Exploring Sex Differences in Cardiac ICU Admission and Length of Stay Using Multiple Correspondence Analysis

Sex differences in intensive care unit (ICU) admission and treatment are reflected in the underrepresentation of women, poorer outcomes, and higher mortality risk. Historical exclusion from clinical trials has limited evidence for providing sex-specific care. This study aims to examine demographic and admission data from MIMIC-IV electronic health records to understand sex differences in ICU admissions, treatments, and outcomes, addressing critical gaps in care. Using the MIMIC-IV database of de-identified electronic health records, multiple correspondence analysis (MCA) was applied as an exploratory method to identify patterns and associations across categorical variables, including ICU type and demographic factors, for both heart failure (HF) and myocardial infarction (MI) patients. The study aimed to determine whether demographic or socioeconomic factors influenced assignment to specialty cardiac ICUs, and to evaluate differences in critical care delivery among patients with HF and MI. We observed that women were associated with admission to non-specialty ICUs, whereas men were admitted to specialty cardiac ICUs. MCA uncovered subgroup patterns showing that admission to cardiac ICUs is not uniform across sex, race, insurance, age, or marital status. There were also racial discrepancies noted in assignments to specialty ICUs, with minority groups being underrepresented in specialty ICUs. Marital status was a factor in admissions: widowed and single people were more frequently admitted to non-specialty ICUs. Length of stay (LOS) was comparable between HF and MI patients; however, chi-square analysis showed LOS being significantly associated with sex, age and marital status in HF, and with insurance status in the MI cohort. This data exploration illustrates systemic inequities in ICU allocation based on sex, race, and socioeconomic factors. Evidence of sex-based bias in the medical community highlights the need to address these inequalities in treatments and critical care to achieve better outcomes in the future.