home

Author(s) Year	Sample Size & Condition	Intervention	Control Group	Main Outcome Measured	Outcome Intervention vs. Control	Comment
Joyce & Weldon 1965	48 patients with rheumatic disease or psychological conditions.	Prayers in Christian tradition	Standard care	Clinical or attitudinal improvements	Not Significant	Conclusion limited by small size of study.
Collipp 1965	18 children with acute leukemia	Prayers in Christian Tradition	Standard care	Survival	Favored intervention	Inconclusive because intervention and control groups not uniform. Triple blind - even intercessors did not know they were in a study.
Byrd 1988	393 patients admitted to coronary care unit	Prayers of “born again Christians”	Standard care	Frequency of complications and grading of hospital course	Favored intervention	Significant outcome variables not independent. Adequacy of blinding questioned.
Beutler et al 1988	96 patients with uncomplicated hypertension	Paranormal healing or laying on of hands with verbal communication	Standard care	Blood pressure level	Not Significant	Possible Hawthorne effect because BP decreased in all groups even before intervention began. Laying on of hands patients not double blinded.
Walker et al 1997	40 alcohol abuse patients	Prayers in Christian & Jewish traditions	Standard care	Level of alcohol consumption	Not Significant	Conclusion limited by small size of study. Heavier drinking among patients who knew someone outside of study praying for them.
O'Loire 1997	406 volunteers in response to advertisement	Directed prayers or non-directed prayers	No prayer	Psychological factors	Not Significant	Possible Hawthorne effect. Intercessors that prayed more improved more.
Sicher et al 1998	40 AIDS patients	Directed healing intentions-Christian, Jewish, Buddhist, Native Am., shamanistic traditions & bioenergetic & meditative healing	Standard care	AIDS defining illnesses, severity, and hospitalizati on	Favored intervention	Intervention and control groups differed at baseline. Study size small. CD4 count increases not significantly different in intervention and control groups
Harris et al 1999	990 patients admitted to cardiac care unit	Prayers in Christian tradition	Standard care	Outcome scoring system	Favored intervention	No informed consent. Statistics and validity of scoring system questioned.
Matthews, D et al 2000	40 rheumatoid arthritis patients	Christian prayers: Audible prayers with laying on of hands and supplementary distant prayers	Standard care	Signs, symptoms, and lab tests	Favored audible intervention. Distant intervention Not significant.	Not double blinded for audible prayers. Improvements not accompanied by improvements in lab tests. Small size limiting.
Harkness et al 2000	84 patients with warts	Flow/channeling distant healing	Standard care	Number and size of warts	Not Significant	Limited by small size of study.
Matthews, W et al 2001	95 patients with end-stage renal disease	Prayers in Catholic tradition or non-religious positive visualization	Standard care	Physical and psychological variables	Not Significant	Some patients received sham prayer and sham positive visualization. Those who expected to receive prayers reported feeling better.
Aviles et al 2001	799 patients discharged from coronary care unit	Prayers of local “religious” groups	No prayers	Death, cardiac arrest, coronary revascularization, cardiovascular ER visit or rehospitalization	Not Significant	Prayers began at discharge when outside prayers expe cted to be decreased.
Cha et al 2001	169 in vitro fertilization patients	Christian intercessors in US,Canada and Australia but patients all in Korea	Standard care	Pregnancy	Favored intervention	No informed consent. Second author indicted for fraud unrelated to this study. Existence of prayer groups questioned.
Leibovici 2000	3393 patients with blood stream infections	Retrospective prayers offered in 2000 for patients hospitalized in 1990-1996	No prayers	Death, length of stay, duration of fever	Favored intervention	Author’s justification for study - God not limited by linear time. Author intended to show that prayer should not be tested in controlled trials.
Krucoff et al 2005	748 patients having elective percutaneous coronary intervention or elective coronary catheterization	Prayers in Christian, Muslim, Jewish and Buddhist traditions. Music, imagery, and touch therapies	Standard care	Major cardiovascular event, readmission or death	Prayers not significant. Music, imagery, touch,. see comment.	Mortality at 6 months slightly lower with music, imagery, touch therapies.
Benson et al 2006	1802 patients admitted for coronary artery bypass graft surgery	Prayers in Catholic and Protestant traditions	Standard care	Post-operative complications within 30 days of surgery	Favored no intervention	Intentional partial blinding. Possible randomization problem. Could not control for background prayers.