Racial and ethnicity differences in health outcomes exist across many areas in medicine. Few studies have investigated race/ethnicity differences among critically ill patients and have not fully accounted for severity of illness, socioeconomic status (SES), or end‐of‐life preferences.
Thirty‐five California hospitals retrospectively collected data from 11,300 adult ICU admissions with an ICU stay ≥ 4 hours, We excluded: (1) bum, trauma, CABG patients; (2) patients without linkage to the statewide discharge database; and (3) Native American patients because of the small sample. To estimate the independent association of race/ethnicity with hospital mortality and ICU length of stay (LOS), we fit staged hierarchical logistic and linear regression models, respectively. The initial model included race/ethnicity (white, black, Hispanic, and Asian), age, and sex, We then examined the effect of severity of illness by adding the APACHE IV predicted probability of death that accounts for acute physiology and comorbidities. Next, we added admission do‐not‐resuscilale (DNR) order status and finally zip code‐based SES index and insurance type.
Nine thousand five hundred and eighteen patients met inclusion and exclusion criteria (6334 white, 655 black, 1917 Hispanic, and 612 Asian). White patients were the most likely to have insurance (93.6%) and a DNR order (5.7%), Hispanic patients had the lowest SES index. Black patients had the lowest percentage of DNR orders (2.9%). Hospital mortality was 15.9% among the entire cohort. Asian and Hispanic patients had the highest crude mortality at 18.6% and 17.6%, respectively. After initial adjustment for severity of illness, the increased odds of death for Asian and Hispanic patients were no longer statistically significant. Further adjustment for ESS, DNR status, and insurance indicated no differences between groups. The median ICU LOS was 2.0 days (IQR 1.0–4.1 days). Black patients' ICU LOS was 0.64 days (95% Cl 0.2–1.1 days), longer than whites in unadjusted analysis. The adjusted model, for age, sex, APACHE predicted mortality, DNR status, SES, and insurance status, continued to see increases in ICU LOS by 0.41 days (95% Cl 0.01–0.82). When restricted only to those who died, the adjusted length of stay was considerably longer for black patients than for whites, averaging an additional 1.2 days in the ICU (95% Cl, ‐0.24 to 2.6).
Our study did not find any significant differences in mortality between ICU patients of varying race/ethnicities after adjustment for severity of illness, SES, and DNR status. Black patients were found to have low rates of DNR orders and prolonged length of stay, especially among those who died. Although we did not examine factors underlying the differences in DNR rates and ICU LOS, variations may reflect patient care preferences, cultural traditions valuing life‐prolonging measures, or measures of trust in the health care system.
E. Vasilevskis, none; S. Erickson, none; M. W. Kuzniewicz, none; M. L. Dean, none; T. Clay, none; D. Rennie, none; R. A. Dudley, none.
To cite this abstract:Vasilevskis E, Erickson S, Kuiniewicz M, Cason B, Lane R, Dean M, Clay T, Rennie D, Dudley R. Racial/Ethnicity Disparities in ICU Outcomes. Abstract published at Hospital Medicine 2010, April 8-11, Washington, D.C. Abstract 148. Journal of Hospital Medicine. 2010; 5 (suppl 1). https://www.shmabstracts.com/abstract/racialethnicity-disparities-in-icu-outcomes/. Accessed January 26, 2020.