fit.

Sources and further reading

[1] Xuan Wang, Hao Ma, Yoriko Heianza and colleagues. Associations of loneliness, social isolation and healthy lifestyle with life expectancy free of major physical disease and mental disorder. Nature Communications. 2026;17:6816. Primary publication record. Full text.

Continue exploring

Read Brain health, Fitness and Sleep, or explore the magazine collection.