Skip to main content

Table 1 Study dataset linkage by patient demographic and clinical factors. Data are expressed as number (%) or median [IQR]. Standardised differences of 0.2, 0.5 and 0.8 reflect small, medium and large standardised differences respectively

From: Do routine hospital data accurately record comorbidity in advanced kidney disease populations? A record linkage cohort study

Variable

Linked dataset

N = 5506

Non-linked dataset

N = 148

P

Standard diff.

Age (n = 5654)

53 [43–63]

51 [41–61]

0.09

0.15

Sex (n = 5654)

 Male

3422 (62)

84 (57)

0.18

0.11

Ethnicity (n = 5632)

 White

4192 (76)

100 (69)

< 0.001

0.47

 Black

497 (9)

35 (24)

 Asian

750 (14)

10 (7)

 Mixed

48 (1)

0 (0)

Index of Multiple Deprivation (n = 5654)

 1 – Most deprived

1420 (26)

31 (21)

0.51

0.11

 2

1169 (21)

29 (20)

 3

1052 (19)

35 (24)

 4

983 (18)

27 (18)

 5 – Least deprived

882 (16)

26 (17)

ATTOM cohort (n = 5654)

 Dialysis

2150 (39)

49 (33)

0.14

0.16

 Transplant

1780 (32)

59 (40)

 Wait listed

1576 (28)

40 (27)

PRD (n = 5590)

 Polycystic kidney disease

676 (13)

22 (15)

0.005

0.38

 Diabetes

1026 (19)

14 (10)

 Glomerulonephritis

1057 (19)

36 (24)

 Pyelonephritis

460 (8)

15 (10)

 Hypertension

340 (6)

9 (6)

 Renovascular disease

97 (2)

7 (5)

 Other

1090 (20)

33 (22)

 Uncertain

697 (13)

11 (8)

Charlson comorbidity index (n = 5571)

 0

3031 (56)

100 (68)

0.007

0.33

 1–2

1518 (28)

37 (25)

 3–4

583 (11)

7 (5)

 5+

292 (5)

3 (2)

  1. Abbreviations: PRD Primary renal diagnosis