Studies on Intercessory Prayer
Some religions claim that praying for somebody who is sick can have positive effects on the health of the person being prayed for.
Intercession or intercessory prayer is the act of praying to a deity on behalf of others. In Western Christianity, intercession forms a distinct form of prayer, alongside Adoration, Confession and Thanksgiving.
Meta-studies of the literature in the field have been performed showing evidence only for no effect or a potentially small effect. For instance, a 2006 meta analysis on 14 studies concluded that there is “no discernible effect” while a 2007 systemic review of intercessory prayer reported inconclusive results, noting that 7 of 17 studies had “small, but significant, effect sizes” but the review noted that the most methodologically rigorous studies failed to produce significant findings.
In comparison to other fields that have been scientifically studied, carefully monitored studies of prayer are relatively few. The field remains tiny, with about $5 million spent worldwide on such research. If and when more studies of prayer are done, the issue of prayer’s efficacy may be further clarified.
The third party studies discussed here have all been performed using Christian prayers. Some have reported null results, some have reported correlations between prayer and health, and some have reported contradictory results in which beneficiaries of prayer had worsened health outcomes. The parameters used within the study designs have varied, for instance, daily or weekly prayers, whether to provide patient photographs, with full or partial names, measuring levels of belief in prayer, and whether patients underwent surgery.
In 1872, the Victorian scientist Francis Galton made the first statistical analysis of third-party prayer. He hypothesized, partly as satire, that if prayer were effective, members of the British Royal Family would live longer than average, given that thousands prayed for their well-being every Sunday, and he prayed over randomized plots of land to see whether the plants would grow any faster, and found no correlation in either case.
Byrd and Harris
A 1988 study by Randolph C. Byrd used 393 patients at the San Francisco General Hospital coronary care unit (CCU). Measuring 29 health outcomes using three-level (good, intermediate, or bad) scoring, the prayer group suffered fewer newly diagnosed ailments on only six of them. Byrd concluded that “Based on these data there seemed to be an effect, and that effect was presumed to be beneficial”, and that “intercessory prayer to the Judeo-Christian God has a beneficial therapeutic effect in patients admitted to a CCU.” The reaction from the scientific community concerning this study was mixed. Several reviewers considered Byrd’s study to be well-designed and well-executed, while others remained skeptical. A criticism of Byrd’s study, which also applies to most other studies, is the fact that he did not limit prayers by the friends and family of patients, hence it is unclear which prayers, if any, may have been measured.
The Byrd study had an inconsistent pattern of only six positive outcomes amongst 26 specific problem conditions. A systematic review suggested this indicates possible Type I errors.
A 1999 follow-up by William S. Harris et al. attempted to replicate Byrd’s findings under stricter experimental conditions, noting that the original research was not completely blinded and was limited to only “prayer-receptive” individuals (57 of the 450 patients invited to participate in the study refused to give consent “for personal reasons or religious convictions”). Using a different, continuous weighted scoring system – which admittedly was, like Byrd’s scoring, “an unvalidated measure of CCU outcomes” – Harris et al. concluded that “supplementary, remote, blinded, intercessory prayer produced a measurable improvement in the medical outcomes of critically ill patients”, and suggested that “prayer be an effective adjunct to standard medical care.” However, when they applied Byrd’s scores to their data, they could not document an effect of prayer using his scoring method. Critics have suggested that both Byrd’s and Harris’s results can be explained by chance. Richard P. Sloan compared the Byrd and Harris studies with the sharpshooter fallacy, “searching through the data until a significant effect is found, then drawing the bull’s-eye.”
A 1997 study by O’Laoire measured the effects on the agents performing daily prayers and reported benefits not only for the beneficiaries, but also for the agents, and the benefit levels correlated with the belief levels of agents and beneficiaries in some cases. The study measured anxiety and depression. This study used beneficiary names as well as photographs.
In 1998 Fred Sicher et al. performed a small scale double-blind randomized study of 40 patients with advanced AIDS. The patients were in category C-3 with CD4 cell counts below 200 and each had at least one case of AIDS-defining illness. The patients were randomly assigned to receive distant intercessory healing or none at all. The intercession took place by people in different parts of the United States who never had any contact with the patients. Both patients and physicians were blind to who received or did not receive intercession. Six months later the prayer group had significantly fewer AIDS illnesses, less frequent doctor visits, and fewer days in the hospital. However, CD4 counts and scores on other physiological tests had no significant variation between the two groups of patients.
A 2001 double-blind study at the Mayo Clinic randomized 799 discharged coronary surgery patients into a control group and an intercessory prayer group, which received prayers at least once a week from 5 intercessors per patient. Analyzing “primary end points” (death, cardiac arrest, rehospitalization, etc.) after 26 weeks, the researchers concluded “intercessory prayer had no significant effect on medical outcomes after hospitalization in a coronary care unit.”
The IVF-ET prayer scandal
In 2001 the Journal of Reproductive Medicine published an experimental study by three Columbia University researchers indicating that prayer for women undergoing in vitro fertilization-embryo transfer (IVF-ET) resulted in a success rate (50%) of pregnancy double that of women who did not receive prayer. Columbia University issued a news release saying that the study had been carefully designed to eliminate bias. The most vocal skeptic was Bruce Flamm, a clinical professor of gynecology and obstetrics at the University of California at Irvine, who found that the experimental procedures were flawed. One of the study’s authors, Cha, responded to criticism of the study in the November 2004 issue of JRM. In December 2001, the U.S. Department of Health and Human Services’ (DHHS) office for Human Research Protections (OHRP) confirmed a report by the Columbia University Health Sciences Division that one of the study’s authors, Rogerio Lobo, only learned of the study six to twelve months after the study was completed, and that he had only provided editorial assistance. The name of Columbia University and Lobo were retracted from the study.
Retroactive intercessory prayer
A 2001 study by Leonard Leibovici used records of 3,393 patients who had developed blood infections at the Rabin Medical Center between 1990 and 1996 to study retroactive intercessory prayer. To compound the alleged miraculous power of prayer itself, the prayers were performed after the patients had already left the hospital. All 3,393 patients were those in the hospital between 1990 and 1996, and the prayers were conducted in 2000. Two of the outcomes, length of stay in the hospital and duration of fever, were found to be significantly improved in the intervention group, implying that prayer can even change events in the past. However, the “mortality rate was lower in the intervention group, but the difference between the groups was not significant.” Leibovici concluded that “Remote, retroactive intercessory prayer was associated with a shorter stay in hospital and a shorter duration of fever in patients with a bloodstream infection.” Leibovici goes on to note that in the past, people knew the way to prevent diseases (he cites scurvy) without understanding why it worked. In saying so, he suggests that if prayer truly does have a positive effect on patients in hospital, then there may be a naturalist explanation for it that we do not yet understand. After many scientists and scholars criticized this retroactive study, Leibovici stated in 2002 that the “article has nothing to do with religion. I believe that prayer is a real comfort and help to a believer. I do not believe it should be tested in controlled trials.” The study has been summarised as being “intended lightheartedly to illustrate the importance of asking research questions that fit with scientific models.”
In 2003, Larry Dossey, the executive editor of the journal Explore: The Journal of Science & Healing and an advocate of faith healing co-authored a paper responding to Leibovici which discussed possible mechanisms to explain the results reported. Olshansky and Dossey invoked quantum mechanics to explain not only the benefits of intercessory prayer, but also how it might operate retroactively, drawing strong criticisms from physicist Victor Stenger and physician Jeffrey Bishop. The observer effect is regularly used to suggest that conscious control of physical reality is predicted by quantum mechanics, but this misconception “can be traced to a misinterpretation of wave-particle duality.” In relation to backwards causality, Stenger noted that “the results of some quantum experiments may be interpreted as evidence for events in the future affecting events in the past at the quantum level, [but] no theoretical basis exists for applying this notion on the macroscopic scale of human experience.” He concluded that while “the atoms in biological systems are quantum in nature … their collective behaviour does not exhibit any quantum effects. … What is more, even if the brain were a quantum system, that would not imply that it can break the laws of physics any more than electrons or photons, which are inarguably quanta.” One further point which illustrates that Dossey and Olshansky do not understand the physics they are using is seen in their invocation of quantum nonlocality in explaining backward causation, stating that “[r]etroactive prayer may be less absurd than [Leibovici] supposes, in the light of the discovery of non-local phenomena.” Unfortunately, the two are mutually incompatible: in allowing reverse causality in a model, the phenomenon of nonlocality ceases. Dossey has written in Explore about coining the term “nonlocal mind” in 1987, though quantum nonlocality goes back to a 1935 paper by Einstein, Podolsky, and Rosen. Olshansky and Dossey defended their work from various critics in the British Medical Journal’s rapid response section
Olensky and Dossey are not alone amongst alternative medicine proponents in having missed the point which Leibovici was making. In 2004, Stephen Wright described the Olshansky and Dossey contribution as a “thoughtful essay,” and it was praised by an editorial in the Journal of Alternative and Complementary Medicine the same year. In 2005, Olshansky and Dossey’s work was included in a critical review published in Explore which concluded that “Religious activity may improve health outcomes.” Their work was also defended in the British Medical Journal itself in 2004. Dossey authored an Explore paper defending experiments on the medical effects of prayer in 2005.
The MANTRA study
A 2005 MANTRA (Monitoring and Actualisation of Noetic Trainings) II study conducted a three-year clinical trial led by Duke University comparing intercessory prayer and MIT (Music, Imagery, and Touch) therapies for 748 cardiology patients. The study is regarded as the first time rigorous scientific protocols were applied on a large scale to assess the feasibility of intercessory prayer and other healing practices. The study produced null results and the authors concluded, “Neither masked prayer nor MIT therapy significantly improved clinical outcome after elective catheterization or percutaneous coronary intervention.” Neither study specified whether photographs were used or whether belief levels were measured in the agents or those performing the prayers.
The STEP project
Harvard professor Herbert Benson performed a “Study of the Therapeutic Effects of Intercessory Prayer (STEP)” in 2006. The STEP, commonly called the “Templeton Foundation prayer study” or “Great Prayer Experiment”, used 1,802 coronary artery bypass surgery patients at six hospitals. Using double-blind protocols, patients were randomized into three groups, individual prayer receptiveness was not measured. The members of the experimental and control Groups 1 and 2 were informed they might or might not receive prayers, and only Group 1 received prayers. Group 3, which served as a test for possible psychosomatic effects, was informed they would receive prayers and subsequently did. Unlike some other studies, STEP attempted to standardize the prayer method. Only first names and last initial for patients were provided and no photographs were supplied. The congregations of three Christian churches who prayed for the patients “were allowed to pray in their own manner, but they were instructed to include the following phrase in their prayers: “for a successful surgery with a quick, healthy recovery and no complications”. Some participants complained that this mechanical way they were told to pray as part of the experiment was unusual for them. Complications of surgery occurred in 52 percent of those who received prayer (Group 1), 51 percent of those who did not receive it (Group 2), and 59 percent of patients who knew they would receive prayers (Group 3). There were no statistically significant differences in major complications or thirty-day mortality. In The God Delusion, evolutionary biologist Richard Dawkins wrote, “It seems more probable that those patients who knew they were being prayed for suffered additional stress in consequence: performance anxiety’, as the experimenters put it. Dr Charles Bethea, one of the researchers, said, “It may have made them uncertain, wondering am I so sick they had to call in their prayer team?'” Study co-author Jeffery Dusek stated that: “Each study builds on others, and STEP advanced the design beyond what had been previously done. The findings, however, could well be due to the study limitations.” Team leader Benson stated that STEP was not the last word on the effects of intercessory prayer and that questions raised by the study will require additional answers.
A meta-analysis of several studies related to distant intercessory healing was published in the Annals of Internal Medicine in 2000. The authors analyzed 23 trials of 2,774 patients. Five of the trials were for prayer as the distant healing method, 11 were with noncontact touch, and 7 were other forms. Of these trials, 13 showed statistically significant beneficial treatment results, 9 showed no effect, and 1 showed a negative result. The authors concluded that it is difficult to draw conclusions regarding distant healing and suggested further studies.
A 2003 levels of evidence review found “some” evidence for the hypothesis that “Being prayed for improves physical recovery from acute illness”. It concluded that although “a number of studies” have tested this hypothesis, “only three have sufficient rigor for review here” (Byrd 1988, Harris et al. 1999, and Sicher et al. 1998). In all three, “the strongest findings were for the variables that were evaluated most subjectively. This raises concerns about the possible inadvertent unmasking of the outcomes assessors. Moreover, the absence of a clearly plausible biological mechanism by which such a treatment could influence hard medical outcome results in the inclination to be skeptical of results.” This 2003 review was performed before the 2005 MANTRA study and the 2006 STEP project, neither of which were conclusive in establishing the efficacy of prayer.
Various broader meta-studies of the literature in the field have been performed showing evidence only for no effect or a potentially small effect. For instance, a 2006 meta analysis on 14 studies concluded that “There is no scientifically discernable effect for intercessory prayer as assessed in controlled studies”.However, a 2007 systemic review of 17 intercessory prayer studies found “small, but significant, effect sizes for the use of intercessory prayer” in 7 studies, but “prayer was unassociated with positive improvement in the condition of client” in the other 10, concluding that based upon the American Psychology Association’s Division 12 (clinical psychology) criteria for evidence-based practice, intercessory prayer “must be classified as an experimental intervention.” The review noted that the most methodologically rigorous studies had failed to produce significant findings.
Adapted from Wikipedia, the free encyclopedia