Religious services at churches, temples and mosques boast various features that can be beneficial to health — meditation, a social network, a set of values that discourage smoking, infidelity and other unhealthy behaviors. Many of the studies have found that the health benefits of weekly attendance accrue more heavily to women than to men, perhaps because women make greater use of religious social networks.

Of course, people who attend weekly religious services are by definition well enough to get out of the house regularly, suggesting that they may enjoy an inherent health advantage. Indeed, studies show no health advantage for people who watch religious services on television.

But it isn't simply that people showing up for church are healthier; they also are more likely to improve their health habits. When compared with nonweekly attendance, "weekly attendance was associated with a statistically significant improvement in quitting smoking, becoming often physically active, becoming not depressed, increasing the number of individual personal relationships and getting married," said one of the examined articles, which was published in the Annals of Behavioral Medicine in 2001. That study gathered health and mortality data over a period of 30 years on 2,676 Californians, 26 percent of whom attended religious services weekly.

Not everyone is convinced that religious services account for the more robust health and survival documented in these articles. The same health benefits could be derived from belonging to a bingo club or socializing at the local library, says Emilia Bagiella, a Columbia University assistant professor of biostatistics. Also, she says, "it's hard to correct for the fact that people who go to church may have a better health status" before they arrive.

But the studies supporting a link between religious-service attendance and health come from such secular institutions as the universities of Texas, Michigan and California at San Francisco. And their authors don't necessarily go to church or perceive the mortality benefits of doing so as the handiwork of God.

"Being religiously involved can confer certain health benefits, and I don't think there's any divine intervention involved," says Robert A. Hummer, a non-churchgoing University of Texas sociology professor whose studies have shown a health benefit for regular religious-service attendance.

Moreover, Powell says that she and her colleagues excluded from their review any study that failed to control for the social benefits of church attendance as well as the healthier habits of those who go regularly. Even after excluding those factors, they found a significant health and mortality benefit from regular attendance. "There's an unknown mechanism" contributing to the benefit, she says, adding that she doesn't believe that that mechanism is God.

Powell says that a continuing study of hers is suggesting that that mechanism might be the practice, encouraged in nearly all religions, of turning to prayer or meditation in moments of anger and distress, thereby diminishing the harmful effects of negative emotion. She tells of a Sikh cab driver who told her that any time another driver cuts him off, he reaches for his prayer beads. In doing so, he told her, "I feel closer to God."