Table 1

Characteristics of 16 UK observational studies and reported medication administration error (MAE) rates

StudyStudy settingData collectionWere participants told the purpose of the study?IV or non-IV dosesSample sizeReported MAE rate* (95% CI)
Dean et al101 hospital, 6 wards: 2 medical, 2 surgical, 2 MfEConsecutive weekday drug rounds (May and June 1993)Partially (a study of medication administration and work sampling conducted concurrently)Non-IV2756 OE3.0% (2.4 to 3.7)
Ridge et al201 hospital, 6 wards: 2 medical, 2 surgical, 2 MfE1 week on each ward (at least 10 drug rounds), between January and April 1993No (work sampling study)Both3312 drug administrations3.5% (2.9 to 4.1)
Gethins211 hospital, 5 wards: 4 medical, 1 renal6 week periodNo (a time and motion survey)Non-IV2000 drug administrations2.8%
Ho et al221 female MfE wardTwo 8-day periods with 1 week break in betweenPartially (a study of the problems associated with the medication distribution system)Non-IV2170 OE5.5% (4.5 to 6.4)
Cavell and Hughes232 hospitals, 2 medical wards42 drug rounds on ward with handwritten charts (H). 35 drug rounds on ward with computer-printed administration form (C)Partially (a study of computerised prescribing on the drug use process)Non-IV1206 OE (H)
1295 OE (C)
5.5% (H)
5.7% (C)
Hartley and Dhillon241 hospital. 3 wards: 2 surgical, 1 medical39 consecutive days in June and July 1996Partially (a study to understand the constraints the nurses operated under and to improve the provision of information for their needs)IV320 prescribed doses26.9% (20.3 to 30.7)
Taxis and Barber251 hospital, 2 general medical5 weekdays on each ward in May and July 1997. All scheduled drug rounds except where two nurses administered using separate trolleysPartially (a study of advantages and disadvantages of each system)Non-IV842 OE8.0% (6.2 to 9.8)
Dean and Barber261 hospital: 1 vascular surgery and 1 renal medical wardTotal 27 days before bedside lockers were implemented, and 17 days post. All four scheduled drug rounds 7 days a week (January–June 1998)Partially (a study to find out how often medication was unavailable, could not be found, or whether any other problems occurred)Non-IV3576 OE (pre)
2491 OE (post)
4.3% (pre)
4.2% (post)
Bruce and Wong271 acute admissions ward4 weeks, each weekday in December 1998No (a study of time spent on drug administrations)IV (except 1 SC, 1 IM)107 OE10.3% (3.8 to 14.9)
Taxis and Barber282 hospitals, 10 wards: 1 renal, 2 medical, 1 CTS, 1 surgical, 1 ICU, 1 oncology, 1 neonatal, 1 CICU, and 1 paediatric.6–10 consecutive days on each ward between June and December 1999. Included weekends and all times of drug rounds on each wardPartially (a study of common preparation and administration problems of IV drugs)IV430 observed doses49% (45 to 54)
Wirtz et al111 hospital. Number and types of ward not stated6 consecutive days in each ward, May–June 2000Partially (a study of problems associated with preparing and administering IV drugs)IV77 preparations, 63 administrations22% (prep) (13 to 31)
27% (admin) (16 to 38)
Franklin et al291 mixed medical ward4 weeks pre internet-education for nursing staff (June 2004), 4 weeks post (Oct/Nov 2004)Partially (a study of drug administration problems)Both1188 OE (pre)
1308 OE (post)
6.9% (pre)
5.0% (post)
Franklin et al301 general surgical ward2 weeks pre-EPMA (spring 2007) and 2 weeks post-EPMA (spring 2008)Partially (a study of any problems associated with the medication system)Both1644 OE (pre)
1178 OE (post)
8.6% (pre)
4.4% (post)
Conroy et al311 children's hospital. Included PICU, NICU, medical, surgical, ED6 weeks, usually two drug rounds each weekdayYesBoth752 administrations1.2%
Ghaleb et al325 hospitals, 10 wards: 4 medical, 1 adolescent, 2PICU, 2NICU, 1surgical2 week period on each ward (2004/2005) each day, including weekendsYesBoth1554 doses; 2249 OE27.6% of doses19.9% of OE (17.5 to 20.7)
Kelly et al354 hospitals, 8 wards: 1 MfE and 1 stroke ward per hospitalMarch to June 2008. Morning and lunchtime drug rounds on some weekdays and weekendsYesNon-IV2129 OE10.7%
  • *Wrong time errors were excluded from reported MAE rates where applicable.

  • †Comparison study of UK and USA hospital (only UK data is presented).

  • ‡Comparison study of UK and German hospital (only UK data is presented).

  • CICU, cardiac intensive care unit; CTS, cardio-thoracic surgery; ED, emergency department; EPMA, electronic prescribing and medication administration system; ICU, intensive care unit; IM, intramuscular; IV, intravenous; MfE, medicine for the elderly; NICU, neonatal intensive care unit; OE, opportunities for error; PICU, paediatric intensive care unit; SC, subcutaneous.