Skip to main content

Table 2 The composition of COPD stages and exacerbation risk for each cluster

From: Quantitative CT and COPD: cluster analysis reveals five distinct subtypes with varying exacerbation risks

 

Training

Validation

Cluster 1

Cluster 2

Cluster 3

Cluster 4

Cluster 5

Cluster 1

Cluster 2

Cluster 3

Cluster 4

Cluster 5

GOLD_STAGE

          

 0

131 (54%)

83 (64%)

83 (24%)

6 (5.3%)

1 (0.9%)

123 (50%)

82 (66%)

84 (23%)

4 (3.8%)

0 (0%)

 1

63 (26%)

2 (1.5%)

57 (17%)

11 (9.7%)

1 (0.9%)

59 (24%)

5 (4.0%)

60 (17%)

13 (12%)

2 (1.9%)

 2

18 (7.4%)

20 (15%)

142 (41%)

47 (42%)

29 (26%)

35 (14%)

18 (14%)

158 (44%)

37 (35%)

22 (21%)

 3

0 (0%)

6 (4.6%)

46 (13%)

41 (36%)

60 (54%)

0 (0%)

0 (0%)

33 (9.2%)

35 (33%)

50 (49%)

 4

0 (0%)

0 (0%)

4 (1.2%)

8 (7.1%)

20 (18%)

0 (0%)

0 (0%)

5 (1.4%)

17 (16%)

29 (28%)

Unknown†

30 (12%)

19 (15%)

11 (3.2%)

0 (0%)

0 (0%)

31 (13%)

20 (16%)

18 (5.0%)

0 (0%)

0 (0%)

Exacerbation (%)

52 (21%)

47 (36%)

155 (45%)

76 (67%)

88 (79%)

56 (23%)

43 (34%)

182 (51%)

72 (68%)

79 (77%)

Exacerbation (OR)

1

2.069**

3.012***

7.505***

13.980***

1

1.798*

3.545***

7.261***

11.286***

Exacerbation (Adjusted OR)

1

2.140**

1.951**

3.353***

4.621***

1

2.016**

2.252***

2.711**

3.091**

Hazards ratio

1

1.942**

2.493***

4.906***

6.936***

1

1.620*

2.662***

4.375***

5.680***

  1. Notes: GOLD unknown†: control group with less than 1 pack-year
  2. ***: p < 0.001, **: p < 0.01, *: p < 0.05
  3. OR (odds ratio): calculated from logistic regression
  4. Adjusted OR (adjusted odds ratio): calculated from logistic regression, adjusting for GOLD stage
  5. hazards ratio: calculated from cox proportional hazards model, cox regression analysis was adjusted for race