Older people

Social isolation increases death risk in older people

"Social isolation is associated with a higher risk of death in older people regardless of whether they consider themselves lonely," BBC News has reported.

Previous research has suggested that people who have limited social contact are at increased risk of death. Many researchers have suggested that this is possibly due to the emotional effects of isolation – that feeling lonely is bad for health.

But this new, large UK study implies that the relationship between social isolation, feeling lonely and risk of death is more complex. The researchers looked at whether these factors were linked in a large group of UK adults aged 52 or above.

The study found that both social isolation and loneliness were associated with increased risk of death. However, if demographic factors and initial health were taken into account, loneliness was no longer significantly associated with risk of death. There was still a significant link between social isolation and risk of death, however, after these other factors and even loneliness had been taken into account.

This suggests that factors other than loneliness – such as having no-one to check on a person’s health – may contribute to increased risk of death.

Efforts to reduce social isolation are likely to have positive outcomes for wellbeing and mortality rates.

Where did the story come from?

The study was carried out by researchers from University College London. The English Longitudinal Study of Ageing, from which the study participants were taken, was funded by the National Institute on Aging and a consortium of UK government departments co-ordinated by the Office for National Statistics.

The study was published in the peer-reviewed journal Proceedings of the National Academy of Sciences of the United States of America (PNAS).

This article is open access, meaning that it can be accessed for free from the PNAS website.

In general, the story was well reported by the UK media. But The Daily Telegraph’s headline on “The toll of loneliness” falls into the trap of confusing loneliness and social isolation. This is precisely the distinction the researchers were trying to make. You can be socially isolated without feeling lonely and you can feel lonely even when surrounded by people.

The Telegraph's mistake is understandable given that it is possible that there is an intricate link between loneliness and social isolation that this study has not been able to evaluate fully.

What kind of research was this?

This was a cohort study. It aimed to determine whether there is an association between social isolation and loneliness, and death from any cause, in a representative sample of the UK population.

The researchers also aimed to determine whether loneliness is partly responsible for the association between social isolation and mortality.

A cohort study is the ideal type of study to address this question.

However, a cohort study cannot show causation. There is still a possibility that other factors (confounders) explain the relationship seen.

There is likely to be an intricate link between social isolation and loneliness. It is difficult to tell whether the methods used by the researchers fully accounted for the complexity of the association.

What did the research involve?

This study included 6,500 men and women aged 52 or over who were part of the English Longitudinal Study of Ageing between 2004 and 2005. The researchers assessed social isolation using a social isolation index, assigning one point for each marker of isolation, for example:

  • unmarried/not cohabiting
  • less than monthly contact with family and friends
  • non-participation in ‘civic organisations’ (such as social clubs or religious groups)

They assigned an overall isolation score on a scale of 0 to 5.

Loneliness was assessed with the three-item short form of the revised UCLA (University of California, Los Angeles) loneliness scale. One example question was “How often do you feel you lack companionship?”. The response options were:

  • hardly ever or never
  • some of the time
  • often

The overall loneliness score ranged from 3 to 9. Participants who scored in the top 20% were defined as being socially isolated or lonely, respectively.

Death from any cause was monitored up to March 2012 (mean follow-up 7.25 years).

The researchers looked at the association between social isolation or loneliness and death from any cause.

The researchers adjusted their analyses for the following confounders:

  • age
  • sex
  • demographic factors (such as wealth, education, marital status and ethnicity)
  • baseline health indicators (including long-standing illness, mobility impairment, cancer, diabetes, chronic heart disease, chronic lung disease, arthritis, stroke and depression)

What were the basic results?

By March 2012, 918 participants had died.

  • Mortality was higher among more socially isolated and more lonely participants.
  • Social isolation was significantly associated with mortality (hazard ratio (HR) 1.26, 95% confidence interval (CI) 1.08 to 1.48) after adjusting for demographic factors and baseline health indicators.
  • Loneliness was not significantly associated with mortality (HR 0.92, 95% CI 0.78 to 1.09) after adjusting for demographic factors and baseline health indicators.
  • The association of social isolation with mortality was unchanged when loneliness was adjusted for (HR 1.26 95% CI 1.08 to 1.48).

How did the researchers interpret the results?

The researchers conclude that “although both isolation and loneliness impair quality of life and well-being, efforts to reduce isolation are likely to be more relevant to mortality”.


This cohort study found that social isolation in older people was associated with increased risk of death from any cause in the UK, and this relationship was independent of demographic factors and baseline health.

It also found that loneliness, which is often thought to be a result of social isolation, is not the reason why social isolation is linked with risk of death.

This study included a large group of people representative of the UK population. The researchers took into account demographic and health factors. However, this was a cohort study and, as such, it cannot show causation. There may be other factors that explain the relationship seen (confounders), that cannot be excluded.

A particular difficulty with this piece of research is that feelings of social isolation and loneliness are highly subjective. It is not possible to say whether they have been satisfactorily assessed by the methods used in this study.

The researchers assessed isolation by creating a social isolation index and giving a score for certain factors. However, these particular factors may not have been relevant to the individual being assessed and may not accurately represent how isolated they feel. For example, the researchers report that they gave equal weight to all social contacts, whereas some relationships may be more important than others.

Similarly, loneliness was assessed using a three-item scale and it is not possible to know whether this could accurately assess loneliness. Overall, there is likely to be an intricate link between the subjective experiences of social isolation and loneliness, which the objective methods used in this study have not been able to evaluate fully.

Nevertheless, efforts to reduce the social isolation of older people are likely to have positive outcomes for wellbeing, and this research suggests that they could also reduce mortality.

NHS Attribution