Having few friends and family and little social contact appears to increase the risk that heart disease patients will die from a heart attack, a Swedish researcher reported Saturday.

Dr. Kristina Orth-Gomer of the Karolinska Institute in Stockholm said a 10-year study of middle-aged Swedish men found heart disease patients who went on to die were much more likely to suffer from "social isolation.""To be socially active appears to be an important factor for survival," said Orth-Gomer in an interview before presenting her findings at a meeting of The Society of Behavioral Medicine.

Orth-Gomer and her colleagues studied 150 men ages 40 to 65 from 1976 to 1987, including 50 healthy men, 50 men with heart disease and 50 men at high risk for developing heart disease.

The researchers tested the men's physical health in a variety of ways and examined a range of psychosocial factors with the aim of seeing which factors were most strongly associated with those who eventually died of heart disease.

After 10 years, 37 of the men had died from heart disease. Of those who died, nearly all had been initially categorized as suffering from "social isolation," she said.

In fact, social isolation was as strong as the strongest physical factor having an irregular heartbeat in predicting who would eventually die from heart disease, she said.

The findings held true even after the researchers took into consideration how much the men smoked, drank alcohol and other potentially unhealthy activities, she said.

The researchers are uncertain why being socially isolated would increase the risk for dying from a heart attack, she said. But being socially isolated may be stressful, which can increase the risk for heart disease.

The findings indicate that, especially for those already at high risk for a heart attack because they have heart disease, doctors may want to try to decrease patients' levels of social isolation, she said.

In fact, Orth-Gomer said the reason treatments aimed at alleviating so-called Type A behavior decrease the risk for a heart attack may be that they decrease the amount of social isolation.

"Perhaps it's possible some of the interventions going on in Type A behavior have a component of social isolation and with Type A modification there is a positive effect on social support," she said.