Skip to main content

Table 1 Baseline characteristics before and after propensity score matching

From: Deep learning as a novel method for endoscopic diagnosis of chronic atrophic gastritis: a prospective nested case–control study

Characteristic

Before matching

After matching

CAG (n = 338) (%)

CNAG (n = 793) (%)

Standardized difference

CAG (n = 338) (%)

CNAG (n = 338) (%)

Standardized difference

Sex (%)

  

− 0.0875

  

− 0.0266

 Male

70.4

61.7

 

70.4

67.8

 

 Female

29.6

38.3

 

29.6

32.2

 

Age

  

0.0158

  

0.0030

 Distribution (%)

      

 < 40 yrs

8.9

10.6

 

8.9

9.8

 

 40–59 yrs

48.8

46.4

 

48.8

45.3

 

 60–75 yrs

33.4

34.7

 

33.4

38.2

 

 > 75 yrs

8.9

8.3

 

8.9

6.8

 

Indication (%)

  

− 0.0370

  

0.0030

 Screening

37.9

34.2

 

37.9

38.2

 

 Diagnosis

62.1

65.8

 

62.1

61.8

 

HP (%)

  

− 0.0149

  

0.0059

 Yes

26.6

28.1

 

26.6

26.0

 

 No

73.4

71.9

 

73.4

74.0

 

Smoking (%)

  

0.0294

  

− 0.0207

 Yes

31.1

28.1

 

31.1

33.1

 

 No

68.9

71.9

 

68.9

66.9

 

Drinking (%)

  

− 0.0480

  

− 0.0030

 Yes

21.3

26.1

 

21.3

21.6

 

 No

78.7

73.9

 

78.7

78.4

 

HT (%)

  

− 0.0150

  

0.0059

 Yes

32.5

34.0

 

32.5

32.0

 

 No

67.5

66.0

 

67.5

68.0

 

CHD (%)

  

− 0.0326

  

0.0059

 Yes

25.7

29.0

 

25.7

25.1

 

 No

74.3

71.0

 

74.3

74.9

 

Diabetes (%)

  

0.0026

  

0.0030

 Yes

24.9

24.6

 

24.9

24.5

 

 No

75.1

75.4

 

75.1

75.5

 
  1. CAG chronic atrophic gastritis, CNAG chronic nonatrophic gastritis, HP helicobacter pylori, HT hypertension, CHD coronary heart disease