People with adequate social relationships — friends, family and community involvement — were 50% less likely to die during study periods than those with sparse social support, the authors found. It's an effect comparable to that of quitting smoking.
And, turning it around, people with little social support have a mortality risk equal to alcoholism and even higher than either obesity or physical inactivity, the study found.
The analysis, by researchers at Brigham Young University and the University of North Carolina at Chapel Hill, compiled data from 148 studies. More than 300,000 people were in the data pool, followed for an average of 7.5 years. The link between social support and mortality risk was found for men and women of all ages, regardless of initial health condition, years of a study or cause of death.
In concrete terms, that 50% number means that socially connected people would live an average of 3.7 years longer than less-connected people, says study co-author Timothy B. Smith, a psychology professor at Brigham Young.
Of course, the 50% survival edge is not an absolute number; it was an average across results from all the studies pooled together, says Julianne Holt-Lunstad, associate professor of psychology at Brigham Young and lead author of the review, which was published in the journal PLoS Medicine. Some studies measured social connectedness in simple ways, such as whether a person lived alone. Others used complex measures, such as tallying how many people were included in a person's social network, the extent to which people were involved in a community, a person's perceived degree of loneliness and the extent to which people felt they made a contribution.
The more sophisticated measures found social support was linked to an even higher survival edge.
The study is "very telling," says Dr. Antonio Gomez, assistant clinical professor of pulmonary and critical care medicine at UC San Francisco. Physicians and the public should take note, he says. But it's still not clear what the practical implications are for improving people's health.
The studies, Gomez notes, have their limitations — primarily the trickiness of teasing out cause and effect. Does social connectedness foster good health or are people in good health simply more likely to be socially connected? "We can't make the broad, sweeping claim that social relationships cause increased survivability — at least, not yet," he says.
Gomez adds that the studies don't explain how social contacts could drive good health. And they don't rule out the possibility of unknown differences that may exist between people who are social and those who are not — and that those differences, not the social links that ride along with them, could be the real things driving health outcomes.
Yet there is mounting evidence in the scientific literature that social relationships do affect health.
In one classic 2003 study by Sheldon Cohen of Carnegie Mellon University, a group of volunteers was infected with the cold virus. Results were compelling: People with a greater diversity of social relationships were less likely to develop symptoms of a cold.
In an earlier study by Janice Kiecolt-Glaser, professor of psychiatry at Ohio State University, medical students who described themselves as lonely had poorer immune responses than their colleagues who described themselves as less lonely.
"As humans, we have many different regulatory systems — blood pressure, metabolism, stress hormones," says Teresa Ellen Seeman, professor of medicine at the UCLA School of Public Health. "There are data that suggest all these systems are affected by social relationships. People who report more supportive and positive social relationships have .... lower blood pressure, lower cholesterol levels, better glucose metabolism and lower levels of various stress hormones."
To study the effect of social support on blood pressure, which is a predictor of cardiovascular disease, Bert Uchino, psychology professor at the University of Utah, put a portable blood pressure cuff on study subjects, monitoring their blood pressure throughout the day. During the study period, subjects also filled out diaries. Those who recorded feeling more loved and cared for had lower blood pressures than those who recorded feeling lower levels of support.
"Friends and supportive people can make life easier on a basic, everyday level," Uchino says. "They can lend you money, offer rides or provide baby-sitting. They can also encourage you to have better health practices, see a doctor, exercise more. They may also help you indirectly by making you feel you have something to live for. A good example is a new parent. You might want to take better care of yourself so you can see your daughter graduate from high school."
The emotional support people receive from friends and loved ones "can help you think about problems in ways that decrease their [perceived] severity or even make them non-problems," Uchino adds. "By having a secure relationship and feeling loved, people live much more secure, calm lives."
Gomez maintains there are still questions that must be resolved before the medical system can better help people at medical risk from feelings of loneliness or isolation.
"What type of social relationships are effective?" he asks. "If physicians create an artificial support system, is that better than an organic network that patients create on their own? Another question is: Does Facebook count? Is a social network that is more virtual-based sufficient for survivability, or is it better to have a face-to-face interaction?"
Clinical trials would be the way to answer such questions — randomly assigning a large number of people who are socially isolated to one of two groups: one with intervention and another with no intervention. Researchers could then follow the groups over the years to see how their social connections change and whether there is an effect on their mortality.
But in the meantime, regardless of the unsettled state of the science, there are lessons. "The real message in a study like this is that people who are isolated should be screened for depression and anxiety," Gomez says. Another may be: If you've got good friends, it may not hurt to have more. Holt-Lunstad and colleagues found that the more social resources a person has, the more their survival edge goes up.
"There is very little downside to engaging in social relationships," Gomez says. "If patients do have good social relationships, they should maintain them. If they don't, they should try to foster them."