Table 3.

Variable importance for random forest model for prediction of biologic therapy switch at follow-up (final reduced model)

Variable	Mean decrease accuracy	Mean decrease Ginia
Change in BSA involvement from baseline to follow-up	38.1	55.9
BSA involvement at baseline	28.1	40.9
Fatigue at baseline	16.8	133.3
NBSM added from baseline to follow-up	16.2	16.1
DLQI at baseline	13.7	115.3
Patient global assessment at baseline	12.1	139.0
Age at baseline	12.1	181.6
Employment at baseline	11.4	29.7
EQ-5D at baseline	11.1	134.1
Year of baseline visit	3.6	78.8
Ethnicity (Hispanic) at baseline	2.7	18.2
PsA at baseline	2.6	29.4
Number of prior topical therapies as assessed at baseline	2.4	65.7
Duration of biologic therapy at baseline	1.3	50.7
Gender at baseline	− 1.4	30.3

BSA body surface area, DLQI Dermatology Life Quality Index, EQ-5D EuroQOL Five Dimensions Questionnaire, NBSM non-biologic systemic medication, PsA psoriatic arthritis

^aMean decrease in Gini impurity, a method of evaluating prediction fit, measures how much each variable contributes to the homogeneity of the nodes and leaves of a random forest. The smaller the Gini, the more “pure” the resulting forest