Koh 2009.

Methods	RCT (cluster randomised)
Participants	Setting: two acute care hospitals, Singapore N = 1122 patients. 2 clusters. Sample: 641 nurses in medical, surgical and geriatric units in the two hospitals (% female patients not stated) Age (years) patients: mean 68 Inclusion criteria: all patients Exclusion criteria: none stated
Interventions	Multifaceted strategy for implementation of Ministry of Health Fall Prevention Clinical Practice Guideline (CPG) Revision of hospital's fall prevention policy in line with CPG Identification of change champions from within staff Educational sessions for staff aimed at promoting and supporting the adoption of the recommendations Reminders and identification systems, e.g. mandatory fall risk‐assessment tool in nursing assessment notes, posters in ward toilets, high‐risk patients identified by pink name card above the bed, pink stickers on clinical/nursing notes, and pink identification bracelets Audit and feedback on incidence of falls and compliance with use of risk assessment tool Control: routine dissemination strategies for implementation of CPG
Outcomes	Rate of falls
Duration of the study	6 months
Notes	Intervention targeted nursing staff. Age of patients not stated in Koh 2009. Obtained by personal communication with author
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "The two study hospitals were randomly allocated either to the "intervention" site... or the "control” site". Author states carried out by supervised coin toss; heads gets the intervention
Allocation concealment (selection bias)	High risk	No concealment. After first site randomised, second site automatically becomes the control group
Blinding of participants and personnel (performance bias) All outcomes	High risk	Judgement comment: not blinded
Blinding of outcome assessment (detection bias) All outcomes	High risk	Quote: "Fall incidence and fall‐associated injury rates were obtained from the hospitals’ fall incidence database"
Incomplete outcome data (attrition bias) All outcomes	High risk	Judgement comment: falls data for a random sample of medical records used. How representative these are of all patients and what proportion unknown
Selective reporting (reporting bias)	Unclear risk	Judgement comment: no protocol identified
Method of ascertaining falls	High risk	Judgement comment: falls determined through audits of hospital records. Definitions and practices may vary between hospitals.
Baseline imbalance	Unclear risk	Judgement comment: baseline characteristics of patients not reported
Other bias	Low risk	Judgement comment: none detected