Researchers found that among 503 women with symptoms of coronary artery disease at the study's start, those with relatively small social circles were more than twice as likely to die over the next two to four years as women with more social ties.

More socially isolated women also had more severe narrowing in their heart arteries at the beginning of the study, as well as higher rates of smoking, high blood pressure and diabetes -- but neither these factors nor rates of depression explained the link between social relationships and death risk.

The one factor that did stand out was income, according to findings published in the journal Psychosomatic Medicine.

Women with the fewest family and social ties were much more likely than other study patients to have an annual income below $20,000, and low income was strongly linked to an increased risk of death over the study period.

Still, that does not mean that income fully explains the link that dozens of studies now have found between social "networks" and health and mortality, according to Dr. Thomas Rutledge of the V.A. San Diego Healthcare System, the lead author of the new study.

The issue is complex, he said in an interview with Reuters Health, and there are probably many factors at work, including the emotional effects of having few social ties.

Low income and social isolation, Rutledge and his colleagues note in their report, probably each increase the risk of the other-and poor health could contribute to each as well.

For the study, the researchers used data on 503 U.S. women taking part in a national study; all had been referred for X-ray tests of their blood vessels due to chest pain or other signs of coronary artery disease. The women completed a battery of questionnaires at the outset, including questions meant to estimate the size of their social circles -- which encompassed relationships with close family, spouses, friends, co-workers and community groups.

Over the next two to four years, women with the smallest social networks were more than twice as likely to die as those with the most social ties, although there were only 30 deaths total in the study group.

According to the researchers, the findings should encourage doctors to consider social and economic factors when treating a woman with suspected heart disease, since each may affect her future health "as powerfully" as traditional heart risk factors.

"Doctors could recommend that a patient gets involved socially -- it doesn't just help with boredom," Rutledge said.

Even having a dog or cat could help, he added, noting that research has shown that pets may help ease stress and lower blood pressure.

However, Rutledge said, it would be too simplistic to "just recommend we all get married and have more friends for our health."

One reason, he noted, is that the quality of relationships, and not merely quantity -- which is what is captured by the social-network measure used in this study -- is also clearly important.

SOURCE: Psychosomatic Medicine, December 2004