Refer to The God Delusion by Richard Dawkins, or directly to the source: American Heart Journal vol. 151:4, "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". Here's the abstract.
The results were roughly as follows:
- Those who didn't know whether they were receiving prayers showed no difference between the sub-group which really was receiving prayers and the sub-group which wasn't.
- Those who knew they were NOT receiving prayers did better than those who were receiving prayers and knew it.
The second point doesn't suggest that prayers are harmful in and of themselves. The going hypothesis is that the knowledge of prayers was a kind of social pressure which tended to result in complications.
- SmartLX