Pray For the Health of It
By Nick Gerlich

A study recently reported in the media suggests that intercessory prayer for the sick really works. Furthermore, it doesn't even matter if the patients know they're being prayed for, or know the people praying for them.

The study was conducted in 1998 by researchers at the Mid-America Heart Institute at Saint Luke's Hospital in Kansas City. The article was published 25 October 1999 in The Archives of Internal Medicine. A syndicated story circulated by Knight Ridder Newspapers concluded that the study's results offered the most compelling evidence to date of the link between improved health and what is known as "remote intercessory prayer."

The study involved a sample of 990 patients, 524 of whom were randomly assigned to a control group. The remainder, 466, were prayed for daily by five separate people. Seventy-five volunteers, split into groups of five, helped out by praying for specific individuals.

Patients and physicians were not aware of the study or that patients were being prayed for. The volunteers, who prayed away from the hospital, were not told any details about the persons for whom they were praying, other than first names and that the person was ill. Each patient was the unknowing recipient of prayer for 28 days, commencing one day after their admittance to the hospital.

Results from the study revealed that, "The patients who were prayed for just did better," according to William Harris, who led the study. A total of thirty-five factors were examined, including the numbers and types of medications patients needed, whether they needed respirators or pacemakers, the length of their hospital stay, speed of recovery, and also whether the patient lived or died.

Statistically, the results showed that those who were prayed for scored 11-percent better than the control group. The odds of such an occurrence were reported as being 1 in 27.

So, What Can We Make of This?

While this study seems to support the practice of praying for the sick, it does nothing more than trivialize medical research. Unfortunately, people of a religious nature are claiming it as proof positive of the power of prayer. For example, in the Amarillo (TX) Globe-Times publication of this story, the article was amended with remarks from local ministers, each of whom concluded that the study verified the value of Christian prayer. Their Christian anecdotes were now seemingly proven by this one study, destined now to become a landmark in all Christendom.

The study suffers from a number of shortcomings, including the following:

  • While the study and its results may be fodder for the pulpit, nowhere in the study did it mention to whom these prayers were directed, be it the Christian God, or anyone else. They were just prayers.
  • What is prayer? Is there a specific way to pray, or does it not matter? And to whom should these prayers be directed? Are they just scattered to the great unknown, and then telepathically received by the beneficiary? How this variable was operationalized needs to be addressed.
  • While it is usually desirable to have sufficiently large samples, it must also be noted that statistical tests are often vulnerable to the law of large numbers. In other words, as samples get larger and larger, even the most minute differences can become statistically significant because the sample size is inversely related to the value of t-statistics and other measures.
  • Does religious persuasion of either the person praying, or the beneficiary, have any effect? Did healings occur across religions? Did deaths and worsening health occur across religions?
  • If the treatment (in this case, prayer) were tangible, it would be easier to accept the results. However, the treatment is intangible, and may vary considerably across supplicants. Furthermore, if the results are indeed true, do the findings then indicate that a person could win the lottery or strike gold on Wall Street, all because some people were secretly pulling for him behind the scenes?
  • What about the patients themselves? What if they did not want to be prayed for? Can we conclude that the volunteer's prayers overcame personal aversion to prayer anyway?
  • There was nothing done to control for external sources of variation. For example, what if a patient (either control or treatment group) had family and friends praying for him? What would this do to the results? Furthermore, nothing was done to measure whether the patient was praying for himself, and whether there may have been any effects from positive thinking.
  • What is causing the results? Are the patients receiving the prayers directly, or is there an intermediary (i.e., God) who receives them, and doles out doses of good health to those specicifically named?
While the results are certainly good for those whose conditions improved, the study does nothing to improve our understanding of the effects of prayer. If anything, it only clouds our view.

As a person of the Christian persuasion, I agree with the practice of prayer. However, I am also a pragmatist, and recognize prayer for what it really is: the well-intentioned, loving best wishes of the supplicant for the beneficiary. There's nothing wrong with praying for someone, because it is one way we show our concern for a person.

As one of my colleagues once remarked, "Prayer is not for God. It's for the person praying." It brings the person closer to his God, and is an outpouring of his heart.

To even attempt to study this phenomenon is stretching the capabilities of the scientific method, and misses the point of prayer. It is impossible to control for all possible sources of variation, and, furthermore, beyond the realm of most theologies to ascertain whether God (whose ever) can be moved to act in response to prayer. More than anything, doctors should stick to healing the way they know how, and leave the subject of prayer to the faithful.

That's my prayer, and I'm sticking to it.