Monsey Chan McLeod1, Nick Barber, Bryony Dean Franklin. 1. Centre for Medication Safety and Service Quality, Pharmacy Department, Imperial College Healthcare NHS Trust,Charing Cross Hospital, London W6 8RF, UK. monsey.mcleod@imperial.nhs.uk
Abstract
BACKGROUND: Medication administration errors (MAEs) are a problem, yet methodological variation between studies presents a potential barrier to understanding how best to increase safety. Using the UK as a case-study, we systematically summarised methodological variations in MAE studies, and their effects on reported MAE rates. METHODS: Nine healthcare databases were searched for quantitative observational MAE studies in UK hospitals. Methodological variations were analysed and meta-analysis of MAE rates performed using studies that used the same definitions. Odds ratios (OR) were calculated to compare MAE rates between intravenous (IV) and non-IV doses, and between paediatric and adult doses. RESULTS: We identified 16 unique studies reporting three MAE definitions, 44 MAE subcategories and four different denominators. Overall adult MAE rates were 5.6% of a total of 21 533 non-IV opportunities for error (OE) (95% CI 4.6% to 6.7%) and 35% of a total of 154 IV OEs (95% CI 2% to 68%). MAEs were five times more likely in IV than non-IV doses (pooled OR 5.1; 95% CI 3.5 to 7.5). Including timing errors of ±30 min increased the MAE rate from 27% to 69% of 320 IV doses in one study. Five studies were unclear as to whether the denominator included dose omissions; omissions accounted for 0%-13% of IV doses and 1.8%-5.1% of non-IV doses. CONCLUSIONS: Wide methodological variations exist even within one country, some with significant effects on reported MAE rates. We have made recommendations for future MAE studies; these may be applied both within and outside the UK.
BACKGROUND: Medication administration errors (MAEs) are a problem, yet methodological variation between studies presents a potential barrier to understanding how best to increase safety. Using the UK as a case-study, we systematically summarised methodological variations in MAE studies, and their effects on reported MAE rates. METHODS: Nine healthcare databases were searched for quantitative observational MAE studies in UK hospitals. Methodological variations were analysed and meta-analysis of MAE rates performed using studies that used the same definitions. Odds ratios (OR) were calculated to compare MAE rates between intravenous (IV) and non-IV doses, and between paediatric and adult doses. RESULTS: We identified 16 unique studies reporting three MAE definitions, 44 MAE subcategories and four different denominators. Overall adult MAE rates were 5.6% of a total of 21 533 non-IV opportunities for error (OE) (95% CI 4.6% to 6.7%) and 35% of a total of 154 IV OEs (95% CI 2% to 68%). MAEs were five times more likely in IV than non-IV doses (pooled OR 5.1; 95% CI 3.5 to 7.5). Including timing errors of ±30 min increased the MAE rate from 27% to 69% of 320 IV doses in one study. Five studies were unclear as to whether the denominator included dose omissions; omissions accounted for 0%-13% of IV doses and 1.8%-5.1% of non-IV doses. CONCLUSIONS: Wide methodological variations exist even within one country, some with significant effects on reported MAE rates. We have made recommendations for future MAE studies; these may be applied both within and outside the UK.
Authors: Josephine Henry Basil; Chandini Menon Premakumar; Adliah Mhd Ali; Nurul Ain Mohd Tahir; Noraida Mohamed Shah Journal: Drug Saf Date: 2022-10-03 Impact factor: 5.228
Authors: Tera L Reynolds; Patricia R DeLucia; Karen A Esquibel; Todd Gage; Noah J Wheeler; J Adam Randell; James G Stevenson; Kai Zheng Journal: JAMIA Open Date: 2019-01-04