Skip to main content
Fig. 2 | Health Economics Review

Fig. 2

From: Predicting high health-cost users among people with cardiovascular disease using machine learning and nationwide linked social administrative datasets

Fig. 2

Important variables extracted from the random forest model. Notes: PRM = Pharmaceuticals; NAP = Non-admitted patients (i.e., outpatients and emergency department [ED] visits); Lab = Laboratory tests; NMD = Public hospitalisations (National minimum dataset); PHO = Primary health organisation (i.e., GP) enrolments. Gini index provides a relative ranking of variable importance [46]. Chronic conditions other than CVD: diabetes, cancers and traumatic brain injuries. Deprivation quintiles on a one to five scale with five being the most deprived

Back to article page

Health Economics Review

ISSN: 2191-1991

Contact us