Background: Numerous equations are used for estimation of renal function, and many electronic medical records report multiple clearance estimates to assist with drug dosing. It is unknown whether the presence of multiple clearance estimates affects clinical decision-making.
Objective: To determine whether the presence of multiple renal clearance estimates affects pharmacist drug dosing decisions.
Methods: A randomized trial in the form of an electronic survey including 4 clinical vignettes was delivered to hospital pharmacists. Vignettes consisted of a patient presenting with an acute pulmonary embolism requiring enoxaparin therapy. Pharmacists were randomized to receive a single estimate of renal function or multiple estimates for all vignettes. The primary outcome was deviation from approved recommendations on at least 1 vignette. The χ2 test was used to detect differences in deviation rates between groups. Logistic regression was performed to adjust for the effects of potentially confounding variables.
Results: A total of 154 studies were completed (73 in the multiple-estimate group and 81 in the single-estimate group). Pharmacists presented with multiple renal estimates were significantly more likely to deviate from recommended dosing regimens than pharmacists presented with a single estimate (54.7% vs 38.2%; P = 0.04). The results were driven primarily by the 2 vignettes that included discordance among Cockcroft-Gault equation creatinine clearance estimates. Logistic regression identified multiple estimates as the only independent predictor of deviation (P = 0.04).
Conclusion and relevance: Pharmacists provided with a single renal clearance estimate were more likely to adhere to approved dosing recommendations than pharmacists provided with multiple estimates.
Keywords: adult medicine; anticoagulation; clinical decision-making; clinical pharmacy; dosing; drug information; electronic information; nephrology.