James records in his gospel that “the prayer of faith shall save the sick” 1, and according to the U.S. Centre for disease control and prevention, 45% of adults use prayer for personal health reasons in the modern age.2 Unfortunately, it is very difficult to measure the power of prayer as a part of the treatment of someone, for quite a number of reasons: firstly, because some religious groups claim that praying as part of a scientific study is artificial, so the prayers will not be answered. There are also various confounding variables, such as ability to pray (according to some, prayer is a skill to be learned )3 and fervour of prayer (this may affect the effectiveness of prayer)4. Another reason that it is difficult to study is that it receives very little funding; according to the Washington Post, “about $5 million is spent worldwide on such research”5. Critics of scientific study of prayer argue that part of the point of prayer is that it is not always answered, and that god is not a mechanism that responds in a way that is predictable, or even measurable, by humans. For these reasons, and because in such a religiously charged subject, some studies are bound to be biased, research into prayer is still very much being refined, and there are conflicting results at the moment.
There have been three main types of prayer studied scientifically to gauge their efficacy. The most obvious one is second person prayer, which is performed by someone who knows the person they are praying about. Second person prayers directly ask a higher power for the person to be helped, as do third person prayers, which are performed by a stranger. First person prayer not only includes the petition to god but also can include self-meditation beforehand, and is a separate type of prayer.
Third person prayer and divine intervention
In order to test second and third person prayers, both blinded and non-blinded experiments have been set up. The general format is as such: two groups of people in hospital for similar problems are chosen (for example, people with the same cancer). One group of people is prayed for, while the other group is not. The health of the patients is measured, either using variables like blood pressure and heart rate, or by more subjective measures such as perceived comfort or number of visits required by a doctor. In third party experiments, the patients are prayed for by strangers, whereas in second person experiments, the friends and family are asked to pray for their loved ones. Again, there is difficulty in ensuring that a fixed number of people are praying - there is nothing stopping friends who are not part of the study from asking for the patient's good health. In a blinded experiment, the only people that know who is being prayed for are the people doing the praying, i.e. neither the doctor nor the patient knows if the patient is being prayed for. This is the most scientifically rigorous method; in non-blinded studies it is impossible to rule out the placebo effect.
The best way to test whether there was any godly interference in aiding a patient's recovery is a third party, blinded trial. These come as close to having one independent variable as any prayer study can for a number of reasons. Firstly, the people praying have no affiliation to the patient, rather than having any personal attachment, so fervour of the prayer is not affected, as it might be in second party studies, where there could be differing levels of attachment to the patient by the people praying. Secondly, by artificially selecting who is doing the praying, it is possible to ensure everyone is of the same denomination, piety, and prayer skill level, simply by using one congregation from the same church. This also allows participants to be randomly selected, rather than only those with religious family and friends. Lastly, because it is blinded, there is no chance that the placebo effect could contribute to the findings. In experiments like this, there is also the opportunity to test the effect of a "placebo prayer".
Despite the relatively little amount of money spent on studying spiritual healing, there have been thousands of studies in the past 30 years.5 Not all of them fit these criteria, however, and not all are published in peer reviewed journals, either. One of the biggest and most thorough studies was STEP, the study on therapeutic effects of prayer,6 which studied 1802 patients recovering from the same surgery in 6 different hospitals. There were three groups: two which were told they may or may not receive intercessory prayer, and one which was told it definitely would. Of the first two, one group received prayer and the other did not. They were then monitored for 30 days following the surgery for any complications. Significant events and deaths were also measured. The group certain of receiving prayer had a 59% chance of complications. The two groups uncertain had a 51% (if they were not prayed for) or 52% (if they were not) chance of complications in the 30-day period. That is to say, there was no significant difference between whether or not they were actually prayed for, but knowing that they would be prayed for decreased the chance of an uncomplicated recovery. Part of the reason that the STEP program was important was because it is one of the few studies that primarily looked for negative results (or the lack thereof). This eliminates the bias prevalent in other studies towards looking for a benevolent god; in those studies, the only outcomes are positive results or no significant results. The existence of this bias highlights an interesting assumption in prayer studies, which is that the results, if there are any, are presumed to be beneficial to humans. The trial still did not address other issues, for example it still treats god as an impersonal, mechanical being who responds in calculable ways to given stimuli. Critics argue that there is no room for any answer other than "yes, prayers are answered" or "no, prayers are never answered"; there should be provisions for an answer along the lines of "not now", or "not always". 7
There are also various meta-analyses on the studies conducted in order to corroborate their results. These reviews usually select the literature they use with the criteria listed above, but some are more strict, and there is also a comment on the methodology of each experiment in the paper. The most recent results used 17 trials, and found significant effects of prayer in 7 of them.8 This is the general trend in such papers: there is some evidence that religious healing helps the patient, but those results tend to come from the more subjective measurements or less rigorous trials.8 Even these summaries of research are unclear and regularly contradict each other, unfortunately.
First person prayer and meditation
Even without miraculous help from a god, it is conceivable that the actual act of prayer could have some therapeutic effect. It was found that recitation of a Latin prayer or a mantra increased heart rate variability and baroreflex sensitivity (the receptors that effect changes in heart rate) in a 2001 study by Bernadi et al..9 This is known to be associated with better prognosis in heart conditions and is a good indicator of exercise tolerance, and oxygenation of the blood, amongst other things. The mechanism by which it effects this change is thought to be to do with the rate of respiration and its synchronisation with other rhythms in the body, such as rise and fall of blood pressure. It is known that a slow respiratory rate decreases myocardial ischaemia, and the waves in blood pressure are about 0.1 Hz, i.e. one every 10 seconds. Therefore, in order for respiration to be synchronous with blood pressure, an ideal breathing rate would be 6 breaths per minute.
In the Bernardi study, various factors were measured during normal breathing, normal talking, and recitation of religious mantras. As is shown in the figure, talking slowed down respiration, and recitation further slowed it down and synchronised it with blood flow and blood pressure waves.
Both the Ave Maria (in Latin) and a yoga mantra were studied, in order to test whether the different disciplines affected the results. Remarkably, both slowed the breathing down to almost 10 second cycles, or 6 per minute, which coincides with the spontaneous cycles of blood pressure (see figure), which suggests that they both evolved to have roughly the same rhythm. This breathing system also makes people feel calmer, and so it makes sense for people to prefer those prayers, which is how they have come in to common usage. There is also historical evidence that the mantras of
influenced those of the middle east, which the crusaders then brought back to
the western world to form the Christian prayers that we still use today.
Therefore, there is some evidence that the act of saying some prayers is
physiologically beneficial, especially in people who have some form of heart
disease or are at risk of a heart attack.
Prayer is often recorded through anecdotal evidence to be relaxing or empowering; it is more difficult to measure psychological symptoms, but studies have also been conducted on this. For example, in 1988, a study was conducted by Carlson et al., designed to test stress reduction effects of meditative prayer compared to relaxation training (and also a control group), using tension in the frontalis muscle and skin temperature as indicators of anxiety.10 It was found that this meditation produced significantly more stress reduction than the control group or the group undergoing relaxation training. It has been recorded, however, that prayer is not always successful, particularly if done alone.11 This is because people praying alone are more likely to only dwell on their guilt and failings, whereas in a group, there is a curative environment, guidance and support from others in the group. Thus is it possible to pray "incorrectly", from a medical point of view, although the physiological mechanisms behind the psychological symptoms are still unclear; it is associated with increased cortical activity, as when the body is alert (Levin, 2001), but little else is known.
There is some further evidence that "integrating prayer into day-to-day life" makes one a kinder, more empathic person.12 According to the study, based on qualitative judgements, they were less likely to get very irate, and more likely to be more friendly towards others, even individuals who were hard to get along with. This is probably because the same group of people are more likely to live by the principles of their religion (the trial only included Christians), which involves turning the other cheek and not judging other people.
The current research into intercessory prayer has inconclusive results, due in part to the methodology and inherently supernatural nature of the trials. In general there does seem to be a positive effect of distant prayer, although not statistically significant, and potentially because of the way results are recorded. In order for studies to progress, there needs to be an objective, quantitative measure of general health, which can be measured as "good" or "bad", and multiple repeats using different religions, methods of prayer, and levels of ability of prayer. Even then it is unlikely such a trial would be accepted by the religious community because of the problems with "artificial prayer".
Trials involving first person prayer are more successful - stress is reduced by it and recitation of some prayers helps to improve heart conditions and hypertension, amongst other things. There is a proposed mechanism for the latter, but the former would merit more research as well in order to help differentiate the biological methods of prayer and relaxation training as a stress reducer.
1.James 5:15, King James Version
2.P.M. Barnes et al., "Complementary and Alternative Medicine Use Among Adults:
United States, 2002", 2004
3. Pope Jean Paul II, "Apostolic letter, novo millennio ineunte", 2000
4. St. Louis de Montfort, "The secret of the rosary", c.17th century
5. Michelle Boorstein, "Study of health and religosity growing despite critiscism", 2008
6. Benson et al., "Study of the Therapeutic Effects of Intercessory Prayer (STEP) in cardiac bypass patients: a multicenter randomized trial of uncertainty and certainty of receiving intercessory prayer", 2006
7. M. Jantos and H. Kiat, "Prayer as medicine: how much have we learned?", 2007
8. D. R. Hodge, "A Systematic Review of the Empirical Literature on Intercessory Prayer", 2007
9. Bernardi et al., "Effect of rosary prayer and yoga mantras on autonomic cardiovascular rhythms: comparative study", 2001
10. J. Martin and C. Carlson, "Spiritual dimensions of health psychology", 1988
11. Parker and St. Johns, "Prayer Can Change Your Life", 1957
12. S. Morgan, "A research note on religion and morality: Are religious people nice people?", 1983