The long awaited Study of the Therapeutic Effects of Intercessory Prayer in
Cardiac Bypass Surgery (STEP), which saw the light of day March 30 on the
website of the American Heart Journal, presented Americans with the
sort of news they don’t like to get. American journalists did what they
could to soften the blow.
STEP was not the first study of its kind, but it was by far the largest and
most methodologically punctilious. Under the leadership of Harvard
cardiologist Herbert Benson, author of The Relaxation Response and
advocate of the beneficial effects of religion on health generally,
investigators divided 1,802 subjects into three groups: (1) patients who
were told they might be prayed for and were; (2) patients who were told they
might be prayed for and weren’t; and (3) patients who were told that they
would be prayed for and were.
group of Catholic monks, a group of Catholic nuns, and a Protestant prayer
ministry signed on to pray systematically for each patient to see if there
might be an impact above and beyond the prayers that, it was assumed, most
if not all patients would be receiving from friends and family members (not
to mention themselves).
Fifty percent of all recipients of heart bypass operations normally
experience some medical complication within 30 days. Going in, STEP
hypothesized that Group 2 (uncertain, unprayed-for) would experience
complications at the normal rate, while Group 1 (uncertain, prayed-for)
would be at 40 percent and Group 3 (certain, prayed-for) at 30 percent. In
fact, the hypothesis proved wrong in every respect.
The first two groups both did a bit worse than the 50 percent norm, with
Group 1 (52 percent) actually experiencing a higher rate of complications
than Group 2 (51 percent). Most strikingly, Group 3 did worst of all at 59
percent. In other words, the study both failed to show that intercessory
prayer was therapeutic and found that patients who knew they were being
prayed for were at significantly greater risk than those who didn’t. (Far
from instilling Bensonian relaxation, awareness of organized prayer may have
provoked “performance anxiety,” researchers later speculated.)
The results can hardly have pleased the
Templeton Foundation, which backed the study to the tune of $2.4 million as
part of its longstanding commitment to promoting better relations between
science/medicine and religion/spirituality. Templeton’s readiness to post
the study on its website despite the lack of positive results is testimony
to a certain intellectual coming of age on the foundation’s part.
Not long after being established in 1987, Templeton began supporting an
official sounding outfit called the National Institute for Heathcare
Research (NIHR). Headed by a former National Institute of Mental Health
psychiatrist named David Larson (who enjoyed a close relationship with the
foundation’s eponymous founder, Sir John Templeton), NIHR devoted itself not
to conducting research but to spreading the gospel of spirituality and
health through educational programs and, most notably, media outreach.
$20,000 Templeton grant for getting articles into, say, popular magazines
would be collected in increments—$2,000 per article “placed.” If this
incentivized philanthropic regime did not exactly result in the most
objective portrayal of the available research, it certainly stimulated a lot
of journalism touting the therapeutic benefits of religion.
Some of the claimed benefits were plausible whether or not you happened to
be a person of faith. That belonging to a religious congregation might be
good for your health could as easily be explained on sociological as
supernatural grounds. But intercessory prayer looked to be another matter.
If it could be demonstrated empirically that such prayer was likely to
enhance one’s medical condition, even if one were unaware of it—well, wasn’t
that tantamount to scientific proof of divine intercession?
STEP seemed to show the contrary, but the journalists—medical writers, for
the most part—were at pains to suggest that it ain’t necessarily so.
They turned to experts in both the medical and theological communities to
cast doubt on the legitimacy of trying to use science to measure religion,
and to pastors and layfolk for testimony that yes, prayer, has worked for
them. (The essence of this vox populi approach could be found
on Beliefnet.com, which ran the Religion News Service news story on the
study and then invited readers to send in their stories about how their
prayers had helped others and vice versa.)
The St. Petersburg Times cut right to the chase with the headline,
“Is it the prayer that’s flawed, or is it the study?” You decide.
One of the frequently cited critics was Duke University psychiatrist Harold
Koenig, himself a frequent recipient of Templeton largesse, who stressed the
incommensurability of science and the supernatural. “There are no scientific
grounds to expect a result and there are no real theological grounds to
expect a result either,” he told the AP’s Malcolm Ritter. “There is no god
in either the Christian, Jewish or Muslim scriptures that can be constrained
to the point that they can be predicted.”
Similarly, Daniel Sulmasy, a Franciscan priest as well as director of ethics
at two New York hospitals, told the Chicago Tribune, “It seems
fundamentally sinful to conceive of God as our instrument.”
Another oft-quoted expert was Columbia University’s Richard Sloan, a
professor of behavioral medicine and author of a forthcoming book, Blind
Faith: The Unholy Alliance of Religion and Medicine.
“The problem with studying religion scientifically is that you do violence
to the phenomenon by reducing it to basic elements that can be quantified,”
Sloan told Benedict Carey of the New York Times, “and that makes for
bad science and bad religion.”
But bad according to whom? Registering evidence of God’s work in the world
is a pretty central feature of Western religion. It was not so long ago that
American evangelicals embraced the “Baconian” view that common sense
empiricism would demonstrate the truths of their religious beliefs. To this
day, the Catholic Church deploys scientific experts to determine whether
potential saints have performed the necessary miracles.
Nor, for that matter, is it obvious that the organized intercessory prayer
performed incessantly by monastic communities in the Middle Ages for the
wellbeing of their benefactors took a less instrumental view of God than the
STEP prayer communities. Regardless of the relevance of other factors,
including the relative worthiness of the patient and of those doing the
praying, it does not seem so theologically out of bounds to suppose that, as
a result of all the praying, at least a few more patients would have avoided
Such considerations aside, there is reason to question the unspoken
assumption that STEP was really about religion at all. Back in 1998, when
the study was getting off the ground, Benson told Bill Hendrick of the
Atlanta Journal Constitution that he thought prayer worked “because of
some unknown energy force that travels from one brain to another, even over
great distances.” At least as far as the principal investigator was
concerned, positive results could just as well have confirmed an hypothesis
of parapsychology as of God.
it happened, the most serious critique of the study came not from the
critics in the press but from an editorial in the American Heart Journal
itself. Written by three Duke Medical School colleagues, the editorial
took Benson et al. to task especially for offering no explanation of Group
3’s statistically significant higher complication rate other than that it
may have been random.
“If the result had shown benefit rather than harm,” the editorial asked,
“would we have read the investigators’ conclusion that this effect ‘may have
been a chance finding,’ with absolutely no other comments, insight, or even
speculation?” By going no farther, the STEP researchers had “allowed
cultural presumption to undermine scientific objectivity.” As with any other
finding of negative effects from an experimental technique, it behooved them
to warn future researchers of the safety issues attendant upon future
inquiry into such “mechanistically undefined ‘frontier’ therapy research.”
With the exception of the Chicago Tribune’s Jeremy Manier, no
reporter took note of the editorial. It fell to the on-line likes of
Slate’s William Saletan and Blogcritics.org’s Anil Menon to push
the interpretive envelope in uncomfortable directions—such as the
possibility that the perils of intercessory prayer might go beyond clinical
Menon put it, “Well it’s official. If you’re scheduled for a coronary bypass
and the local Ned Flanders is busy organizing the congregation to pray for
you, order the bastard to cease and desist immediately.”
America, the cultural presumption in favor of prayer is pretty damn strong.
order to assess the full range of possible interactions between
intercessory prayer (or lack of it) and patient awareness, it would have
been useful to create one group of patients who received prayers but
weren’t told about the study, another who were told they would receive
prayers but didn’t, and another who were told they weren’t but were.
Although there would have been ethical issues involved in creating such
groups, a similar research project (Harris et al., Archives of Internal
Medicine, 1999) made a sufficiently strong case that the hospital
institutional review board permitted the study to proceed without
obtaining patients’ informed consent.
lead author, cardiologist Mitchell W. Krucoff, has collaborated on several
papers with Harold Koenig, including one addressing the therapeutic
effects of stress relaxation and prayer on heart patients. Both are
scientists at Duke’s Center for Spirituality, Theology and Health.