Table 1

Synopsis of 24 identified studies

Author	Country	RCT	Intervention	Type of safety metric	Unit of measurement	Impact
Abramson	USA	No	Transition between EHRs	Medication safety	Clinicians	n.s.
Adelman	USA	Yes	Change in version of EHR	System safety: wrong patient orders	Clinicians	Identification-re-entry function resulted in lower error rate (p<0.001).
Awdishu	USA	Yes	Notification: AKI	Medication safety: AKI	Clinicians	Adjusted prescriptions increased (p<0.001).
Barnett	USA	No	Transition between EHRs	Adverse event reporting: PSI-90, death and readmissions	Patients	n.s.
Boockvar	USA	Yes	Link to community EHR	Medication safety: Reconciliation	Patients	n.s.
Cardozo	USA	No	Notification: Trauma	Clinical pathway: cervical-spine clearance protocol	Patients	Improved compliance rate with pathway.
Cho	USA	No	EHR generated lists	Alerts	Clinical unit	Reduction in catheter related infections (p<0.05).
Cho	Korea	No	Notification: Falls risk assessment	Falls	Patients	Unchanged rate of falls.
Colpaert	Belgium	No	Transition to electronic system	Medication safety	Patients	Reduction in prescription errors (p<0.001).
Cook	USA	No	Transition to electronic system	Medication safety: antibiotic prescribing	Patients	Reduction in nosocomial infections (p<0.07).
Dowding	USA	No	Transition to electronic system	Hospital acquired pressure ulcers and falls	Patients	Increased documentation rates for hospital acquired pressure ulcers.
Fahey	USA	No	Change in version of EHR	Medication safety: wrong dosage of chemotherapy	Clinicians	Decrease in dosage error (n=0) compared with manual rounding (n=4).
Hess	USA	No	Transition from paper to electronic system	Medication safety: wrong dosage in chemotherapy	Clinicians	n.s.
Mishra	USA	No	Notification: Medication dosage	Medication safety: monitoring of Vancomycin dosage	Patients	Increase in frequency of trough levels (p<0.01).
Mohsen	USA	No	Change in version of EHR	Venous thrombembolism Reduction in inappropriate alerts	Patients	Alert reduction (p<0.001), increase in alert effectiveness (p<0.001), but decrease in alert efficiency (p=0.007).
Muhlenkamp	USA	Yes	Notification: Dosage alerts	Medication safety: removal of inappropriate or unnecessary alerts	Patients	Decrease in dosage alerts by 3.6%.
Nanchal	USA	Yes	Change in version of EHR	ICU handover: occurrence of non-routine events	Clinicians	Structured sign-out process reduced the occurrence of non-routine events reported by residents (p=0.005).
Nendaz	Switzerland	Yes	Notification: VTE risk assessment	Medication safety: decision support for VTE prophylaxis	Patients	Less overprescribing with e-alerts (p<0.01).
Schnipper	USA	Yes	Medication Reconciliation	Medication safety: adverse drug events	Patients	Changes significant at discharge but not admission.
Silbernagel	Switzerland	Yes	Notification: Complications of Atrial fibrillation	Medication safety: anticoagulation	Patients	Adequate prescription increased from 16% to 22% (p=0.021).
Spirk	Switzerland	Yes	Notification: VTE prophylaxis	Medication safety: VTE prophylaxis	Patients	n.s.
Weiss	USA	Yes	Checklist in EHR	Medication Safety: Antibiotic prescribing	Patients	Increase in number of days with empirical antibiotics (p<0.002).
Westbrook	Australia	No	Implementation of two EHRs	Medication Safety	Patients	44% reduction in serious errors, increase in system errors.
Wilson	USA	Yes	Notification: AKI	Medication Safety: AKI	Patients	Increase in creatinine checks (p<0.05) and reduction in deaths and dialysis (p<0.01) only in surgical stratum.

AKI, acute kidney injury; EHR, electronic health record; ICU, intensive care unit; n.s., not significant; PSI, Patient Safety Indicator (PSI-90); PSI-90, Patient Safety and Adverse Events Composite for the International Classification of Diseases; RCT, randomised controlled trial; VTE, venous thromboembolism.