Measuring Errors in Pediatric Liquid Medication among Caregivers: Magnitudes and Related Factors

Abstract

Objectives: To determine dosing error of pediatric liquid medications among caregivers, its related factors and to identify the measures for the error from the perspective of pharmacists. Method: The research consisted of three studies. The first study collected 17,746 physician orders of pediatric liquid medications over 1 year period within a community hospital. The study regarded orders with the same level of viscosity of medication, unit of measurement, volume and measuring device as the same patterns of prescribing. The researchers selected 34 out of 59 patterns (covering 97.36% of physician orders) for determination of dosing error among caregivers. In the second study, subjects were 400 conveniently chosen parents who visited the hospital. Each was tested with 5 randomly selected patterns of prescribing. The study defined dosing error as measured volume deviating from the prescribed amount more than 20% (error 20%) and large dosing error as >40% deviation from prescribed amount (error 40%). The results of testing were presented and considered within a focus group discussion among pharmacists to identify the measures for reducing the error in the third study. Results: Three patterns of prescribing with the highest dosing error were measuring 1 tablespoonful of low viscosity medication by dosing cup (79.37% of parents), followed by measuring 0.3 cc of low viscosity medication with a dropper (62.29% of parents) and measuring about a teaspoon of moderate viscosity medication with a teaspoon (58.50% of parents). By using the dosing error rate of each prescribing pattern to estimate overall magnitude of error, it was found that, within one year, error 20% was committed in 5,590 orders for pediatric liquid medications or 32.35% with 19.74% (3,410 orders) and 12.62% (2,180 orders) being underdose and overdose errors, respectively. Logistic regression analysis revealed that factors significantly related to higher odds of error 20% were higher age (OR=1.02; 95%CI =1.01-1.03), higher viscosity of medication (OR=1.40; 95%CI =1.06-1.84 and OR=1.87; 95%CI =1.45-2.41 for moderate and high viscosity, compared to low viscosity), prescribing in tablespoon unit (OR=12.00; 95%CI =6.10-23.60) and prescribing in decimal or not in an integer volume (OR=1.94; 95%CI =1.53-2.48). Higher health literacy was related to decreased odds of error 20% (OR=0.90; 95%CI =0.85-0.96). Strategies for reducing measuring errors applicable to all prescribing pattern were the provision of relevant education to parents by pharmacists with the use of advanced counseling techniques, changing prescribing order by using milliliter as a unit of measurement instead of tablespoon and those difficult to measure, changing measuring devices to oral syringe and prescribing drugs with higher concentration to reduce errors from multiple measuring. Conclusion: Dosing error of liquid medications was committed by caregivers one-third of physician orders. Ability of caregivers, drug orders, and medications were related to the occurrence of errors, and should be considered in developing measures to address this issue.