Design: Randomized, controlled, double-blind study.
Patients: 496 volunteers: those who prayed (agents, n = 90) and those who were prayed for (subjects, n = 406).
Intervention: Agents were randomly assigned to either a directed or nondirected prayer group; photos and names of subjects were used as a focus. Subjects were randomly assigned to three groups: those prayed for by nondirected agents, a control group, and those prayed for by directed agents. Prayer was offered for 15 minutes daily for 12 weeks. Each subject was prayed for by three agents.
Main outcome measures: Five pretest and posttest objective measures and six posttest subjective measures were taken.
Results: Subjects improved significantly on all 11 measures. Agents improved significantly on 10 measures. A significant positive correlation was found between the amount of prayer the agents did and their scores on the five objective tests. Agents had significantly better scores than did subjects on all objective measures. Subjects' views of the locus of God's action showed significance in three objective measures. Improvement on four objective measures was significantly related to subjects' belief in the power of prayer for others. Improvement on all II measures was significantly related to subjects' conviction concerning whether they had been assigned to a control or an experimental group. Possible explanations include the placebo/faith effect, the time displaced effect, and extraneous prayer.