Table 3.

Can use of height and weight eliminate the need for race in GFR estimation in the CKD-EPI creatinine equation?

Variables Used in Equations for eGFR	All, n=8254		Black, n=2601, Root Mean Square Errorb	Nonblack, n=5653, Root Mean Square Errorb
	Black Race Coefficient (95% CI)a	Root Mean Square Errorb
Serum creatinine, age, sex, race	1.16 (1.14 to 1.17)	0.236 (0.229 to 0.242)	0.243 (0.232 to 0.254)	0.232 (0.225 to 0.241)
Serum creatinine, age, sex	NA	0.244 (0.238 to 0.251)	0.258 (0.248 to 0.268)	0.238 (0.230 to 0.247)
Serum creatinine, age, sex, race, height, weight	1.15 (1.14 to 1.17)	0.235 (0.229 to 0.242)	0.242 (0.232 to 0.253)	0.232 (0.225 to 0.241)
Serum creatinine, age, sex, height, weight	NA	0.243 (0.237 to 0.250)	0.255 (0.245 to 0.265)	0.238 (0.230 to 0.246)

Data are from the pooled CKD-EPI development and internal validation datasets. Mean (SD) measured GFR =68 (40) ml/min per 1.73 m² (28). Other GFR estimating equations using serum creatinine were not considered because they are not more accurate than the CKD-EPI creatinine equations in external validation datasets. 95% CI, 95% confidence interval; NA, not applicable. Modified from ref. 72, with permission.

^aCoefficient for equation expressed on the multiplicative scale of GFR.

^bRoot mean square error for the regression of measured GFR on eGFR computed on the logarithmic scale. A lower root mean square error indicates higher accuracy of the eGFR.