Fig. 2From: Utilization of anonymization techniques to create an external control arm for clinical trial dataCox model results for the association between overall survival and the confounders in fully anonymized and pseudonymized real world data sets. BMI, body-mass index (kg/m2); COPD, chronic obstructive pulmonary disease; NOAC, novel oral anticoagulant; SSRI, selective serotonin reuptake inhibitor; TIA, transient ischemic attackBack to article page