Spirituality as Therapy
By Howell Sasser, PhD, Dr. Sasser is Director, Research Epidemiology, R. Stuart Dickson Institute for Health Studies, Carolinas HealthCare System, Charlotte, NC; he reports no consultant, stockholder, speaker's bureau, research, or other financial relationships with companies having ties to this field of study.
Part 2 of a Series on Spirituality
Conventional western medical practice is essentially pragmatic, focused on identifying problems and addressing them. This "do-something" approach informs how existing therapies are evaluated (evidence-based medicine) and how new treatments are tested (randomized clinical trials). As complementary and alternative therapies have gained in popularity, their proponents have adopted the prevailing methods to attempt to prove safety and efficacy. This article looks at some of the issues raised when these methods have been applied to prayer and the state of the evidence to date.
Prayer as an Intervention
The expression of a desire for healing, whether in specific or general terms, is surely one of the oldest reactions to illness. Virtually every culture has a tradition of appeal to a Higher Power (divinity, universal life force, Nature) to restore health either as a supplement to, or in place of, human efforts. While the degree of confidence in the chance of success of such practices may vary from one culture to the next, most share a sense that the help being sought is largely beyond human understanding and control.
The range of practices that might be found under the broad rubric of appeal to a Higher Power includes prayer, directed energy healing, shamanic practices, therapeutic touch, and other kinds of spiritual healing.1 Although clinical research has been conducted on a number of these modalities, prayer as practiced in the "revealed" religions (Judaism, Christianity, Islam) has received the most attention in the parts of the world that contribute most to the clinical trials literature. These traditions share a common understanding of a distinct, personal divinity with the capacity to intervene in the natural order.
Even within this comparatively homogeneous setting, prayer is understood and studied in various forms. Two common ones are intercessory and nondirected prayer.2 Intercessory prayer requests a specific outcome, such as the remission of a tumor or the safe delivery of a baby. Nondirected prayer does not seek a specific outcome, but rather expresses the wish that the will of God be done in or through the person being prayed for. In common practice, both forms may be used interchangeably or simultaneously.
Issues in Studies of Prayer
Trials of prayer as a clinical intervention have attracted criticism from both religious and scientific observers. At the most global scale, and common to the two camps, is the perception that the underlying motivation for such studies is broader than the stated objectives. From the scientific perspective, prayer trials could be seen as attempts to prove the existence of God (however understood) with the tools of science. If one or a series of prayer studies shows a significant benefit, this implies the existence of Someone to answer the prayers. The converse—proof of the non-existence of God through negative findings—is a potential concern from the religious perspective.3
Yet, to meet the typical medical expectation that a treatment be specific, it is difficult to avoid expressing an opinion both about the nature and expected outcome of prayer. It would be wasteful and unethical to do a clinical study of an intervention one did not believe to exist. By engaging in research in this area, one admits both to the potential existence of a Higher Power, and also to some conception of who or what that power is. If one understands the divine Other to be properly approached in one way (sacrifice or some other physical action), it would be illogical to design a study in which another approach (prayer or some other non-physical action) were used. Even to come to the study of prayer with no a priori assumption about how it might work is a kind of expression of belief. However, even if the researcher is assumed to have no social, cultural, or religious preconceptions, a number of potential practical issues remain in the design of such research.
Control of the "exposure" is difficult. Most investigators conducting studies of prayer would admit to an a priori assumption that the ultimate source of their experimental intervention (God) is not mutable. It is not possible to define the precise nature, intensity, and duration of the exercise of the divine will. It is also unknown—or at least untested—whether there is a correlation between the number of prayers offered for a specific intention and the likelihood of divine action. Even if these factors were known or could be held constant, there is the further issue that access to the intervention is not completely under the control of the investigator. A study participant may pray for himself or herself, be prayed for by friends and relatives, and benefit from nonspecific prayers offered by some faithful people for all the sick. This makes it difficult, if not impossible, to guarantee homogeneity within groups in a randomized study.
Assessment of the outcome is difficult. The majority of well-designed clinical studies have a clearly defined and measurable outcome. Studies of prayer again face a challenge in the nature of the intervention. Perhaps a study was unable to show that prayer prevented an infection, but perhaps the divine recipient of the prayers heard them and simply chose to respond in a way other than the one requested. Without doing violence to the conventional Judeo-Christian understanding of God, it is not possible to set precise goals and interpret failure to achieve them as evidence that prayer is not efficacious.
Findings are difficult to generalize. The mechanism of action by which prayer might have a clinical effect is understood in many different ways, even within a single faith or denomination. Some would view even the idea of testing prayer scientifically as blasphemous. Others would argue that prayer is intended to have as great or greater an effect on the one who prays as on the one prayed for. In any event, the lack of a biologically plausible explanation of how prayer might work makes it difficult to predict how prayer might affect a given patient. In defense of prayer studies, it should be noted that this limitation is not unique to this intervention. A variety of medical therapies vary in efficacy from patient to patient.
Clinical Studies
Despite these challenges, numerous studies have been conducted and the results published since the 1960s. Space does not permit a full review of the literature, but selected examples illustrate the evolution of the field.
In 1965, investigators in London published a small study (n = 38) of prayer in people with chronic diseases.4 Patients were pair-matched on several variables and one of each pair was assigned randomly to be prayed for by individuals or groups with no prior knowledge of or contact with him or her. Neither the treating physicians nor the participants were aware of the nature of the study. The investigators estimated that each participant in the active arm received at least 15 hours of prayer during the six-month study period. An independent physician who was unaware of treatment assignment evaluated each patient before and after the intervention. No significant difference was shown between the groups. Although this study was flawed—it included patients with both physical and psychological illnesses—it set a precedent for the study of prayer with standard clinical trial methods.
Byrd's 1988 study of prayer for patients in a San Francisco coronary care unit (CCU) is widely cited and reflects overall advances in clinical trial methods.5 He enrolled 393 patients and randomly assigned 192 to be prayed for. Informed consent was obtained and participants were aware of the nature of the study, although patients and study staff were blind to treatment assignment. All those providing the prayer intervention were described as "born again," and actively involved in a Christian community. Patients were prayed for daily until discharge from the CCU, and the prayers were intercessory in nature—asking for a rapid recovery and prevention of complications and death.
From a list of 26 possible complications, including death, cardiac events, infections, and the need for surgery, six (congestive heart failure, need for diuretics, cardiopulmonary arrest, pneumonia, need for antibiotics, and need for intubation/ventilation) occurred less frequently to a statistically significant degree in the prayer group. A composite score of post-enrollment hospital course also showed more favorable results in the prayer group (P < 0.01). Byrd is careful to point out that there was no effective way to preclude exposure to prayer among the controls, and also speculates about whether the appropriate unit of analysis is the individual or the group (i.e., perhaps God acted for the good of the treatment group as a whole, rather than for one or more individual members).
A study of kidney disease patients published in 2001 sought to take into account the element of participant expectancy regarding the prayer intervention.6 Ninety-five end-stage renal disease (ESRD) patients from an outpatient hemodialysis center were enrolled. They were told that during a six-week period, they would be prayed for by an individual or group, or "positively visualized" by a transpersonal (nonreligious) positive visualization group. In fact, only one-third of the participants were randomized to each of the interventions, with the remaining third serving—unwittingly—as a no-intervention control group. This created six intervention-by-expectation study arms (expected prayer/got prayer, expected prayer/got visualization, expected prayer/got nothing, expected visualization/got visualization, expected visualization/got prayer, expected visualization/got nothing). The interventions were conducted remotely by volunteers who were given basic information about, but who had no personal knowledge of, the participants. The structure of the interventions was standardized with respect to timing, duration, and format. Seven medical parameters typically measured in dialysis patients and scores on quality-of-life (SF-36) and psychological functioning questionnaires were used as outcomes.
Those expecting to receive prayer reported feeling significantly better than those expecting to receive positive visualization. Otherwise, there were no significant differences in medical or psychological outcomes by expectancy or by the intervention actually received. The authors' discussion raises the important issue of the relationship between the study participants and the intercessors. If prayer is more than instrumental—that is, if it is efficacious through relationships within and among people as much as through relationships between people and a higher power—it might be inappropriate to study it in an impersonal or anonymous setting.
Perhaps the best known and most ambitious prayer study to date is the Monitoring and Actualisation of Noetic Trainings (MANTRA II) trial.7 Between 1999 and 2002, 748 patients undergoing invasive cardiac procedures were enrolled at nine clinical sites. They were randomly assigned by a 2 × 2 factorial design to prayer or no prayer, and a composite music/imagery/touch (MIT) intervention or no MIT. From study inception until September 2001, an established religious congregation (Christian, Jewish, Muslim, or Buddhist) was given the name, age, and medical condition of each patient assigned to the prayer intervention and asked to pray for that person according to its usual practices— this equated in practice to a period of 5-30 days. In the final year of the study, a second tier of prayer was added, with additional congregations praying for those praying for the study patients. The MIT therapy was standard through the study, with one 40-minute session before the cardiac procedure. Patients were followed for six months after enrollment for in-hospital major adverse cardiovascular events (MACE), death, or readmission to the hospital.
There were no differences in any of the outcomes to six months between the prayer randomization groups, and only in the hazard rate for death by six months between the MIT groups—those receiving MIT were less likely to die within six months, although this difference was not statistically significant. There also were no differences in outcomes between those enrolled under the original prayer schema and those enrolled under the later, two-tiered system. Finally, no factorial combination (prayer only, MIT only, prayer and MIT, or standard care) showed significantly improved outcomes to six months.
A final example will illustrate the extent to which conventional assumptions may be questioned in this area of research. Leibovici conducted a retrospective experiment with in-hospital septicemia.8 Arguing that God might not be limited to a linear conception of time, in 2000 he "enrolled" and reviewed the records of 3,393 patients who had been admitted to a hospital in Israel with a detected blood infection between 1990 and 1996. These he assigned randomly to intervention or control. The first names of those in the intervention group were given to another person who said a prayer for them collectively. There was no difference in mortality between the study groups, but those who were retroactively prayed for had shorter average duration of fever and length of stay (P = 0.04 and P = 0.01, respectively). Even if this study was perhaps a bit tongue-in-cheek, it does emphasize the hazards of looking for God in potentially random events.
Conclusion
This article was not intended to be an exhaustive review of the published findings of randomized research on prayer, or to reach a firm conclusion as to the direction of the evidence. If anything, its purpose was to offer for consideration some of the issues that make such research complex, and that might suggest that any appropriate conclusion cannot be expressed in absolutes. It is worthy of note that both proponents of the scientific approach (or what is sometimes referred to pejoratively as "Scientism") and those with strong religious beliefs find reason to question the validity of clinical trials of spiritual practices. Since the advent of modern medicine empirical evidence seems to show that the clinical effect of prayer is subtle in comparison with drugs and surgical techniques. Perhaps the true effect of prayer awaits demonstration in a much larger trial, one powered to detect a whisper among shouts.
Recommendation
None of the articles reviewed here, or most of the other peer-reviewed studies published to date, advocate replacing conventional treatment with prayer. The evidence also does not appear to favor making prayer a routine part of every patient's medical care. However, for those who are receptive to it, there is some evidence that prayer can have benefit. This seems to be especially true when the patient knows or believes that he or she will be prayed for, and when the pray-er and the prayed-for know each other. A physician who has an ongoing clinical relationship with a patient should probably already know whether the patient has some kind of spiritual life (and if not, should not be shy about asking). With this information in hand, in suitable situations, a question like, "Do you have someone praying for/thinking about you?" should not be too awkward to ask.
References
1. Targ E. Research methodology for studies of prayer and distant healing. Complement Ther Nurs Midwifery 2002;8:29-41.
2. O'Laoire S. An experimental study of the effects of distant, intercessory prayer on self-esteem, anxiety, and depression. Altern Ther Health Med 1997;3:38-52.
3. Dossey L. Running scared: How we hide from who we are. Altern Ther Health Med 1997;3:8-15.
4. Joyce CR, Welldon RM. The objective efficacy of prayer: A double-blind clinical trial. J Chronic Dis 1965;18:367-377.
5. Byrd RC. Positive therapeutic effects of intercessory prayer in a coronary care unit population. South Med J 1988;1:826-829.
6. Mathews WJ, et al. The effects of intercessory prayer, positive visualization, and expectancy on the well-being of kidney dialysis patients. Altern Ther Health Med 2001;7:42-52.
7. Krucoff MW, et al. Music, imagery, touch, and prayer as adjuncts to interventional cardiac care: the Monitoring and Actualisation of Noetic Trainings (MANTRA) II randomised study. Lancet 2005;366:211-217.
8. Leibovici L. Effects of remote, retroactive intercessory prayer on outcomes in patients with bloodstream infection: Randomized controlled trial. Lancet 2001;323:1450-1451.
Sasser H. Spirituality as therapy: Part 3 in a series on spirituality. Altern Med Alert 2006;9(11):121-124.You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content