News stories, midmorning talk shows, and celebrity figures such as Deepak Chopra claim that prayer can improve health. Biologists, pundits, and others scoff at the same assertion. As is typical of so many debates regarding faith, people often see what they want to see: believers seeing proof of prayer and skeptics seeing disproof. This raises the question: What does the research show?

According to Dr. Harold Koenig, an associate professor of medicine at Duke University and the country's leading authority on faith-and-medicine studies, academic research does show that prayer has beneficial health effects, although mainly for the person who does the praying. Studies of "intercessory" prayer--Person A prays for the health of Person B--find scant if any effect. But studies of "petitionary" prayer, in which a person prays for his or her own health or peace of mind, show tangible statistical results. When you pray for your own health--especially your own mental health, in cases of clinical depression--science suggests you may be on solid ground.

Objective attempts to determine whether prayer has demonstrable effect on health trace to the 1860s, when Francis Galton, a cousin of Charles Darwin, studied the subject. Galton examined English mortality records to determine if ministers, whom he presumed would pray more than the public at large, had longer life expectancies than those in other professions. He found that, at that time in England, the typical cleric lived 69 years, versus 68 years for lawyers and 67 years for physicians. Galton thought this was too small a difference on which to base any claim of prayer benefits.

Galton also studied death records for English aristocracy, whom he assumed would live longest of all if prayers were answered--because standard Anglican liturgy of the period asked parishioners to offer prays for "the nobility" of the nation. But rather than enjoying prayer-conferred longevity, Galton found, English nobles of the 1860s died on average after 66 years. From such numbers, Galton concluded that prayer has no effect.
Although Galton's work was rigorous by the standards of a century and a half ago, it had what today would be considered "control" problems. Galton made no attempt to determine whether ministers actually did receive more prayers than others; he just assumed it. Similarly, he assumed that since the Anglican liturgy asked parishioners to pray for the health of nobility, people must be doing so. But we know from sociology that millions of 19th-century Britons despised the titled class: Maybe churchgoers were silently praying that the local lords and ladies would come to woe.

Today, many skeptics who reject prayer-and-health studies cite Galton, without mentioning the flaws in his work. They also don't mention Galton's bias--he was an ardent foe of religion.


Flash forward to the present and the two most oft-cited, "controlled" scientific attempts to determine if "intercessory" prayer improves health. The first was conducted by a California physician named Randolph Byrd, who in 1983 studied patients entering the coronary care unit of San Francisco General Medical Center. Byrd broke the patients into two groups, a control group for whom no praying was done (at least, so far as Dr. Byrd was aware) and a test group whose names were given to volunteers who had offered to pray for the patients. The test was "blind," meaning the patients did not know which group they were in so there could be no placebo effect of feeling better simply because you believed others were praying for your recovery.

One researcher found that patients in the prayed-for group were 11% more likely to do well than patients in the not-prayed-for group.

Byrd found that 85% of those in the prayed-for group had a "good" medical outcome, versus 73% in the non-prayed-for group, while only 14% in the prayed-for group had "bad" outcomes, versus 22% in the no-prayer category. These differences were statistically significant for the size of the group studied and thus seemed to confirm that prayer had helped.

A second, similar study, widely noted in news reports in the fall of 1999, also concerned coronary care patients, this time in a hospital in Florida. William Harris, a medical researcher, again broke patients into prayed-for and non-prayed-for groups. Again, the test was blind, patients not knowing which category they were in. Harris found that patients in the prayed-for group were 11% more likely to do well on a standard scale of coronary health indicators than patients in the not-prayed-for group.

The Byrd and Harris studies form the basis of hard-science assertions that the value of intercessory prayer has been scientifically confirmed. A third study, by Elizabeth Targ of the University of California at San Francisco, where a number of researchers form an important Spirituality and Health Group, claims further proof.

Targ staged a randomized, double-blind study of what she calls "distant healing," in which volunteers prayed for AIDS patients. Targ and her colleagues found that the prayed-for group had fewer hospital and physician visits than a control group and showed other indicators of improved health.