The COVID-19 pandemic has prompted aninternational effort to develop and repurpose medica-tions and procedures to effectively combat the disease.Several groups have focused on the potential treat-ment utility of angiotensin-converting–enzyme inhibitors(ACEIs) and angiotensin-receptor blockers (ARBs) forhypertensive COVID-19 patients, with inconclusive ev-idence thus far. We couple electronic medical record(EMR) and registry data of 3,643 patients from Spain,Italy, Germany, Ecuador, and the US with a machinelearning framework to personalize the prescription ofACEIs and ARBs to hypertensive COVID-19 patients.Our approach leverages clinical and demographic in-formation to identify hospitalized individuals whoseprobability of mortality or morbidity can decrease byprescribing this class of drugs. In particular, the algo-rithm proposes increasing ACEI/ARBs prescriptionsfor patients with cardiovascular disease and decreasingprescriptions for those with low oxygen saturation atadmission. We show that personalized recommendationscan improve patient outcomes by 1.0% compared to thestandard of care when applied to external populations.We develop an interactive interface for our algorithm,providing physicians with an actionable tool to easily as-sess treatment alternatives and inform clinical decisions.This work offers the first personalized recommendationsystem to accurately evaluate the efficacy and risks ofprescribing ACEIs and ARBs to hypertensive COVID-19patients.
The main model fitting script is treatment_cluster_control.py and results are evaluated in treatment_evaluation_control.py