Table 1

Studies using an observational study design investigating double checking of medication administration

Study	Country	Sample size/study duration	Setting	Method of measuring double checking	Method of measuring errors	Findings	Study quality
Jarman 2002*27	Australia	14 months	Inpatient units, operating suites, birthing suite and ED at a 400-bed academic tertiary care hospital	None (before and after study of change in policy)	Incident report forms	4 administration errors were measured from March through September 2000 (when double checking was required) compared with 5 errors from March through September 2001 (when single checking became standard)	Poor
Manias 200531	Australia	175 administrations to 47 patients over 2 months	Metropolitan academic teaching hospital	Direct observation	–	Adherence rate was 97% for double checking of preparation and 80% for double checking to the patient’s bedside	Fair
Conroy 200730	UK	752 administrations to 253 patients over 6 weeks	Medical and surgical wards, PICU, NICU, ED in a 92-bed paediatric hospital	Direct observation	Direct observation	In 84% of patients, nurses were observed to double check administrations Cursory double checks were done on oral drug volumes and intravenous infusions in 3% of patients Independent checks of calculations were not obvious in 2% of patients Student nurses were allowed to administer unsupervised in 1% of patients	Fair
Alsulami 201428	UK	2000 administrations to 876 patients over 4 months	Medical and surgical wards, PICU, NICU in a paediatric hospital	Direct observation	Direct observation	Among 15 steps of independent double checking, adherence rates were equal or greater than 90% for 11 steps For the four other steps, adherence rates were 83% for the actual administration, 71% for rate of intravenous bolus, 67% for labelling of flush syringes and 30% for dose calculation	Good
Bulbul 201429	Turkey	98 nurses	Paediatric emergency, paediatric and neonatology, paediatric surgery wards in two teaching and research hospitals	Self-report	Self-report	64% of nurses reported double checking while preparing or administering high-risk drugs	Poor
Schilp 201432	Netherlands	2154 administrations of intravenous drugs over 1 year	ICU, internal medicine, general surgery and other departments administering intravenous drugs in 19 hospitals (2 academic, 6 tertiary teaching, 11 general)	Direct observation	–	Adherence to double checking was 52% for administrations of intravenous drugs	Fair
Härkänen 201534	Finland	1058 administrations to 122 patients over 2 months	Medical and surgical wards in an 800-bed academic hospital	Direct observation	Direct observation, medical records	In multivariate regression, double checking was significantly associated with a lower odds of any medication error (OR 0.44 (0.27 to 0.72)) Adherence to double checking was 81%	Good
Young 201533	USA	60 administrations to 47 paediatric and 10 adult patients over 24 days	198-bed paediatric inpatient hospital	Direct observation	Direct observation	Adherence to double checking was 75% (9 out of 12) among continuous intravenous administrations	Poor
Cochran 201626	USA	6497 administrations to 1374 patients	12 rural hospitals	Direct observation	Direct observation, medical records	16 of 29 (55%) preparation and administration errors occurred from administrations done with a single check, 9 (31%) with a double check and 4 (14%) with bar-code administration	Poor
Subramanyam 201619	USA	1473 intravenous infusions over 1 year	Paediatric patients undergoing radiological imaging at a tertiary academic paediatric hospital	Self-report	Self-report	Intercepted errors decreased from 4 per month to 1 per month Adherence to double checking was reported to be over 90%	Poor

*This study used a before-and-after design. All other studies were observational cohort studies.
ED, emergency department; NICU, neonatal intensive care unit; PICU, paediatric intensive care unit.