Skip to main content

Table 5 Univariate analysis on endocrine and metabolic diagnosis between CC and non-CC clusters

From: Application of unsupervised deep learning algorithms for identification of specific clusters of chronic cough patients from EMR data

  Non-CC (N = 8588) CC-1 (N = 4658) CC-2 (N = 1737) CC-3 (N = 154) p value
Obesity 99 (1.15%) 491 (10.54%) 103 (5.93%) 7 (4.55%) < .0001
Type 2 diabetes 352 (4.1%) 1272 (27.31%) 294 (16.93%) 21 (13.64%) < .0001
Hyperlipidemia 196 (2.28%) 1650 (35.42%) 423 (24.35%) 23 (14.94%) < .0001
Hypothyroidism 153 (1.78%) 713 (15.31%) 184 (10.59%) 12 (7.79%) < .0001
Hypovolemia 51 (0.59%) 569 (12.22%) 78 (4.49%) 1 (0.65%) (< .0001, < .0001, 0.9293)
  1. The p values are shown in the last column. < .0001 was listed in the last column if all of three p values were < .0001