Farnflower Newsletter
Connect

Do friends help you live longer?

Which social ties went with longer survival, why a wide social life showed a stronger link than marriage alone and what friends may do to blood pressure.

How connected do you feel? Take the one-minute test

What counts as social

Being woven into a wide social life went with better survival than being married or living with someone.

Odds of surviving during follow-up by type of social tie (no difference = 1.00), 148 studies; the living-with-others value is not clearly different from 1.00

  1. Broadly integrated in social life 1.91
  2. Large social network 1.45
  3. Feeling supported 1.35
  4. Married (vs. not married) 1.33
  5. Living with others 1.19
Farnflower Source: Holt-Lunstad et al., 2010 · 308,849 people

Probably. People with stronger social ties were more likely to survive during the years they were followed in a large review of the research, and the link was clearest for people woven into a wide social life, not for those who were simply married or living with someone. Because these are observational studies, they show a strong and consistent link rather than proof that friends themselves add years.

Key takeaways

  • Having many kinds of ties at once went with a better chance of survival than any single tie.
  • Living with someone showed the weakest and least certain link.
  • Feeling that support is there if you need it showed a clear link, while the support people said they had received didn’t.
  • Better-connected people had a lower risk of unhealthy blood pressure, waist size and inflammation in both early and later life.
  • Ties of several kinds are the part of a social life you can widen.

What counts as a social tie in the research?

Social ties in the research include structural ones, such as marriage and network size, and functional ones, such as feeling supported, and the type made a difference. A team led from Brigham Young University pooled 148 studies with 308,849 people and sorted the measures into two families. Structural measures count what exists: marriage, the size of your network, how many groups you belong to, whether you live alone. Functional measures ask what those ties do for you: support you received, support you believe you could get and loneliness (Holt-Lunstad et al., 2010).

The strongest link with survival came from complex measures of social integration, which combine several ties at once, such as a partner, friends, and taking part in groups and activities. Simply living with someone showed the weakest link, and it wasn’t statistically clear. Being married sat in the middle (Holt-Lunstad et al., 2010).

I find this the most practical finding in the field. A single relationship, however good, is a narrower safety net than a web of people, places and roles, and that web is the part you can widen.

Does feeling supported matter more than getting help?

The research can’t say for sure. Believing that support would be there if you needed it went with better survival, while the support people reported actually receiving showed a link that wasn’t statistically clear, and the review didn’t test the two against each other (Holt-Lunstad et al., 2010). One possible reading is that people who receive a lot of help are often the ones who are already ill, which would hide the benefit in the data. The review doesn’t test that, so it stays a guess.

How could friends affect your body?

Social ties are linked to measurable things such as blood pressure and inflammation. Researchers at the University of North Carolina used four nationally representative US samples covering adolescence to old age, compared stage by stage, which measured blood pressure, waist size, body mass index and C-reactive protein, a marker of inflammation. In both early and later life, people who were more socially integrated had a lower risk of unhealthy values, and the more connected they were, the lower the risk. Among teenagers, social isolation was linked to inflammation about as strongly as physical inactivity, and in old age, isolation was linked to high blood pressure more strongly than diabetes (Yang et al., 2016).

These are links in observational data, and in my view the authors use stronger causal language than their design can carry. The 2010 review suggests one everyday route: relationships can encourage or model healthy behavior and give people roles and purpose (Holt-Lunstad et al., 2010). What the blood pressure numbers mean is explained in What blood pressure numbers mean, and which habits lower them.

The survival link is consistent across studies, but it can’t show cause. The 2010 review found that the link held across age, sex, initial health, cause of death and length of follow-up (Holt-Lunstad et al., 2010). But nobody can randomly assign people to have friends, and illness can shrink a social life as easily as a small social life can harm health. How much of the effect remains once health, habits and income are counted is explained in Is loneliness as harmful as smoking?

What I’d do

  • Add one regular group activity to your week, such as a class, a club or volunteering.
  • Keep a standing date with one friend, the same day each week or month.
  • Invest in more than one relationship, so your net doesn’t hang on a single person.
  • Tell someone when you need help, so the people around you know they can be there. Outlook may count too: Do optimists live longer?
  • Check how connected you feel with the three questions researchers use in the test How connected do you feel?

Frequently asked questions

Does online contact count? The 2010 review didn’t look at online contact, so it can’t answer that (Holt-Lunstad et al., 2010).

Does volunteering count as social connection? Taking part in groups and activities is part of the social integration measures that showed the strongest link. Whether volunteering itself helps is explained in Does volunteering keep you healthier?

Do pets count? They weren’t part of these studies. What dog ownership shows on its own is in Do dog owners live longer?

Sources

  • Holt-Lunstad, J., Smith, T. B., & Layton, J. B. (2010). Social relationships and mortality risk: A meta-analytic review. PLoS Medicine, 7(7), e1000316. https://doi.org/10.1371/journal.pmed.1000316
  • Yang, Y. C., Boen, C., Gerken, K., Li, T., Schorpp, K., & Harris, K. M. (2016). Social relationships and physiological determinants of longevity across the human life span. Proceedings of the National Academy of Sciences, 113(3), 578–583. https://doi.org/10.1073/pnas.1511085112

One question a month, answered properly