Skip to main content

Table 3 Logistic regression models for the occurrence of any complication, and of a cardiopulmonary complication

From: Comparison of postoperative complications between segmentectomy and lobectomy by video-assisted thoracic surgery: a multicenter study

ModelsAny complicationCardiopulmonary complication
OR (95% CI)POR (95% CI)P
1VATS segmentectomya0.91 (0.61–1.30)0.580.93 (0.65–1.34)0.71
Propensity scoreb0.42 (0.16–1.17)0.100.46 (0.19–1.13)0.049
2VATS segmentectomya0.93 (0.70–1.24)0.620.86 (0.60–1.24)0.41
  1. Model 1: standard logistic regression, where the propensity score was left as a continuous probability between 0 and 1. Model 2: Conditional logistic regression, with 15 sets matched for propensity (0.100–0.149, 0.150–0.199, 0.200–0.249, etc). Fifty-five patients with a propensity for segmentectomy < 0.10 and > 0.85 were excluded, as all were treated with the same intervention. a vs lobectomy
  2. b probability of 1 vs 0
  3. VATS Video-assisted thoracic surgery, OR Odds ratio, CI Confidence interval