JMIR Publications

ABSTRACT

Machine learning clustering offers an unbiased approach to better understand the interactions of complex social and clinical variables via integrative subphenotypes, an approach not studied in out-of-hospital cardiac arrest (OHCA). We conducted a cluster analysis for a cohort of OHCA survivors to examine the association of clinical and social factors for mortality at one year.We utilized a retrospective observational OHCA cohort identified from Medicare claims data, including area level SDOH features and hospital level datasets. We applied k-means clustering algorithms to identify subphenotypes of beneficiaries who had survived an OHCA and examined associations of outcomes by subphenotype.27,028 unique beneficiaries survived to discharge after OHCA. We derived 4 distinct subphenotypes, finding subphenotype 1 with the highest unadjusted mortality (53.8%) and subphenotype 4 with low mortality (31.7%). Jointly modeling of these features demonstrated an increased hazard of death for subphenotypes 1-3 but not for subphenotype 4 when compared to reference.We identified four distinct subphenotypes with differences in outcomes by clinical and area level SDOH features for OHCA. Further work is needed to determine if individual or other SDOH domains are specifically tied to long-term survival after OHCA.