Development and evaluation of regression tree models for predicting in-hospital mortality of a national registry of COVID-19 patients over six pandemic surges

Table 3 Selected variables in number of bootstraps. Analysed model was based on admission data

Variable	Frequency
Age	200
Number of comorbidities	198
Mechanical ventilated at admission	78
Surge	18
Gender	17
Admission source	10
Immunological Insufficiency	7
BMI	6
Hospital length of stay before ICU admission	5
Referring specialty	3

ISSN: 1472-6947