Byrd 1988.

Methods	Allocation: randomised (computer generated list). Blindness: double (participants, care‐givers and researchers blind to assignment. Duration: unclear ("for the remainder of admission" ‐ mean number of days in hospital ˜ 8, SD 8.8). Consent: given.
Participants	Diagnosis: congestive heart failure (129), cardiomegaly (126), acute myocardial infarction (109). N=393. Age: mean ˜ 59 years. Sex: 265M, 128F. History: just admitted to Coronary Care Unit.
Interventions	1. Intercessory prayer: standard medical care + IP (by 'born again multi‐denominational Christians' outside hospital, daily by 3‐7 intercessors until discharge). N=192. 2. Standard medical care. N=201. Intercessors had no contact with their assigned patients.
Outcomes	Death. Clinical state: good/poor, complications. Service use: number of days in hospital, number of days in CCU, readmission to CCU. Leaving the study early.
Notes	Multiple complications presented. Independent collaborator (Evandro Coutinho), blinded to data, selected 'Re‐admissions to CCU' as proxy for 'Complications'.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer generated list.
Allocation concealment (selection bias)	Unclear risk	Unclear
Blinding (performance bias and detection bias) All outcomes	Low risk	Those rating outcomes not aware group of allocation.
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Full reporting.
Selective reporting (reporting bias)	Unclear risk	Primary outcomes clearly reported, but many secondary outcomes also emphasised in the trial's report.
Other bias	High risk	Clear expression of prior belief in the positive effects of prayer.