A nineteen eighty-eight review of five studies concluded that social relationships predict mortality, and put the effect alongside smoking and blood pressure. That was a strong claim from a small base. Two decades later there was a literature to test it against. Holt-Lunstad, Smith and Layton pooled it. Does the number survive the scale-up?
It does, and it is large. Across one hundred and forty-eight studies covering three hundred and eight thousand participants, people with stronger social relationships had about a fifty percent greater likelihood of survival. The figure held across age, sex, initial health status, cause of death and length of follow-up, which is a stability most risk factors cannot claim.
The quality and quantity of a person’s social relationships has long been linked to mental health, and to illness and death as well. Two decades before this review, a look at five large prospective studies proposed a causal association, and placed social relationships among risk factors like cigarette smoking, blood pressure, blood lipids, obesity and physical activity.
Holt-Lunstad, Smith and Layton set out to measure the extent of the influence, find which aspects of relationships predict most strongly, and identify what the risk. They extracted cause of mortality, initial health status and pre-existing conditions from each study, along with the length of follow-up and the type of assessment used.
Across one hundred and forty-eight studies and three hundred and eight thousand participants, the random effects weighted average came to an of one and a half. That is a fifty percent greater likelihood of survival for people with stronger social relationships. The finding held across age, sex, initial health status, cause of death and follow-up period. What did differ was the type of measurement: complex measures of gave close to double the odds, while a binary indicator of living alone gave little.
The authors place the influence alongside smoking and alcohol consumption, and ahead of physical inactivity and obesity. They also argue the pooled effect may understate the real one, since so many studies used single-item measures of isolation rather than complex ones. Their closing recommendation is addressed to physicians, health professionals, educators and the media: treat social relationships as seriously as the other risk factors that affect mortality.
wat. On the pooled numbers, being socially connected sits in the same range as not smoking, and ahead of being active or thin.
wat wat
The authors conclude the influence is comparable with well-established mortality risk factors, and ask that clinicians and educators treat it with the seriousness those get.
How do we know?
A meta-analytic review in PLOS Medicine pooled one hundred and forty-eight studies, extracting cause of mortality, initial health status, pre-existing conditions, follow-up length and the way each study assessed social relationships.
What is still unknown
Observational studies throughout, so the direction of cause is not settled by the pooling. The authors also note the overall effect may be an underestimate, since many studies used crude single-item measures of isolation.
Learn something?
Where to next
Is being lonely as dangerous as the risk factors doctors already watch?mortality
How fast do adults actually read?meta-analysis
Does an ice bath after training actually speed up recovery?meta-analysis