Comparison of automated and retrospectively calculated estimated glomerular filtration rate in electronic health record data

Table 1 Characteristics of the study cohort

	Overall N = 266,084		Discordant eGFR values N = 86,747		Concordant eGFR values N = 179,337
Characteristic	Mean	SD	Mean	SD	Mean	SD	p-value
Age	54.87	27.00	51.83	18.36	56.33	17.90	< 0.05
Scr	1.04	0.51	1.02	0.39	1.05	0.55	0.74
Retroactively calculated eGFR	83.16	28	84.88	23.04	82.33	22.9	< 0.05
Automated eGFR	84.00	27	87.25	23.48	82.41	22.94	< 0.05
Gender	N	%	N	%	N	%	< 0.05
Female	25,725	9.67	9933	11.45	15,792	8.81
Male	240,359	90.33	76,814	88.55	163,545	91.19
Race binary							< 0.05
Black	46,873	17.62	19,380	22.34	27,493	15.33
Other	219,211	82.38	67,367	77.66	151,844	84.67
Race							< 0.01
Am. Indian	3068	1.15	1034	1.19	2034	1.13
Asian	4058	1.53	1310	1.51	2748	1.53
Black	46,873	17.62	19,380	22.34	27,493	15.33
White	212,085	79.71	65,023	74.96	147,062	82.00

SD standard deviation, Scr: serum creatinine, p-values generated from independent t-tests and chi square tests

ISSN: 1471-2369