BACKGROUND: Infusion devices can be programmed with individual hospitals' "best practice" rules for intravenous (i.v.) drug administration, and alerts can be provided if dosages fall outside pre-established limits. High variation levelsare common in medical care but can increase safety risk if the variation is unnecessary. METHODS: The i.v. best practice data sets of drugs from 100 hospitals using one manufacturer's infusion devices were compared to assess the number of drug names used and thevariation in concentrations, dose units, dose limits, and administration practices. RESULTS: The 100 hospitals showed an average of 64 drugs per data set and an average of 113 different drug/concentration 4 combinations. On average, each hospital had designated 6 profiles or unique patient care areas; there were 4 different names per drug across the hospitals (for example, amiodarone had 45 different names). High levels of variation in concentrations were ubiquitous. Overall, 60% of medications had more than one continuous dosage unit (range, 1-9). Variation was also noted in bolus dosing; 59 (50%) of 119 drugs had more than one unit (range, 1-4). Dose limits also varied substantially but were difficult to assess since the limits typically varied withthe indication. CONCLUSIONS: Substantial unnecessary variation in i.v. medication practicesis likely associated with increased risk of harm. Standardization has the potential to substantially improve i.v. medication safety.
BACKGROUND: Infusion devices can be programmed with individual hospitals' "best practice" rules for intravenous (i.v.) drug administration, and alerts can be provided if dosages fall outside pre-established limits. High variation levelsare common in medical care but can increase safety risk if the variation is unnecessary. METHODS: The i.v. best practice data sets of drugs from 100 hospitals using one manufacturer's infusion devices were compared to assess the number of drug names used and thevariation in concentrations, dose units, dose limits, and administration practices. RESULTS: The 100 hospitals showed an average of 64 drugs per data set and an average of 113 different drug/concentration 4 combinations. On average, each hospital had designated 6 profiles or unique patient care areas; there were 4 different names per drug across the hospitals (for example, amiodarone had 45 different names). High levels of variation in concentrations were ubiquitous. Overall, 60% of medications had more than one continuous dosage unit (range, 1-9). Variation was also noted in bolus dosing; 59 (50%) of 119 drugs had more than one unit (range, 1-4). Dose limits also varied substantially but were difficult to assess since the limits typically varied withthe indication. CONCLUSIONS: Substantial unnecessary variation in i.v. medication practicesis likely associated with increased risk of harm. Standardization has the potential to substantially improve i.v. medication safety.
Authors: Sonia Pinkney; Mark Fan; Katherine Chan; Christine Koczmara; Christopher Colvin; Farzan Sasangohar; Caterina Masino; Anthony Easty; Patricia Trbovich Journal: Ont Health Technol Assess Ser Date: 2014-05-01