Table 3. Analysis via simple logistic regression and multiple logistic regression to determine the predictive factors of the yield of computed tomographic pulmonary angiogram.
Nil: not available, OR: odd ratio, b: regression coefficient, aForward LR multiple logistic regression was applied.
Multicollinearity and interaction were checked and not found.
Hosmer-Lemeshow test, (p-value = 0.486), Pearson chi-square test, (p-value =0.325), classification tale (overall correctly classified percentage = 75.8%), and area under the ROC curve (70.9%) were applied to check the model fit.
| Variables | Simple logistic regression | Multiple logistic regression | ||||
| b | Crude OR (95% CI) | p-value | b | Adjusted OR (95% CI) | p-value | |
| Age | −0.003 | 0.997 (0.982, 1.012) | 0.680 | |||
| Gender | ||||||
| Female (ref) | 0 | 1.00 | 0 | 1.00 | ||
| Male | 0.723 | 2.060 (1.259, 3.371) | 0.004 | 0.804 | 2.235 (1.328, 3.760) | 0.002 |
| Ethnicity | ||||||
| Malay (ref) | 0 | 1.00 | ||||
| Chinese | 0.535 | 1.707 (0.683, 4.263) | 0.252 | |||
| Indian | −20.183 | Nil (Nil, Nil) | >0.950 | |||
| Others | −0.926 | 0.396 (0.048, 3.269) | 0.390 | |||
| Discipline | ||||||
| Medical (ref) | 0 | 1.00 | ||||
| Surgical | 0.896 | 2.449 (1.185, 5.060) | 0.016 | |||
| O&G | 0.608 | 1.837 (0.775, 4.355) | 0.167 | |||
| Ortho | 0.490 | 1.633 (0.538, 4.952) | 0.386 | |||
| Others | −20.019 | 0 (0, Nil) | >0.950 | |||
| Pregnancy | ||||||
| No (ref) | 0 | 1.00 | ||||
| Yes | 0.082 | 1.085 (0.438, 2.688) | 0.860 | |||
| Trauma | ||||||
| No (ref) | 0 | 1.00 | ||||
| Yes | 1.087 | 2.964 (1.139, 7.714) | 0.026 | |||
| Malignancy | ||||||
| No (ref) | 0 | 1.00 | ||||
| Yes | 0.661 | 1.937 (0.983, 3.816) | 0.056 | |||
| Covid | ||||||
| No (ref) | 0 | 1.00 | 0 | 1.00 | ||
| Yes | −0.743 | 0.475 (0.281, 0.805) | 0.006 | −1.260 | 0.284 (0.160, 0.503) | <0.001 |
| Non-covid pneumonia | ||||||
| No (ref) | 0 | 1.00 | 0 | 1.00 | ||
| Yes | −0.999 | 0.368 (0.194, 0.700) | 0.002 | −1.652 | 0.192 (0.096, 0.384) | <0.001 |