Skip to main content

Table 2 Background characteristics of patients identified by Algorithms 1 and 2

From: Validation of algorithms to identify colorectal cancer patients from administrative claims data of a Japanese hospital

Characteristics

Algorithm 1: Colorectal cancer codes

 

Algorithm 2: Site-specific colorectal cancer codes

Total

(n = 1,406)

Samples

(n = 249)

 

Totalb

(n = 524)

Samples

(n = 100)

Age

         

 Mean ± SD

66.60 ± 13.07

66.69 ± 13.82

 

67.63 ± 12.57

67.83 ± 13.32

 Median [min–max]

69.0 [21–96]

69.0 [21–95]

 

69.0 [21–96]

69.5 [21–90]

Men, n (%)

815

(57.97)

135

(54.22)

 

285

(54.39)

50

(50.00)

Diagnosis, n (%)a

         

 Colorectal cancer

512

(36.42)

84

(33.73)

 

15

(2.86)

2

(2.00)

 Rectal cancer

348

(24.75)

55

(22.09)

 

6

(1.15)

–

–

 Sigmoid colon cancer

226

(16.07)

51

(20.48)

 

226

(43.13)

51

(51.00)

 Ascending colon cancer

141

(10.03)

21

(8.43)

 

141

(26.91)

21

(21.00)

 Transverse colon cancer

76

(5.41)

16

(6.43)

 

76

(14.5)

14

(14.00)

 Cecal cancer

56

(3.98)

10

(4.02)

 

56

(10.69)

10

(10.00)

 Descending colon cancer

33

(2.35)

7

(2.81)

 

33

(6.3)

7

(7.00)

 Colon cancer

18

(1.28)

6

(2.41)

 

–

–

–

–

 Postoperative recurrence of rectal cancer

16

(1.14)

4

(1.61)

 

–

–

–

–

 Appendix cancer

15

(1.07)

–

–

 

–

–

–

–

 Ileocecal cancer

–

–

–

–

 

6

(1.15)

2

(2.00)

  1. aThe earliest diagnostic records for colorectal cancer are tabulated. Some patients had multiple diagnoses on the same day. Only diagnoses with a proportion of > 1.0% were listed
  2. bData of 524 patients who met Algorithm 2 within all records obtained are shown instead of 102 patients who met Algorithm 2 within the sampled patients for Algorithm 1
  3. Note: SD, standard deviation