From The Scientific Review of Alternative Medicine, Vol. 6, No. 1, Winter 2002

Study Yields No Evidence for Medical Efficacy
of Distant Intercessory Prayer:
A Follow-up Commentary

by Gary P. Posner, M.D.

The medicine of the future is going to be prayer and Prozac. If we can prove the medical benefits of intercessory prayer, it's going to be page 1 news. You're going to see a revolution. The world of medicine will be turned upside down.

As reported in the New York Times,  the above remark was made several years ago by Dale A. Matthews, M.D., associate professor of Medicine at Georgetown University School of Medicine. Dr. Matthews was discussing his nearly completed study on the medical efficacy of intercessory prayer before "a slightly bewildered crowd" of future doctors at St. Louis University School of Medicine.1

My 1998 Scientific Review of Alternative Medicine  article about Matthews' study-in-progress concentrated on media coverage that the research was receiving even as the data were still being collected.2 The premature news stories by two local television stations provided glimpses of what appeared to be leaky protocol that could have skewed the data in favor of a desired result. Matthews' yearlong study, testing the effects of intercessory prayer on patients with rheumatoid arthritis, has now been published.3

Most previous prayer studies had endeavored to delineate the possible clinical benefits of religious faith and personal prayer versus intercessory prayer. The terms "intercessory," "distant," and "remote" have generally been employed interchangeably, as distinguished from "personal." There is little disagreement that personal prayer, and its attendant emotions and expectations, can affect one's perceptions of well-being. On the other hand, studies claiming favorable clinical effects of prayer by distant intercessors, without the patients' knowledge, have been less persuasive.

The Byrd4 and Harris5 studies were the two most recent efforts to demonstrate the medical efficacy of distant, intercessory prayer. As previously reported in these pages,6 both were fraught with flaws rendering their positive findings valueless.7-11 Thus, the Matthews study was much anticipated.

Those hoping for confirmation of a positive therapeutic effect of distant prayer were to be disappointed. As Matthews explained, "Neither multivariate nor univariate analysis showed a statistically significant overall improvement after intervention in the 10 outcome variables for the group receiving . . . distant intercessory prayer (n = 19) when compared with the group receiving no . . . distant prayer (n = 21)." And the placebo effect was in evidence: "Although only . . . 48% [of patients] actually received distant prayer and individuals were blinded to distant prayer treatment status . . . 73% [believed they received] distant prayer. These patients were more likely than others to have improvement in global function . . . and reductions in pain."

However, those clearly stated findings mask a source of confusion. For example, the "Conclusions" section of the paper's abstract begins, "In-person intercessory prayer may be a useful adjunct to standard medical care for certain patients with rheumatoid arthritis." In addition to the blinded distant prayer, all patients in the study received in-person prayer by intercessors. Thus, authors represent this study as having positive findings with regard to "intercessory" prayer, though they were negative for "distant" intercessory prayer.

How positive were the effects of the in-person prayer? Compared to pre-prayer measurements, Matthews reported statistically significant improvements in patients' grip strength, number of tender and swollen joints, pain, fatigue, and level of functional impairment. Some of these variables are clearly subjective in nature, though measuring and counting swollen joints is objective. In a comparison of baseline and 12-month data for the study's entire population, the forty patients began with a mean of 9.8 swollen joints, and ended with 3.1, a difference significant at the P < .0001 level. All measurements were taken and recorded by a nurse practitioner who also served as the study's clinical director and co-author, and who was the focal point of the premature TV news stories. According to Matthews,

One unexpected and unexplained finding was that the improvement in swollen and tender joints and reduction in pain and functional disability . . . was not accompanied by a parallel reduction in serum inflammatory markers (ESR and CRP) [erythrocyte sedimentation rate and C-reactive protein]. Therefore, it is possible that the detected clinical improvement might be attributable more to alteration of patients' perceptions regarding their illness than to changes in inflammatory pathways affecting their joints.

If the reported reductions in swollen joints (objective), not merely pain/tenderness and fatigue (subjective), are genuine, Matthews' data would suggest the possibility (though he does not) of a "mind-body" phenomenon operating independent of the scientifically established inflammatory mechanisms that are measured by objective blood analysis, or perhaps even the existence of a god who answers in-person intercessory prayers. Given the extraordinary nature of the implications of a "positive" medical prayer study, it would have been preferable to have had an independent, disinterested clinician taking the measurements. One must entertain the possibility of loose protocol and human error.

The blood tests (ESR and CRP) were almost certainly performed objectively and blindly. And we are told that the nurse practitioner had been "blinded to the group status (group 1 vs. group 2) . . . as were all members of her office staff (with the exception of one study coordinator)." Yet the premature television reports raise doubts as to how successfully this was accomplished.

One subset of patients ("group 2") had been "designated as a 'waiting list' control group, receiving no [in-person] prayer intervention in the initial phase of the study," but only after a six-month waiting period. When measurements were taken of the "group 1" patients (those receiving in-person prayer from the outset) six months into the study -- just before the group 2 patients began receiving in-person prayer -- significant differences favoring the prayed-for patients were reported in terms of "tender joints . . . as well as a trend toward greater improvement for group 1 in seven of the other eight variables." Though its P = .128 level of significance is hardly convincing, the number of swollen joints nevertheless trended in that direction, with the group 1 mean dropping from 8.7 to 4.4, and the group 2 patients dropping only from 11.8 to 10.4.

If the nurse practitioner's measurements were indeed taken under strictly blinded conditions, this study's findings imply that in-person prayer may result in objective -- not merely subjective -- improvement that is observable, even if not corroborated by objective chemical analysis.

Given the need for extraordinarily powerful evidence to support such implications, more studies with similar results, employing independent/skeptical oversight, would be required before even "in-person" prayer therapy (not to mention distant prayer) could be persuasively credited with such clinical effectiveness.



1. Sides H. Prescription: Prayer. St. Petersburg Times  (republished from the New York Times Magazine ). December 29, 1997;D1-2.

2. Posner G. An Examination of the Media Coverage of a Prayer Study-in-Progress. Sci Rev Alt Med. 1998;2(2):34-37.

3. Matthews D, Marlowe S, MacNutt F. Effects of Intercessory Prayer on Patients with Rheumatoid Arthritis. South Med J. 2000;93(12):1177-1186.

4. Byrd R. Positive Therapeutic Effects of Intercessory Prayer in a Coronary Care Unit Population. South Med J. 1988;81(7):826-829.

5. Harris WS, Gowda M, Kolb JW et al. A Randomized, Controlled Trial of the Effects of Remote, Intercessory Prayer on Outcomes in Patients Admitted to the Coronary Care Unit. Arch Int Med. 1999;159(19):2273-2278.

6. Posner G. Another Controversial Effort to Establish the Medical Efficacy of Intercessory Prayer. Sci Rev of Alt Med. 2000;4(1):15-17.

7. Posner G. God in the CCU? Free Inquiry. 1990;10(2):44-45.

8. Witmer J, Zimmerman M. Intercessory Prayer as Medical Treatment? An Inquiry. Skeptical Inquirer. 1991;15(2):177-180.

9. Sloan RP, Bagiella E, Powell T. Religion, Spirituality, and Medicine. Lancet. 1999;353:664-667.

10. Tessman I, Tessman J. Efficacy of Prayer: A Critical Examination of Claims. Skeptical Inquirer. 2000;24(2):31-33.

11. Courcey K. Medical Claims for Intercessory Prayer Remain Elusive. Sci Rev Alt Med. 2000;4(2):9-11.