Using machine learning to develop a clinical prediction model for SSRI-associated bleeding: a feasibility study

Table 4 Clinically significant feature clusters for each drug cohort

Cohort	Clinically significant feature clusters^a
Clopidogrel	Bleeding history
Warfarin	Bleeding history, antithrombotics
Escitalopram	Health literacy
Citalopram	Bleeding history
Fluoxetine	Bleeding history, socioeconomic
Sertraline	Bleeding history
Paroxetine	Bleeding history
Combined SSRIs	Bleeding history, socioeconomic

^a Clinically significant feature clusters resulted in a > 0.01 decline in AUC score among 3 out of 4 machine learning models (frequency ≥ 0.75)

ISSN: 1472-6947