Machine Learning Readmission Risk Modeling: A Pediatric Case Study

Table 1

Descriptive statistics of the dataset.

Dataset characteristic
Total number of admissions	56,558
Number of unique individuals	35,064
Percent readmission within 30 days	3.72%
Number of unique procedures (ICD-10 AM)	1,124
Number of unique diagnoses (ICD-10 AM)	4,370
Variables used in prediction
Age (years), mean (SD)	5.78 (5.04)
Male (%)	59.2
Public facilities	1
Number of Transfers (SD)	0.61 (0.8)
Length of Stay (days), mean (SD)	3.77 (10.03)