Secondary Logo

Journal Logo


Should Academic Medical Centers Conduct Clinical Trials of the Efficacy of Intercessory Prayer?

Halperin, Edward C. MD

Author Information


A recent article in Time described a clinical trial conducted in the cardiology unit of a Veterans Administration Medical Center.1 Patients undergoing invasive procedures were randomized into two groups. Some patients were prayed for by Buddhist Monks in Nepal, at a Roman Catholic Carmelite Convent near Baltimore, at an interdenominational Christian prayer center in Missouri, by Jews at the Western Wall in Jerusalem, and by supplicants at several other sites. The remaining patients served as controls. The study was designed to determine “whether prayer by strangers might influence the medical outcomes of 30 patients” in the cardiac catheterization laboratory. The magazine report indicated that “the outcomes of those prayed over were 50% to 100% better than those of a control group.” The primary investigator felt these results were “sufficient … to be intriguing.” A larger-scale study is planned.

Upon reading the above, the reader might ask whether clinical trials of the efficacy of intercessory prayer fall within the realm of scientific medicine or are an attempt to use the methods of science to answer a theological question. This essay is my attempt to confront that question, which underlies the question stated in the essay's title. To do so, I define the concept of intercessory prayer (see the sidebar for definitions), contrast it with other forms of prayer, and review the literature concerning clinical trials of the efficacy of intercessory prayer. The arguments for and against conducting such trials are described so that the reader may consider their relative merits. Last, I discuss the potential power of faith in healing, review the philosophical basis and pitfalls of clinical trials of intercessory prayer, and reflect on the place of these trials in academic medicine.


The work described in Time is one of many studies of intercessory prayer published in the medical literature. The point of the brief survey that follows is not to engage in a detailed critique of the methods and results of these studies. I wish, rather, to give the reader a sense of the literature. In this way we can subsequently turn our attention to the question reflected in the title of this essay: Is the academic medical center the appropriate forum for the study of intercessory prayer?

Sir Francis Galton (1822–1911), explorer, anthropologist, student of human intelligence, and eugenicist, conducted one of the first studies of intercessory prayer.2

It is asserted by some that men possess the faculty of obtaining results over which they have little or no direct personal control, by means of devout and earnest prayer, while others doubt the truth of this assertion. The question regards a matter of fact, that has to be determined by observation and not by authority; and it is one that appears to be a very suitable topic for statistical inquiry … Are prayers answered or are they not? … Do sick persons who pray, or are prayed for, recover on the average more rapidly than others?3

Galton considered that, of all classes of English society, the groups most prayed for were the Royal Family and the clergy's children. If prayer is effective, he reasoned, these people should live longer than other persons exposed to similar health risks. Galton, therefore, compared the average lifespans of kings with those of lords and the lifespans of children of clergy with those of children of other professional men. He found, contrary to the hypothesis, that much-prayed-for persons had slightly shorter lives than did those with whom he compared them. Galton also assessed the frequency with which ships carrying missionaries experienced disaster at sea compared with the frequency of this event with other ships. He found that missionaries sank slightly more often and with greater loss of life.4–6

Moving to the present, in 1965, Joyce and Whelldon reported a small double-blind study of 11 pairs of adult out-patients with physiologic and rheumatic diseases. They assigned one patient in each pair to receive intercessory prayers and the other not to receive prayers, as a control. Over a 15-month interval the patients were evaluated using psychiatric and physical measures. In seven pairs of patients the prayed-for individual did better, in five the control did better.7

In 1969 in a study of children with leukemia reported by Collipp,8 ten were prayed for in a Protestant church and eight not. After 15 months of prayer, seven of the ten prayed-for children were alive compared with two of the eight control children. Byrd of the University of California at San Francisco reported on 393 patients entered into a prospective double-blind randomized trial conducted in 1982–83 and reported in 1988. All patients were admitted to the coronary care unit at San Francisco General Hospital. Half were assigned to be prayed for by “born again” Christians and half not. Of the 29 variables studied (such as days in the coronary care unit, days in the hospital, number of discharge medications, development of congestive heart failure, need for antibiotics), there was no difference between the two groups in 23. For six variables the results favored the prayed-for group. By combining the data into a good-, intermediate-, or bad-outcome composite score, the study is reported to favor prayer (p < .01).9,10

In 1992, O'Laoire took 496 volunteers, attracted by advertisements in San Francisco Bay Area newspapers, and assigned half to a prayed-for group and half to a control group. Several psychological metrics were used to assess anxiety, self-esteem, mood, and depression. The 1997 report of the study indicated that there was no difference in outcomes between the two groups.11

In a 1997 report, Walker et al., from the University of New Mexico, randomized 40 patients admitted for therapy for alcoholism so that 22 were the recipients of distant prayer by volunteers (who reported “more than five years of regular intercessory prayer experience”) and 18 were in the control group. No difference was found between prayer intervention and nonintervention with respect to the patients' alcohol consumption over time.12

In 1998, Sicher et al. reported a randomized double-blind trial of the effect of distant healing in a population with advanced AIDS. Forty pairs of patients were randomized to prayer by “healers from Christian, Jewish, Buddhist, Native American, and shamanic traditions as well as graduates of secular schools of bioenergetic and meditative healing.” At six months a review of the medical records showed no difference in the CD4 counts of the prayed-for group versus those in the control group. The prayed-for patients, however, were stated to have significantly fewer AIDS-defining illnesses and less severe illness, and to require fewer doctor visits and hospitalizations, as well as showing improved mood compared with the controls.13

In 1999 Harris et al. reported a trial involving 1,019 patients admitted to a coronary care unit.14 The 990 analyzed cases were randomized to receive distant intercessory prayer by Christian intercessors or to be in a control group. The institutional review board at the Mid-America Heart Institute granted permission to conduct the trial without the individual patients' giving informed consent to participate. This decision was based on the basis of the absence of risk in the study and the possibility that obtaining consent might increase anxiety in some patients. Thirty-seven outcome measures were studied (such as need for interventional cardiac procedure, antibiotics, temporary pacemaker placement, and two global outcome scores). There was no significant difference in outcomes (p > .05) for 36 of the 37 measures and one of the two global outcome scores. Because there was a benefit in what the authors view as the more important global outcome score, the trial is reported as being positive, favoring the prayer group. The trial has been extensively critiqued on statistical design, technique of randomization, lack of validation of global outcome scoring system, and interpretation.15–30


One can mount several arguments in favor of and several against the conduct of clinical trials concerning the efficacy of intercessory prayer. I consider these in turn, leaving the reader to judge how persuasive they are individually or in the aggregate.

In Favor

  1. The academic medical center should seek new knowledge concerning therapeutics, irrespective of the source of the original hypotheses.
  2. Academic medical centers exist for the purpose of the generation, conservation, and dissemination of new knowledge concerning the causes, prevention, and treatment of human disease. There is a long history of treatment of human disease's being faith-based. While faith-based treatment has been supplanted, in many circles, by reliance on scientific medicine, it is frequently relied upon by patients and some physicians. Academic medical centers should rigorously investigate such a widely used treatment modality. Interest in alternative or complementary medicine is growing rapidly in the United States. Prayer “therapy” may fall into this category.14
  3. An obvious corollary to this line of argument is that it might open the door to the testing of other hypotheses that may be less palatable, to some academicians, than investigations of prayer. If one accepts, for example, the premise that alleged forms of therapeutics are worthy of study if they are of interest to a significant portion of the public, this might lead to agreement to test an infinite dilution of a bacterial extract in water (what some would call a homeopathic therapy and others might call distilled water) for the treatment of pneumococcal pneumonia.
  4. Medical research has an obligation to respond to the demands of the body politic in selecting topics for research.
  5. Academic medical centers are well accustomed to seeking research grants. It is, in fact, a large part of their raison d'être. These research grants derive from taxpayer dollars funneled through federal agencies and philanthropy. If the public demands new therapies for AIDS, breast cancer, or mental illness, then their elected officials will pressure federal agencies to set up mechanisms to fund research in these areas. Similarly, if the public has a strong interest in the value of prayer as medical therapy, it is appropriate for grant-funding agencies to promote such research and for academic medical centers, which have a fiduciary obligation to the society that ultimately supports them, to perform such research.
  6. Clinical trials of intercessory prayer take no position on the existence of God.
  7. A clinical trial of the efficacy of prayer is designed to test whether or not prayer influences the clinical course of the patients in the study. The trial is not designed to prove or disprove the existence of a God who answers prayer. If the trial shows that the prayed-for patients have a better outcome than do those in the control group, then a second-generation trial might look at the mechanism of action of the observed outcome. The investigator, however, is not taking a position with respect to the existence or non-existence of God by studying, initially, whether prayer influences clinical outcomes.
  8. One way of framing this argument is to assert that studies of intercessory prayer are designed to explore a phenomenon and not a mechanism. If “prayer works,” that is if a clinical trial of prayer is positive, one may then consider natural or supernatural explanations. Are the effects of prayer attributable to currently unknown physical forces associated with intercessory prayer and received by patients? Are the effects of prayer related to a supernatural cause, i.e., a God who responds to prayer? Clearly one must do clinical trials of prayer first and show they are positive before addressing the mechanism of action. Physicians administered aspirin and digitalis before fully understanding their pharmacology and biochemistry, used radiation therapy before understanding DNA damage and clonogenic death, and did oophorectomies for breast cancer before understanding hormone receptors. It is clear that medicine is comfortable basing therapies on observational studies, with mechanistic explanations following decades or centuries later.14

Not in Favor

  1. It is impossible to design a controlled trial of the efficacy of intercessory prayer.
  2. A clinical study of intercessory prayer must face the problem of identifying statistically similar groups of subjects for the treatment and control arms of the study. Even if one found wholly comparable groups of patients based upon type and stage of disease, it would be impossible to find equivalent groups based upon faith or “worthiness” to be healed by prayer. It is also difficult to imagine how one would find equivalent groups of intercessors and prayers.31
  3. The end-point of the trial must also be specified. Death is an unequivocal end-point. Measuring end-points such as number of days in the intensive care unit, the need for cardiac catheterization, or administration of antibiotics is somewhat more subjective. Some investigators try to solve this problem by using “outcome scores.” Such scoring systems need to be generally agreed upon, validated, and reproducible. Finally, the laws of statistical reasoning caution us that if we measure enough outcome measures for two groups of patients, some measures will show a statistically significant difference, by chance, when no real difference exists.
  4. Intercessory prayer experiments might be associated with a dose—response relationship. How “much” prayer works? How is prayer measured? Is it measured by the time, by the degree of fervency of the prayer, and/or by the degree of sincerity or faith of the prayer?31–37 Does the faith tradition (e.g., Buddhism, Christianity, etc.) of the intercessor matter? Unless we have some concept of dose, we cannot study the effect of prayer as therapy.
  5. Establishing a true control group of persons for whom there is no prayer may be impossible. Patients and their friends and family will certainly pray for recovery. Enemies may pray for the reverse. If an investigator believes that prayer matters, how does one account for the “unintentional doses” of such possible unplanned-for interventions? If a patient who is participating in a study of the effects of prayer asks a hospital chaplain to pray with or for him or her, will the patient be dropped from the study?
  6. An academic medical center conducting a clinical trial of intercessory prayer is offensive to religion.
  7. Studies of intercessory prayer are evaluating one particular type of prayer—a request for a specific action. In attempting a study of the probability of miraculous divine intervention, academic medical centers are, in fact, trying to prove the existence of a God who answers prayers. People of faith generally do not demand that God perform healing and they do not give God a timetable to perform the task.22 For example, the injunction against man's asking God to perform on demand is found in both the Old Testament and the New Testament. “You shall not test the Lord thy God, as you tested him in Messah” (Deuteronomy 6:16); and “Jesus said to him. It is written again. Thou shalt not test the Lord thy God” (Matthew 4:7). This principle of faith is also described when God addresses man in the Book of Job: “Where wast thou when I laid the foundations of the earth? Declare, if thou has understanding … Shall he that contendeth with the Almighty instruct him? He that reproveth God, let him answer it” (Job 38:2–40:2).
  8. K. S. Thompson describes an attempt at a “scientific” proof of the existence of God as an effort
  9. to create a situation in which God must show himself or herself and perform a miracle—something that gods in general do only very rarely. Even more rarely do they perform on demand. If God or gods feel the need to give us a sign, he or she or they tend to choose the time, place, and form; not us. Simply put, I am sure that such tests are hideously arrogant at best and certainly blasphemous.31
  10. Intercessory prayers for the good health of others raise philosophical and theological problems. Will God, who is all-knowing, all-powerful, and all-good (by definition for the good of His people), turn away from His intended purpose because of a human's expressing his or her desires?31 One would have difficulty accepting the concept of a God who preferentially heals people who, in a clinical trial, are selected to be prayed for by strangers rather than healing those randomly assigned to receive no prayer. God should not be conceived of as so capricious.23
  11. Clinical trials of the efficacy of prayer are attempts to prove the existence of God.
  12. It is true that some effective drugs were discovered serendipitously. Some forms of therapy were employed for years before there was any sound scientific explanation for their efficacy. “If prayer works,” continues the argument, “then it doesn't matter if we cannot derive the explanation for its mechanism of action.” This is the so-called “black box” argument—if something works clinically, it does not matter if we can explain why or how.
  13. Remote intercessory prayer, if it works, could be explained only by the intervention of God in the physical world by a supernatural mechanism or by telekinesis. Neither mechanism can or will be demonstrable by a credible replicable scientific experiment.19 A clinical trial of the efficacy of prayer is an attempt to prove the existence of God. To claim that these studies are exploring only a phenomenon and not a mechanism is a ruse.
  14. Experimental therapeutics should be based on pathophysiology and a likely mechanism of the proposed therapy's action. Physicians should not expend effort in the pursuit of irrational treatments. In a world of scarce resources, we should select subjects for clinical research on the basis of evidence-based hypotheses. Political pressure or public infatuations should not hold sway.
  15. The methods of scientific medicine are inapplicable to theology.
  16. Theology concerns itself with truths derived by deductive reasoning: An apparently rational universe exists. What are its origins? Whence arose its first cause? A higher power or higher being directs the world and our actions. Who or what is it? Is it appropriate to worship it? The theologian reasons back from what currently is to what caused it.38–40
  17. Scientific knowledge is different. Science acquires information by feel and touch. The physician is required to see, handle, weigh, and measure phenomena. After extensive categorization of information, physician—scientists seek to identify truths generated by artful arrangement of the data. The theologian is interested in explaining why things are as they are based on logical deductions from the truths acquired through revelation. The physician—scientist, on the other hand, attempts to discover how things are by observing and measuring them.40
  18. Medicine has the possibility of continual acquisition of knowledge and a change in its dominant paradigms. New understandings of biology lead to new hypotheses, new experiments, new therapeutic targets, and new clinical practices. Theology, on the other hand, is concerned with fundamental truths. While some religions accept changes over time in certain aspects of their practices and rituals because they believe in “progressive revelation,” the fundamental underpinnings are generally thought of as immutable.
  19. Theology and scientific medicine, therefore, intellectually operate in different spheres.40,41 Attempting to use an established truth of molecular biology to explain the truths endorsed by the faithful as their theology, or vice versa, seems analogous to asking the astrophysicist to use his or her special expertise to lecture on 16th-century Spanish art history.33–36,42,43
  20. The distinction between theology and science was summarized by Cardinal Newman over 150 years ago:
  21. Theology begins, as its name denotes, not with sensible facts, phenomenon or results, not with nature at all, but with the Author of nature—with the one invisible approachable basis and source of all things. It begins at the other end of knowledge and is occupied not with finite but with the infinite. It enfolds and epitomizes what He Himself has told of Himself, of His nature, His attributes, His will, and His acts … physical science is experimental, theology traditional; physical science is the richer, theology the more exact; physics is the bolder, theology the shorter; physics progressive, theology in comparison, stationary; theology is loyal to the past, physics has visions of the future.40


Sir William Osler clearly understood the power of faith to contribute to healing. Osler, however, saw faith as a general concept involving faith in authority. Osler realized that patients may be aided by faith in the power and knowledge of the physician as well as in God or the saints. In an essay, “The Faith That Heals,” Osler wrote

As Galen says, confidence and hope do more than physick— ‘he cures most in whom most are confident’ … Faith in the Gods or in the Saints cures one, faith in little pills another, suggestion a third, faith in a plain common doctor a fourth … The cures in the temples of Aesculapius, the miracles of the Saints, the remarkable cures of those noble men, the Jesuit missionaries, in this country, the modern miracles at Lourdes and at St. Anne de Beaupré in Quebec, and the wonder-workings of our latter day saints are often genuine and must be considered in discussing the foundations of therapeutics. We physicians use the same power every day. If a poor lass, paralyzed apparently, helpless, bed-ridden for years, comes to me having worn out in mind, body and estate a devoted family, if she in a few weeks or less by faith in me, and faith alone, takes up her bed and walks, the Saints of old could not have done more.44

There is clearly a human desire to fill a void science cannot. The problem is well described by Normal Levitt.

The rift between the scientific world view and the common need for some assurance that human existence is not a pointless accident ensures that, well into the future, a substantial portion of the population will contrive ideological and psychological defenses against science. Large numbers of people, perhaps amounting to a majority in even the most scientifically advanced societies, will remain alienated from science, though that condition will often be covert and inarticulate, embodied in quiet reservations rather than public manifestos … in the presumptions and prejudices that undergird our social structure, science is something of a foreign body, consequently it draws antibodies to itself.45

As Osler recognized, physicians must not denigrate the importance of spirituality or faith in the healing process or the inadequacies of science for some patients. Not all of healing is explicable by medicinal therapeutics. Healing is a complex process that involves the medications we physicians prescribe, the surgery we perform, and the diagnostic x-rays we order. Healing also requires human interaction, faith, and a family support structure. There are facets of healing that are complementary to scientific medicine and deserve to be integrated with it. There can be little objection to the use of harmless complementary forms of healing with scientific medicine so long as these forms are used in addition to scientific medicine and not in lieu of it.46

Scientific understanding does not necessarily contradict religion—depending upon one's understanding of God and sacred texts. Sir Francis Bacon, for example, felt that by separating “the absurd mixture of matters divine and human,” what science would allow us to do is “to render unto faith the things that are faith's.”39 Recourse to prayer during illness is not an indication that the patient or family lacks confidence in traditional medical therapy. In many religious traditions, confidence in the healing power of God is not intended to usurp or negate the essential functions of physicians and medical science.46,47

The community of scholars of the academic medical center must understand the limits of both science and religion.38,48 There are various ways to seek truth that cannot, by their nature, overrule each other.38,41,47,48 The academic medical center should be devoted to the generation, conservation, and dissemination of knowledge and should utilize intellectual tools suitable for the purpose. Physicians should have an informed appreciation of theology as it affects their patients, their own lives, and both the physicians' and the patients' understanding of their place in the world.

The debate over clinical trials of intercessory prayer raises questions concerning the exploration of the universe of faith utilizing the scientific method. Medical academicians must engage in a serious weighing of the arguments for and against such trials.

Definition of Terms Used in This Article

Prayer is “a solemn and humble request to God, or to an object of worship; a supplication, petition, or thanks-giving, usually expressed in words.”49Petition prayers request preservation of or return to health, material goods, prosperity, or success in undertakings. Confession prayers express faith and a recognition of the state of sin. Intercessory prayers call for aid to others. Such prayers presume a belief in a God or other entity worthy of worship that can be influenced by correctly and sincerely expressed requests. Prayers of praise and thanks-giving note the majesty of God and God's creation. Adoration prayers are a spiritual prostration before God and may be accompanied by a bow, kneeling, or touching the forehead to the ground. Prayers of mystical union or ecstasy attempt a departure from human limitations and a meeting with the Divine.50 Clinical studies of prayer generally concentrate on the efficacy of intercessory prayer.

In this essay, I use the term theology as it was derived from French (theos, a god + legin, to speak). Theology is religious knowledge or belief, especially when methodically formulated. I define scientific medicine as that field of study concerned with observation and classification of facts related to the causes, prevention, and treatment of human disease. This knowledge is particularly acquired by induction, hypothesis formulation and testing, and experimentation.51


1. Van Biema D. A test of the healing power of prayer. Time. 1998;152:72–3.
2. Sir Francis Galton. In: Encyclopaedia Britannica, 15th ed. Chicago, IL: Encyclopaedia Britannica, vol. 7, 1974:858–9.
3. Galton F. Inquiries into human faculty and its development. London, U.K.: Macmillan, 1883:277–94.
4. Helig J. Testing the power of belief. Science. 1997;276:891.
5. Haldane JBS. Possible worlds and other papers. London, U.K.: Chatto and Winduis, 1928:237–52.
6. Galton F. Fortnight Review. 1872;12:125.
7. Joyce CRB, Whelldon RMC. The efficacy of prayer: a double-blind clinical trial. J Chronic Dis. 1965;18:367–77.
8. Collipp PJ. The efficacy of prayer: a triple-blind study. Med Times. 1969;987:201–4.
9. Byrd RC, Sherrill J. The therapeutic effects of intercessory prayer. J Christ Nurs. 1995;12:21–3.
10. Byrd RC. Positive therapeutic effects of intercessory prayer in a coronary care unit population. South Med J. 1988;81:826–9.
11. O'Laire S. An experimental study of the effects of distant, intercessory prayer on self-esteem, anxiety, and depression. Alt Ther Health Med. 1997;3:38–53.
12. Walker SR, Ronigan JS, Miller WR, Corner S, Kahlich L. Intercessory prayer in the treatment of alcohol abuse and dependence: a pilot investigation. Alt Ther Health Med. 1997;79–86.
13. Sicher F, Targ E, Moore II D, Smith HS. A randomized double-blind study of the effect of distant healing in a population with advanced AIDS: report of a small scale study. West J Med. 1998;169:356–63.
14. 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 Intern Med. 1999; 159:2273–8.
15. Sloan RP, Bagiella E. Data without a prayer. Arch Intern Med. 2000; 160:1870.
16. Karis R, Karis D. Intercessory prayer. Arch Intern Med 2000;160:1870.
17. Goldstein J. Waiving informed consent for research on spiritual matters? Arch Intern Med. 2000;160:1870–8.
18. Van der Does W. A randomized, controlled trial of prayer? Arch Intern Med. 2000;160:1871–2.
19. Sandweis DA. P Value out of control. Arch Intern Med. 2000;160:1872.
20. Hamm RM. No effect of intercessory prayer has been proven. Arch Intern Med. 2000;160:1872–3.
21. Price JM. Does prayer really set one apart? Arch Intern Med. 2000; 160:1873.
22. Pande DN. Does prayer need testing? Arch Intern Med. 2000;160:1873–4.
23. Galishoff ML. God, prayer, and coronary care unit outcomes: faith vs. works? Arch Intern Med. 2000;160:1877.
24. Hammerschmidt DE. Ethical and practical problems in studying prayer. Arch Intern Med. 2000;160:1874.
25. Rosner F. Therapeutic efficacy of prayer. Arch Intern Med. 2000;160:1874.
26. Waterhouse WC. Is it prayer, or is it parity? Arch Intern Med. 2000; 160:1875.
27. Hoover DR, Margolick JB. Questions on the design and findings of a randomized, controlled trial of the effects of remote, intercessory prayer on outcomes in patients admitted to the coronary care unit. Arch Intern Med. 2000;160:1875–6.
28. Smith JG, Fisher R. The effect of remote intercessory prayer on clinical outcomes. Arch Intern Med. 2000;160:1876.
29. Zimmerman SM. Prayer can help. Arch Intern Med. 2000;160:1876.
30. Harris WS, Gowda M, Kolb JW, et al. In reply [to comments]. Arch Intern Med. 2000;160:1877–8.
31. Targ E, Thompson KS. Can prayer and intentionality be researched? Should they be? Alt Ther Health Med. 1997;3:92–105.
32. DeLashmutt M, Silva MC. The ethics of long-distance intercessory prayer. Nursing Connec. 1998;11:37–40.
33. Mayer D. Testing the power of belief. Science. 1997;276:891
34. Cox BM. Testing the power of prayer. Science. 1997;276:1630–1.
35. Kling J. Testing the power of prayer. Science. 1997;276:1631.
36. Fish S. Can research prove that God answers prayers? J Christ Nurs. 1995;12:24–46.
37. Davis T. The research evidence on the power of prayer and healing. Canad J Cardiovasc Nurs. 1994;5:34–6.
38. Heard A. Speaking of the University: Two Decades at Vanderbilt. Nashville, TN: Vanderbilt University Press, 1995.
39. Podhoretz N. Science hasn't killed God. Wall Street Journal. Dec. 30, 1999, p.A12.
40. Newman JH. The Idea of a University. New Haven, CT: Yale University Press, 1996.
41. Giamatti AB. A Free and Ordered Space: The Real World of the University. New York: W. W. Norton, 1990.
42. Roush W. Herbert Benson: mind—body maverick pushes the envelope. Science. 1997;276:357–9.
43. Targ E. Evaluating distant healing: a research review. Alt Ther Health Med. 1997;3:74–8.
44. Osler W. The faith that heals. BMJ. 1910; 1470–2 [as quoted in Bliss M. William Osler: A Life in Medicine. Oxford, U.K.: Oxford University Press, 1999:275–6].
45. Levitt N. Prometheus Bedeviled: Science and the Contradictions of Contemporary Culture. New Brunswick, NJ: Rutgers University Press, 1999.
46. Prayer. In: Werblowsky RJ, Wigoder G (eds). The Encyclopedia of the Jewish Religion. New York: Amanda Brooks, 1986:306–7.
47. Rosner F. Complementary therapies and traditional Judaism. Mt Sinai J Med. 1999;66:102–5.
48. Blazer D. Freud vs. God: How Psychiatry Lost Its Soul and Christianity Lost its Mind. Downers Grove, IL: Inter Varsity Press, 1998.
49. The Compact Edition of the Oxford English Dictionary. Oxford, U.K.: Oxford University Press, 1985:2,268.
50. Hamman AG. Prayer. In: Encyclopaedia Britannica, 15th ed. Chicago, IL: Encyclopaedia Britannica, Vol. 14, 1974:948–53.
51. Webster's New Collegiate Dictionary. Springfield, MA: G. & C. Merriam, 1960:757,881.
© 2001 Association of American Medical Colleges