Data mining application to healthcare fraud detection: a two-step unsupervised clustering method for outlier detection with administrative databases

Table 4 Variables of interest for cross validation

Variable	Dataset	Notes
Age	Patients	Indicator of patients’ complexity
length of Stay	Patients, HDC	Indicator of patients’ complexity
Comorbidity	Patients, HDC	Indicator of patients’ complexity
Total Costs	Patients, HDC	Patients’ expensiveness
Cost / length of Stay	HDC	Expenses in relation with intensity of care
Cost / Comorbidity	Hospital	Expenses in relation with intensity of care

ISSN: 1472-6947