Impact of Removing the Race Coefficient in Renal Function Estimate Equations on Drug Dosage Recommendations

Abstract

Background The appropriateness of including the race coefficient in glomerular filtration rate (GFR) equations in Black patients is debated, and the impact on drug dosing is unknown. Objective This study explored the impact of removing the race coefficient on drug dosing in Black patients in comparison to conventional methods. Methods This was a retrospective study of hospitalized patients who self-identified as Black/African American and were prescribed an antimicrobial that includes renal dosage recommendations in the product labeling. The primary end point was the discordance between drug dosing recommendations derived by body surface area deindexed GFR estimated by the CKD-EPI equation (Chronic Kidney Disease Epidemiology study) with and without race versus recommendations derived from Cockcroft-Gault (CG). Results A total of 210 Black patients were included. There was an 18% rate of discordance when GFR was estimated with the race coefficient (GFR w/Race) versus without the race coefficient (GFR w/out Race). GFR w/out Race had a higher level of agreement with dosing by creatinine clearance (CrCl; κ = 0.779) than GFR w/Race versus CrCl (κ = 0.651). GFR w/out Race had less within-patient difference than GFR w/Race in comparison to CrCl (mean difference: −6.3 vs −18.0 mL/min). Conclusions and Relevance This represents the first report to examine the removal of the race coefficient and its implication on drug dose discordance. GFR w/out Race had a higher level of agreement and less drug dose discordance than GFR w/Race, in comparison to CrCl estimates. If GFR equations are considered comparable to CrCl for the purposes of guiding drug dosing, GFR w/out Race should be considered.