Social links may counter health risks.

Certain health patterns in the United States stand out clearly. Death rates rise sharply among people age 65 and older, for instance, and hypertension plagues many black adults.

These trends have been attributed to populationwide factors such as the greater susceptibility to physical illness among elderly people and the poverty and discrimination that disproportionately afflict blacks. Now, two studies find that, on closer inspection, death and high blood pressure take their greatest toll on those elderly and black people, respectively, who maintain few social contacts.

Both investigations appear in the September Epidemiology.

"These papers give us a better picture of how social relationships help to maintain the health of people who in other ways may experience disadvantage," writes Lisa F. Berkman of the Harvard University School of Public Health in Boston in an accompanying comment.

In the first study, elderly men and women who reported having consistently interacted with two or fewer people over a 3-year span died at a much higher rate over the next 8 years than those who had operated in a larger social network. Moreover, the small number of older adults who managed to go from few to many social contacts in the initial 3 years died at the same rate as those who had sustained numerous relationships throughout that time.

Older people whose numerous social ties unraveled dramatically during the same 3-year period did not experience elevated death rates, after statistically accounting for age and preexisting health problems, report James R. Cerhan and Robert B. Wallace, both of the University of Iowa College of Medicine in Iowa City.

"Continued social isolation may be a more important determinant of mortality risk than recent changes in social ties," they suggest.

Their study consisted of 2,575 adults. age 65 to 102 living in two primarily rural Iowa counties. In surveys conducted in 1982 and 1985, each participant described his or her social ties, including marital status, number of close friends and relatives, church attendance, and membership in clubs or other social groups.

Deaths in the sample were then tracked through 1993.

Volunteers who cited persistent social isolation died at a substantially higher rate than the others, taking into account age, education, history of cigarette smoking, symptoms of depression, and changes in physical health, the researchers note.

Biological paths by which a lack of social ties might prove deadly to older people remain unknown, Cerhan and Wallace add. Further work needs to address whether separate factors, such as an introverted or hostile personality, lie at the root of prolonged social isolation, they say.

In the second study, black adults reporting few social contacts and lives dominated by worry and aggravation over largely uncontrollable circumstances exhibited significantly higher blood pressure readings, whether or not they suffered from hypertension.

Chronic mental stress and social isolation may trigger hormonal responses that produce narrower blood vessels and an increase in blood volume. These changes, in turn, raise blood pressure, proposes a research team headed by David S. Strogatz of the University at Albany (N.Y.) School of Public Health.

Strogatz and his coworkers collected data in 1988 from a representative sample of 1,750 black adults age 25 to 50 living in Pitt County, North Carolina, a region noted for high mortality from stroke. Family incomes for participants ranged from the poverty level to relative affluence.

Blood pressure readings were obtained in each volunteer's home. Social support and stress surveys focused on immediate concerns, such as the likelihood of getting help from others on simple house repairs or advice from friends on personal problems and the extent of daily worrying about money and personal safety.

Blood pressure was higher in volunteers who had either little contact with others or lots of worries, but it rose even further in those who reported both, the researchers report. The results held for both sexes.

This potentially deadly duo of social isolation and mental stress appeared most often in men and women from poor households. In addition, the findings applied whether volunteers' blood pressure was in the normal range, in the hypertensive range, or being treated with medication.