Table 3.
Logistic regression results of potential risk factors for severity of coronavirus disease 2019
|
|
Wald χ2
|
P
value
|
OR
|
95%CI
|
|
|
Upper bound
|
Lower bound
|
||||
| Intercept (Moderate) | 1.800 | 0.180 | 9.816 | - | - |
| Intercept (Severe) | 1.122 | 0.289 | 0.179 | - | - |
| Age | 5.477 | 0.019 | 1.055 | 1.009 | 1.104 |
| High-sensitivity cardiac troponin I | |||||
| Normal (≤ 15.61, ≤ 34.22) | |||||
| Higher (> 15.61, > 34.22) | 4.094 | 0.043 | 4.019 | 1.045 | 15.467 |
| Neutrophil to lymphocyte ratio | |||||
| Higher (> 3.13) | 5.477 | 0.042 | 10.126 | 1.088 | 94.247 |
| Normal (≤ 3.13) | |||||
The normal and abnormal range applied to female patients.
The normal and abnormal range applied to male patients.
Ordinal logistic regression model was adopted in JMP15.0 software to explore the risk factors associated with severity of coronavirus disease 2019 patients. Step-by-step regression method (P for inclusion = 0.05, P for exclusion = 0.05) was applied and 3 variables were included in the final model. Results showed that total model test χ2 = 17.380, P < 0.001, which means at least one β in the equation did not equal to 0 and this model was preceded to constant 1. Good-to-fitness test showed χ2 = 59.137, P = 0.968 > 0.05, which means this model fit well. OR: Odds ratio.