Forty percent of American scientists believe in a personal God, one who listens to their individual prayers and answers them favorably, according to a 1997 survey conducted by E. J. Larson and L. Witham and reported in Nature. With such a large pool of believers, it is not surprising to find scientists testing the existence of a personal God, possibly to confirm their belief that supernatural phenomena exist a sensational result that would rattle the other 60 percent of the scientific community.

In 1897, Andrew Dickson White, the first president of Cornell University, provided a clue to why one might be motivated to study intercessory prayers. At the beginning of the second volume of his monumental study A History of the Warfare of Science with Theology in Christendom, he wrote: Nothing in the evolution of human thought appears more inevitable than the idea of supernatural intervention in producing and curing disease.

It is this idea that Dr. Herbert Benson, president of the Mind/Body Institute at Harvard Medical School, and his collaborators set out to test in their just-published Study of the Efficacy of Intercessory Prayer, also known as STEP. The study used a sample of 1,802 cardiac bypass patients from six hospitals to measure the effects of intercessory, or third-party, prayer.

The history of prayer studies

Many writers on this subject ritually endorse two pioneer studies that claim intercessory prayers protect cardiac patients. The first is by Dr. Randolph C. Byrd, a cardiologist, at San Francisco General Hospital in 1988. The second is by William S. Harris, a medical researcher, and his colleagues at St. Lukes Hospital in Kansas City in 1999. Like many in the media, the STEP study appears to accept these two studies at face value.

Both, however, have been widely criticized. For example, I co-authored a complete analysis in the Skeptical Inquirer in 2000. We showed that the two papers contradict each other at nearly every point. Additionally, although the Byrd paper asserted that it was performed double-blind, we learned that there were major breakdowns in blindedness that Byrd inadvertently failed to report, and an essential protocol was violated.

The problem with the Harris paper was quite different. We argued that they had simply miscalculated the probability that their results could be explained by statistical fluctuations. Our recalculations showed that their results were completely consistent with a null result prayer showed no significant benefit.

I personally provided Benson and his colleague Jeffery A. Dusek, both co-authors of STEP, with copies of our critical review. Presumably, they disagree with our conclusions, for in STEP, they wrote that their negative results were novel. Referring to Byrd and Harris, they wrote, Our results are not consistent with prior studies showing that intercessory prayer had a beneficial effect on outcomes in cardiac patients. I assert that their results in fact are consistent with the properly analyzed results of Harris and colleagues.

There is now a remarkable irony: Harris and colleagues, though unaware of it, were in reality the first to publish studies showing that intercessory prayers are ineffective. The work of STEP becomes confirmatory.

Knowing when to quit

Where do we go from here? The idea of supernatural intervention may be so firmly embedded in the human mind that the scientific search for supernatural cures may never end.

One careful study published in 2001 by Jennifer M, Aviles and colleagues from the Mayo Clinic found no evidence that intercessory prayer is effective. Nevertheless, these authors reacted in the tradition of dedicated scientific inquiry: they advocate widening the search, not abandoning it. They propose further study to define endpoints that better measure the efficacy of prayer. A similar attitude was taken in 2005 by Duke University associate professor of medicine Mitchell W. Krucoff and colleagues, who also failed to detect a beneficial effect of intercessory prayer.

To advocate the continued search for effective intercessory prayer gives one the appearance of traveling the high road of open-mindedness. But appearances can be deceptive. One would do well to ponder the well-worn words popularized by the space scientist James Oberg: Let us be open-minded, but not so open-minded that our brains fall out.