Summary of baseline patient characteristics and final diagnoses
| All patients | Development set | Validation set | p-value# | |
| Subjects n | 244 | 114 | 130 | |
| Age years | 56.15±14.76 58 (16–94) | 55.93±15.18 57.5 (16–87) | 56.34±14.44 58 (21–94) | 0.830 |
| Male/female | 150/94 | 68/46 | 82/48 | 0.583 |
| Diagnosis | 0.512 | |||
| Malignant | 108 (44.26) | 53 (46.49) | 55 (42.31) | |
| Benign | 136 (55.74) | 61 (53.51) | 75 (57.69) | |
| Final diagnosis | ||||
| Malignant¶ | 108 (44.26) | 53 (46.49) | 55 (42.31) | |
| Definitive diagnosis | 52 (48.15) | 33 (62.26) | 19 (34.55) | |
| Probable diagnosis | 56 (51.85) | 20 (37.74) | 36 (65.45) | |
| Benign | 136 (55.74) | 61 (53.51) | 75 (57.69) | |
| Pneumonia | 44 (18.03) | 22 (19.30) | 22 (16.92) | |
| Tuberculous pleurisy | 45 (18.44) | 33 (28.95) | 12 (9.23)§ | |
| Congestive heart failure | 11 (4.51) | 2 (1.75) | 9 (6.92) | |
| Hepatic disease | 22 (9.02) | 3 (2.63) | 19 (14.62) | |
| Renal failure | 8 (3.28) | 0 (0) | 8 (6.15) | |
| Other benign disease+ | 6 (2.46) | 1 (0.88) | 5 (3.85) |
Data are presented as n, mean±sd, median (range) or n (%), unless otherwise stated. #: comparison of the development set with the validation set; ¶: refers to lung cancer (n=40), mesothelioma (n=5), thymic tumour (n=1), breast cancer (n=1), gastrointestinal cancer (n=1) and unclassified malignancy (n=5) in the development set, and refers to lung cancer (n=34), mesothelioma (n=4), thymic tumour (n=3), breast cancer (n=6), gastrointestinal cancer (n=4) and unclassified malignancy (n=4) in the validation set; +: refers to Sjogren syndrome (n=1) in the development set and systemic sclerosis (n=1), systemic lupus erythematosus (n=3) and Sjogren syndrome (n=1) in the validation set; §: one patient suffered from tuberculous pleurisy and empyema and was listed only once, under tuberculous pleurisy.