Nearly as many said that on occasion the influence is attributable to divine intervention, said Farr A. Curlin, M.D., of the University of Chicago, and colleagues. Yet only a few said that these beliefs change "hard" medical outcomes.

"We find it notable, particularly in light of perennial discussions about the relationship between science and faith, that most physicians apply medical science while maintaining a belief that God intervenes in patients' health," said Dr. Curlin and colleagues.

The survey also found that the physicians' perceptions were strongly influenced by their own religious convictions. "Patients are likely to encounter quite different opinions about the relationship between their religion and spirituality and their health, depending on the religious characteristics of their physicians," the authors wrote.

Although many patients draw on prayer and other religious resources to manage the spiritual challenges that arise from illness, controversy has remained about whether, and to what extent, religion and spirituality help or harm patients, Dr. Curlin and colleagues said.

To study this relationship, the researchers mailed a cross-sectional survey in 2003 to a stratified, random potential sample of 2,000 practicing U.S. physicians, 65 or younger, representing all specialties.

Physicians were asked to estimate how often patients mentioned religion and spiritual issues, how much these issues influenced health, and in what ways the influence manifested itself.

The survey also included questions to determine the physicians' own religious characteristics, general observations, and interpretations of religion.

Among eligible physicians, the response rate was 63% (1,144 of 1,820), and the average age of the physicians was 49. Most physicians (56%) believed that spiritual issues had much or very much influence on health, while 54% believed that at times a supernatural being intervenes, the researchers reported.

However, although 85% of the physicians believed that the influence of spirituality is generally positive, only 6% perceived that these beliefs often changed "hard" medical outcomes.

Rather, the study found that 76% of the physicians believed that spirituality helps patients cope, 74% said that it gives patients a positive state of mind, while 55% reported that spirituality and religion provide emotional and practical support via the religious community.

Only 7% of the physicians said that spirituality often causes guilt, anxiety, or other negative emotions, while 2% said it leads patients to decline medically indicated therapy, and 4% reported that patients use it to avoid responsibility for their own health. Finally, about one-third said it can have these harmful influences sometimes.

The physicians' observations and interpretations were strongly influenced by their own religious beliefs, the researchers said.

Compared with those with low religiosity, highly religious physicians were substantially more likely to report that patients often mention spiritual issues (36% versus 11%; P<0.001).

They were also more likely to believe that religion and spirituality strongly influence health (82% versus 16%; P<0.001), and to interpret the influence of religion and spirituality in positive rather than negative ways, the researchers found.

These associations persisted in multivariate analyses that controlled for religious affiliation, region of practice, age, sex, ethnicity, and specialty.

In further analyses, comparing physicians with religious affiliations with those with no religious affiliation, Protestant physicians were more likely to report that their patients bring up spiritual issues and are more likely to believe that God intervenes, that spirituality helps patients cope, and sometimes prevents hard medical outcomes.

Catholic physicians put their faith in God's intervention first and also agreed that belief helps patients cope. They were less likely to say that belief causes negative emotions.

Physicians of other religious affiliations were more likely to report that their patients bring up spiritual issues, that God intervenes, and that spirituality strongly influences health and sometimes prevents hard medical outcomes.

Finally, physicians who practiced in the South, followed by the Midwest, were more likely to report that their patients often mention religious beliefs, with those in the West and Northeast not as likely to do so.

This survey indicated, said Dr. Curlin and colleagues, that religious issues may influence end-of-life care in which some patients and families express hopes for miracles. Because religious physicians may be more likely to share such hopes, further study is needed to explore how these differences may affect the care patients receive.

As a cross-sectional survey, this study was not able to explain why religious and non-religious physicians differed so markedly in their observations and interpretations, the researchers said.

Yet it is possible, they said, that other factors being equal, physicians with different religious or secular commitments may interpret the same evidence in different ways. What the secular physician may not notice or ignore, the religious physician may emphasize or exaggerate.

The study had important limitations, the investigators wrote. Although the study had a better-than-average response rate and there was no substantial evidence to suggest response bias, religious and other characteristics may have affected physicians' willingness to respond in unmeasured ways.

There may also have been other ways to define physicians' religiosity. However, the analyses found similar relationships for frequency of attendance at religious services and self-reported religiousness, the investigators said.

Limitations, notwithstanding, the investigators said, these findings challenge any attempt to create a single interpretation of the relationship between religion and health. The study lends support to recommendations by the Association of American Medical Colleges that physicians recognize how their own beliefs affect the way they provide care for their patients.

"Future studies should examine the ways physicians' religion (and secular) commitments shape their clinical engagements in these and other domains," Dr. Curlin concluded.

No financial disclosures were reported. The study was funded by the Greenwall Foundation of New York, the Robert Wood Johnson Clinical Scholars Program, and the National Center for Complementary and Alternative Medicine.