Skip to main content
. 2021 Dec 9;11(12):e053983. doi: 10.1136/bmjopen-2021-053983

Table 4.

Binary logistic regression for mortality and SARS-CoV-2 postoperative infection.

Binary logistic regression for mortality
Variables OR 95% CI P value
 Age* 1.05* 1.034 to 1.068 <0.001
 Community COVID-19 incidence (cases/105 people/week)* 1.006* 1.002 to 1.009 <0.001
 SARS-CoV-2 perioperative infection 4.7 1.81 to 12.1 <0.001
 Postoperative neurological worsening 5.9 3.27 to 10.66 <0.001
 GCS 3–8 2.82 1.34 to 5.94 0.006
 Postoperative airway support 5.38 2.81 to 10.3 <0.001
 ASA grade ≥3 2.5 1.31 to 4.79 0.005
Nagelkerke R square=0.338
Binary logistic regression for postoperative SARS-CoV-2 infection
Community COVID-19 incidence (cases/105 people/week)* 1.013* 1.008 to 1.018 <0.001
Screening swab test <72 hours preoperatively 0.098 0.012 to 0.778 0.028
Preoperative cognitive impairment 2.784 1.037 to 7.471 0.042
Postoperative sepsis 3.807 0.968 to 14.976 0.056
No postoperative complications 0.188 0.068 to 0.521 0.001
Nagelkerke R square=0.351

The OR, 95% CI and level of significance (P) are provided for each variable, as well as the constant of the model and the Nagelkerke R square.

*OR provided per unit increase.

ASA, American Society of Anesthesiologists; GCS, Glasgow Coma Scale.