In the study, researchers from the Human Population Laboratories of the Public Health Institute and the California Department of Health Services, and from the University of California, Berkeley, found that people who attended religious services once a week had significantly lower risks of death compared with those who attended less frequently or never, even after adjusting for age, health behaviors and other risk factors. The study will be published April 4 in the International Journal of Psychiatry in Medicine.

"We found this difference even after adjusting for factors such as social connections and health behaviors, including smoking and exercising," said Doug Oman, lead author of the study and a lecturer at UC Berkeley's School of Public Health. "The fact that the risk of death by several different causes is lower for those who attend religious services every week suggests that we should look to some psychological factor for answers. Maybe frequent attendees experience a greater sense of inner peace, perhaps because they can draw upon religious coping practices to help them deal with stressful events."

Oman conducted the study while he was a research scientist at the Public Health Institute and a post-doctoral researcher at UC Berkeley's School of Public Health.

Researchers specifically looked at the risk of death from certain diseases, including cancer and heart disease. While they found no significant difference for risk of death by cancer, they did find that people who attended religious services less than once a week or never had a 21 percent greater overall risk of dying, as well as a 21 percent greater risk of dying from circulatory diseases.

There was also a strong trend towards lower mortality from respiratory and digestive diseases, although there is more of a possibility that chance might have played a role in producing those results. When compared with weekly attendees, those who attended less than weekly or never had a 66 percent greater risk of dying from respiratory diseases and a 99 percent greater risk of dying from digestive diseases.

Adherents to Christian religions made up the bulk of the study participants, with 51.9 percent reporting themselves to be liberal Protestants. Twenty-seven percent of the participants were Roman Catholics and 2.5 percent were Jewish. Members of other Western religions made up 7.2 percent of the group, while those practicing non-Western religions made up 0.8 percent of the group. Those with no religious affiliations comprised 10.6 percent of the study participants.

Oman said there are still unanswered questions he hopes will be addressed in future studies, including the significance of spirituality or devoutness. "Several non-Western religions, including Buddhism, place less emphasis on going to a temple or church," he said. "So people of those faiths may be just as devout in their tradition, and that may revolve around a household shrine. They may go to a temple only a few times a year, but they could still be getting the psychological benefits of inner peace."

Still, Oman said this study adds to a growing body of evidence that religious practices are generally linked to better health. He also pointed out that this study is one of the few that has investigated the relationships between religious involvement and several specific causes of death.

Prior studies of Mormons and Seventh Day Adventists noted that beneficial health behaviors, such as abstention from smoking and high fat diets, were strongly integrated in their beliefs. Past studies have also associated positive health effects with the high level of social support found in religious communities.

"The picture that is developing is that religious activity is affecting health through several pathways," said Oman. "Whether it is encouraging better health habits such as exercising, providing a strong social support network, providing a sense of psychological well-being, or all those factors combined, it seems clear that the effects of faith deserves more study."

The researchers used data taken over 31 years from a longitudinal survey of 6,545 adult residents from California's Alameda County. Researchers from the state health department began the survey, which is ongoing, to study the interrelationships among health status and social, familial, environmental and other factors.

Although the study only looked at residents of Alameda County, Oman said the area's ethnic diversity and mix of rural, suburban and urban neighborhoods make the findings relevant to communities throughout the United States.

Data was obtained through mailed questionnaires or through interviews in the participants' homes. Participants aged 21 and over at the start of the study in 1965 were questioned again in 1974, 1983 and 1994. Death records were obtained from state and federal mortality files through 1996.

Co-authors of the study are William Strawbridge and Richard Cohen from the Public Health Institute, and John Kurata from the California Department of Health Services.