Table 2.

 Summary of 2×2 tables indicating correct and incorrect diagnoses of endobronchial intubation and correct and incorrect diagnoses of excluding endobronchial intubation by different methods for assessment of position of endotracheal tube.* Each of 20 patients in each group assessed independently by experienced and inexperienced anaesthetists resulting in 40 independent observations

Tube position and diagnosis	Auscultation	Observation	Depth	All three
Endobronchial position:
Correct diagnosis	26	17	35	40
Incorrect diagnosis	14	23	5	0
Tracheal position:
Correct diagnosis	37	36	39	38
Incorrect diagnosis	3	4	1	2
Odds ratio (95% CI)†	10.5 (2.3 to 47.5), P=0.002	19.9 (4.5 to 88.5), P<0.001	3.2 (0.6 to 17.0), P=0.18	1

*Bilateral auscultation of chest; observation of symmetrical chest movements; checking cm scale (depth); or combination of all three.

†Odds ratio to predict incorrect tube position according to bedside test with “all three” as baseline category from logistic regression model with 95% confidence intervals calculated from robust standard errors to allow for correlation within patients.