Let me admit to my skeptical readers here in this first sentence: Prayer experiments are riddled with scientific paradoxes. I get that. They prove nothing absolutely. Then again, neither do experiments in quantum reality. Do we debunk them as “not real science” because we can observe actions and reactions but can’t thoroughly explain them? Of course not. Science is just not as certain as it once was. It now too is a matter of faith…

That said, I want to post a few accounts of some of the most famous or infamous prayer/medical studies completed in the last 20-30 years. Again, I understand there are problems with these studies. My point isn’t to use them as proof of God or prayer but simply to highlight that SOMETHING is going on here…

One landmark study in intercessory prayer was conducted in 1988 by cardiologist Randolph Byrd at San Francisco General Hospital. Byrd built and executed a model that not only withstood (mostly) the scrutiny of methodological critique, but also demonstrated the first clear case of statistical significance from intercessory prayer. You can read his article at: Byrd, R .B. (1988). Positive therapeutic effects of intercessory prayer in a coronary care unit population.  Southern Medical Journal, 81, 826-829. 

Byrd, a cardiologist and practicing Christian began with two questions: 1) Does intercessory prayer to the Judeo-Christian God have any effect on the patient’s medical condition and recovery while in the hospital?  2) How are these effects characterized, if present?  To measure for these effects, Byrd randomly assigned 393 patients in his coronary care unit into two groups.  One group of 192 patients was prayed for by outside intercessors from around the country).  The intercessors were informed of the patients’ names and clinical status and asked to pray daily (until the patient was discharged) “for a rapid recovery and for prevention of complications and death, in addition to other areas of prayer they believed to be beneficial to the patient”. The second group of 201 patients, the control group, did not receive this experimental prayer.  All patients knew they were participating in a study on prayer but only the research nurse administrating the study knew which patients were in which group. Thus, the study met the double-blind, placebo-controlled criteria of typical clinical trials. 

Byrd collected data on each patient “in a blinded manner, without knowledge of the spiritual status, condition, or ideas of the entrants during the study”.  Once gathered, the data was entered into a PDP-11 computer using a medical statistical package called Biomedical Data Processing. 

Patients entering the study were assessed according to 30 specific health criteria and were then randomly assigned to one of the two groups.  Univariant and multivariant analysis of each patient’s condition at the time they entered the study revealed no statistical difference between these groups, anchoring the claim that an evaluation of the effects of intercessory prayer, the controlled for treatment, would be valid. 

After entry all patients were monitored for the remainder of their hospitalizations.  New problems, diagnoses, and interventions were catalogued.  Byrd used descriptive statistics to report percentages of medical conditions and medical procedures.  He used means +/- one standard deviation (stepwise logistic regression) for data utilizing multivariant analysis.  Results were also inferentially reported through chi-squares for categorical data and through unpaired t-tests for interval data.   

In addition, Byrd built a health scoring code grading the treatment course for each patient during their stay: good, intermediate, or bad.  This score was an interpretation of the cumulative health categories measured during hospitalization.  The hospital stay (course) was considered good if no new diagnoses were recorded for the patient.  The course as considered intermediate if there were somewhat higher levels of risk of death.  The course was considered bad for patients who had high risk of death. 

According to Byrd’s multivariant analysis of data gathered during the patient’ hospitalizations revealed a significant difference between the two groups (P < .0001).  Specifically, fewer patients in the prayer group required ventilatory support, antibiotics, or diuretics.   On the scoring levels, 85% of those in the prayer group were graded with a good course compared to 73% in the control group.  1% of the prayer group had intermediate courses against 5% in the control group.  And 14% of the prayer group had a bad course, while 22% of the control group took this score.   A 2 by 3 chi-square analysis of these data found that P <.01.  Thus, the statistical variance was significant: patients in the control group were nearly twice as likely to suffer complications, more than twice as likely to suffer heart failure, three times more likely to require diuretics and to suffer pneumonia, and nearly five times more likely to need antibiotics as were the patients in the prayer group.

Proof? No. But Byrd’s chutzpah to dare to test prayer in the laboratory did yield some tantalizing results.

What do you think?